CN113926571B - Tailing activation process system based on drying, grinding and selecting integrated intelligent production equipment - Google Patents
Tailing activation process system based on drying, grinding and selecting integrated intelligent production equipment Download PDFInfo
- Publication number
- CN113926571B CN113926571B CN202111217407.6A CN202111217407A CN113926571B CN 113926571 B CN113926571 B CN 113926571B CN 202111217407 A CN202111217407 A CN 202111217407A CN 113926571 B CN113926571 B CN 113926571B
- Authority
- CN
- China
- Prior art keywords
- grinding
- sensor
- selecting
- intelligent
- box body
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000227 grinding Methods 0.000 title claims abstract description 128
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 72
- 238000001035 drying Methods 0.000 title claims abstract description 64
- 238000001994 activation Methods 0.000 title claims abstract description 52
- 239000000843 powder Substances 0.000 claims abstract description 46
- 238000005265 energy consumption Methods 0.000 claims abstract description 43
- 239000000463 material Substances 0.000 claims abstract description 39
- 230000004913 activation Effects 0.000 claims abstract description 29
- 238000000034 method Methods 0.000 claims abstract description 24
- 230000008569 process Effects 0.000 claims abstract description 22
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 17
- 238000012216 screening Methods 0.000 claims abstract description 12
- 238000007885 magnetic separation Methods 0.000 claims abstract description 10
- 230000003213 activating effect Effects 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims abstract description 9
- 239000010410 layer Substances 0.000 claims description 65
- 238000007726 management method Methods 0.000 claims description 34
- 230000002159 abnormal effect Effects 0.000 claims description 29
- 238000012544 monitoring process Methods 0.000 claims description 28
- 238000012549 training Methods 0.000 claims description 27
- 238000006073 displacement reaction Methods 0.000 claims description 23
- 239000013598 vector Substances 0.000 claims description 21
- 238000010606 normalization Methods 0.000 claims description 18
- 238000012795 verification Methods 0.000 claims description 18
- 230000000694 effects Effects 0.000 claims description 15
- 230000007246 mechanism Effects 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 10
- 230000000149 penetrating effect Effects 0.000 claims description 10
- 238000011282 treatment Methods 0.000 claims description 10
- 238000013507 mapping Methods 0.000 claims description 9
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 claims description 8
- 239000012190 activator Substances 0.000 claims description 8
- 238000002347 injection Methods 0.000 claims description 8
- 239000007924 injection Substances 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 7
- 238000011478 gradient descent method Methods 0.000 claims description 6
- 238000012886 linear function Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 239000000203 mixture Substances 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 4
- 229910052742 iron Inorganic materials 0.000 claims description 4
- 230000009467 reduction Effects 0.000 claims description 4
- 230000000087 stabilizing effect Effects 0.000 claims description 4
- 239000000725 suspension Substances 0.000 claims description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 4
- 230000009471 action Effects 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000000889 atomisation Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 claims description 3
- 239000000428 dust Substances 0.000 claims description 3
- 239000011229 interlayer Substances 0.000 claims description 3
- 230000000306 recurrent effect Effects 0.000 claims description 2
- 238000005096 rolling process Methods 0.000 claims description 2
- 238000005507 spraying Methods 0.000 claims description 2
- 230000003993 interaction Effects 0.000 claims 1
- 238000004148 unit process Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 5
- 238000000926 separation method Methods 0.000 description 5
- 238000005498 polishing Methods 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 3
- 238000011068 loading method Methods 0.000 description 2
- 239000006148 magnetic separator Substances 0.000 description 2
- 239000002994 raw material Substances 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 239000011120 plywood Substances 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 239000007921 spray Substances 0.000 description 1
- 239000013076 target substance Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
- B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
- B02C21/00—Disintegrating plant with or without drying of the material
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
- B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
- B02C19/00—Other disintegrating devices or methods
- B02C19/16—Mills provided with vibrators
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
- B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
- B02C23/00—Auxiliary methods or auxiliary devices or accessories specially adapted for crushing or disintegrating not provided for in preceding groups or not specially adapted to apparatus covered by a single preceding group
- B02C23/08—Separating or sorting of material, associated with crushing or disintegrating
- B02C23/14—Separating or sorting of material, associated with crushing or disintegrating with more than one separator
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
- B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
- B02C25/00—Control arrangements specially adapted for crushing or disintegrating
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P40/00—Technologies relating to the processing of minerals
- Y02P40/10—Production of cement, e.g. improving or optimising the production methods; Cement grinding
Landscapes
- Engineering & Computer Science (AREA)
- Food Science & Technology (AREA)
- General Factory Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The tailing activation process system based on the drying, grinding and selecting integrated intelligent production equipment is characterized by comprising tailing intelligent production equipment and an intelligent tailing utilization execution system, wherein the tailing intelligent production equipment is used for carrying out an activation process on tailings, namely drying, grinding and selecting, drying, grinding and selecting of materials under selecting and selecting, activating of materials on selecting and comprehensively utilizing the tailings, and the selecting process comprises magnetic separation and screening which are sequentially carried out. The drying, grinding and selecting integrated processing mode is adopted, the magnetic separation and crushing steps are separated and put into the powder selecting step, the stability of the quality of products is guaranteed, the intelligent algorithm is adopted to monitor and adjust the tailing activation process in real time, the equipment faults and the energy consumption are monitored, the intelligent activation production of the tailings is realized, the average field problem processing time is shortened by 25%, and the production plan execution rate is effectively improved by 15%.
Description
Technical Field
The invention relates to a tailing activation process system, in particular to a tailing activation process system based on drying, grinding and selecting integrated intelligent production equipment, and belongs to the field of activation process equipment.
Background
Tailings are not well utilized due to the low content of the target composition. The activation process can improve the activity index of tailings, the existing activation process separately executes drying, iron removal and grinding powder separation in different equipment, so that the cost of grinding powder separation is increased, the process cannot be monitored in real time, and even if the same ore dressing process is used for producing tailings in the same mining area, the stable fluctuation of the activity index of products caused by the fact that the activation process is finished due to different batches cannot be guaranteed.
Disclosure of Invention
In order to solve the problems, the invention adopts a scheme of drying, grinding and selecting into a whole, separates the traditional magnetic separation and breaking steps, finishes the breaking step in the first step, and then enters the drying and grinding process. The magnetic separation step is carried out in the powder selection step after grinding, and the content of the target substance is greatly ensured by adopting a circulating drying grinding selection process. Meanwhile, an artificial neural network ANN algorithm is introduced, so that the relative stability of the activity index of the activated product is realized.
Therefore, the invention provides a tailing activation process system based on drying, grinding and selecting integrated intelligent production equipment, which is characterized by comprising tailing intelligent production equipment and used for carrying out the following activation processes on tailings, namely drying, grinding and selecting treatment, drying, grinding and selecting circulation treatment of materials under selecting and selecting, activation treatment of materials on selecting, and
the tailings are comprehensively utilized by an intelligent production execution system, wherein the powder selection treatment comprises magnetic separation and screening which are sequentially carried out.
Preferably, the screening comprises rolling screening or vibratory screening.
Further, the tailing intelligent production equipment comprises an intelligent 'drying, grinding and selecting' all-in-one machine and an intelligent activating machine which are sequentially connected, wherein,
the intelligent drying, grinding and selecting integrated machine is provided with a first current sensor, a first voltage sensor, a first displacement sensor for monitoring the vibration displacement of a lower box body, a high-temperature airflow temperature sensor, a first tension sensor for monitoring the tension required by vibration provided for the lower box body of the intelligent drying, grinding and selecting integrated machine, a first vibration frequency sensor and an infrared sensor for monitoring the temperature in the box body of the intelligent drying, grinding and selecting integrated machine, and the drying, grinding and selecting of the tailings, the drying, grinding and selecting of the materials under the selection of the tailings and the activation of the materials on the intelligent drying, grinding and selecting integrated machine are realized through the intelligent drying, grinding and selecting integrated machine;
the intelligent activation machine is provided with a second current sensor, a second voltage sensor, a second displacement sensor for monitoring the vibration displacement of an activation chamber (specifically a grinding tool and a vibratable grinding plate), a temperature sensor for detecting the temperature of the activation chamber, a second tension sensor for monitoring the tension required by vibration provided for the activation chamber, a second vibration frequency sensor and an activator injection quantity sensor.
The tailing comprehensive utilization intelligent production execution system comprises:
the first module is in data communication with the intelligent drying, grinding and selecting integrated machine and the intelligent activation machine, is used for establishing an RNN-ELM combined model by taking the output ends of a plurality of extreme learning ELM models as the input ends of a plurality of corresponding time sequence nodes of a Recurrent Neural Network (RNN) so as to intelligently predict the optimal activation process and activity index of the tailings with unknown composition,
the second module is in data communication with each sensor of the intelligent drying, grinding and selecting integrated machine and the intelligent activating machine, independently uses an ELM algorithm to carry out real-time monitoring on equipment faults, independently uses an LSTM algorithm to carry out real-time monitoring on the energy consumption of the system,
a third module for performing unified management on product quality management, equipment management, energy consumption management and inventory management, an
The system comprises a whole plant management and control module, a user management module and an information main display screen.
In one embodiment, the intelligent integrated drying, grinding and sorting machine specifically comprises:
an outer frame, a mill consisting of a vibration-assisted multilayer grinding mechanism and a suspended grinding box, a multipoint air inlet drying device and a circulating powder selecting system, wherein,
the vibration-assisted multilayer grinding mechanism adopts an energy-saving grinder with a multilayer structure design, the energy-saving grinder comprises a layer plate and a grinding body, the layer plate grinds materials in a layered manner, and the materials are ground in a vibration manner through the vibration of the layer plate and the grinding body; wherein, the plywood includes a plurality of parallel sheet layers from last parallel arrangement down, and except bottom plate layer, every sheet layer all has the unanimous via hole of a plurality of diameters in this sheet layer, and the via hole diameter reduces from top plate layer to penultimate sheet layer in proper order, preferably, reduce to the linearity and reduce. The grinding body is a plurality of grinding blocks which are paired with a plurality of plate layers in a one-to-one mode through first gaps and provided with at least one first through hole, so that the step-by-step screen grinding of each plate layer is realized. Preferably, the first through holes are uniformly 1-3cm in diameter. The maximum diameter of the via hole is 3-5mm, the minimum diameter of the via hole is 0.2-1mm, and the size of the first gap is not larger than the diameter of the via hole of the matched plate layer when measured in a non-vibration state.
After materials enter an intelligent 'drying, grinding and sorting' all-in-one machine, the materials firstly fall onto the uppermost plate layer in vibration through the at least one first through hole of the grinding block on the uppermost layer in vibration to be ground, fine tailing blocks firstly fall into the second plate layer in vibration through the at least one first through hole on the second grinding block below the second grinding block, and so on until the fine tailing blocks cannot pass through the through hole of one plate layer and are left in the plate layer to be ground, larger blocks are gradually ground to fine-diameter powder through the plurality of first through holes after the large blocks pass through the vibration grinding blocks on the uppermost layer and the plate layer after the large blocks fall into the plate layer, and finally all the large blocks fall onto the bottommost plate layer without the through holes to be ground finally.
The vibration grinding is adopted, so that the material dispersion degree is better, and the grinding efficiency of the energy-saving grinder and the specific surface area of the micro powder are greatly improved.
The suspension type grinding box is divided into an upper box body and a lower box body, the lower box body is used for transmitting vibration energy for each grinding block and each plate layer, the elastic mechanism is adopted to be in soft connection with the outer frame and hung on the outer frame, independent vibration of the lower box body is achieved, and the lower box body is excited more easily. When the amplitude is enhanced, the vibration reduction effect can be achieved, and when the amplitude is weakened, the vibration lifting effect can be achieved, so that the lower box body is more stable in the vibration process.
The multi-point type air inlet drying device inputs high-temperature air to vertical multiple points in the upper box body, and the high-temperature air is transmitted to each interlayer in the lower box body, so that the whole grinding box is filled with the high-temperature air, the materials can be fully dried in the grinding process, the moisture content of the materials is greatly reduced, the grinding effect is improved, and the production efficiency is improved.
The circulating powder selecting system is connected with the mill, the ground powder enters the powder selecting system under the action of high-temperature air, magnetic separation, iron removal and screening are sequentially carried out, the selected micro powder according with the granularity is collected by the bag dust collector and enters the micro powder bin, and the selected material after powder selection is input into the drying and grinding system again for drying and grinding treatment, so that the material is fully utilized.
In one embodiment, the intelligent activation machine comprises a material stabilizing bin, a metering device and an activation chamber, wherein the conveying belt arranged in a sealed channel is used for conveying the micro powder in the micro powder bin into the material stabilizing bin, the micro powder is weighed by the metering device and then transferred into the activation chamber, the activation functional material and water are mixed and heated in proportion, and the mixture is pneumatically conveyed to the surface of the tailing micro powder sprayed into the activation chamber through atomization.
Wherein the activation chamber has at least one vibratable grinding plate, at least one vibratable grinding tool provided with at least one second through hole paired with a second gap of the at least one vibratable grinding plate, and an atomizing pneumatic feed spray device. The second gap is sized to be no greater than 0.5mm measured in a non-vibrating state.
The micro powder is transferred into the activation chamber, enters the at least one second through hole to reach the at least one vibratable grinding plate, is simultaneously conveyed into the spraying device through atomization pneumatic conveying to be sprayed into the second through hole to be sprayed onto the surface of the tailing micro powder entering the at least one second through hole, and the sprayed wet micro powder is continuously spread on the surface of the vibratable grinding plate through the vibration of the grinding tool and the vibratable grinding plate, so that the activating agent is uniformly distributed on the surface of the micro powder and generates chemical reaction, and the activity of the tailing micro powder is improved to the maximum extent.
The real-time monitoring of the equipment fault by the ELM algorithm specifically comprises the following steps:
s1, acquiring abnormal data of the intelligent drying, grinding and selecting integrated machine and the intelligent activating machine, and establishing a mapping relation between the fault and the characteristic parameters of the abnormal data after analysis to form a fault library;
s2, performing offline big data analysis and feature extraction on the abnormal data; and finally, calculating through an ELM extreme learning algorithm, comparing a calculation result with a fault library, and determining the fault type of the equipment.
Wherein, S1 specifically includes:
s1-1, the second module respectively collects data of a first current sensor, a first voltage sensor, a first displacement sensor, a high-temperature airflow temperature sensor, a first tension sensor, a first vibration frequency sensor and an infrared sensor, and the data of a second current sensor, a second voltage sensor, a second displacement sensor, a temperature sensor, a second tension sensor, a second vibration frequency sensor and an activator injection quantity sensor in real time, and typical abnormal data are continuously accumulated;
s1-2, extracting the characteristic of the typical abnormal data, and normalizing the extracted characteristic, wherein the normalization of the extracted characteristic of the data of the first and second current sensors and the first and second voltage sensors is per unit, and the normalization of the extracted characteristic of the data of the remaining sensors is a linear function normalization, forming a plurality of typical characteristic parameter column vectors R ═ u (u ═ f) 1c u 2c i 1c i 2c s 1c s 2c T 1c T 2c In c F 1c F 2c ) T Wherein u is 1c 、u 2c 、i 1c 、i 2c 、s 1c 、s 2c 、T 1c 、T 2c 、In c 、F 1c 、F 2c Respectively a first second voltage characteristic parameter, a first second current characteristic parameter, a first second displacement characteristic parameter, a high-temperature airflow temperature characteristic parameter, an air temperature characteristic parameter in the box body, an infrared sensing data characteristic parameter and a first second tension characteristic parameter;
s1-3, establishing mapping relation between fault and abnormal data characteristic parameter
A failure bank H will be formed.
S2 specifically includes:
s2-1, the second module collects abnormal data of a first current sensor, a first voltage sensor, a first displacement sensor, a high-temperature airflow temperature sensor, a first tension sensor, a first vibration frequency sensor and an infrared sensor, and the second current sensor, the second voltage sensor, a second displacement sensor, a temperature sensor, a second tension sensor, a second vibration frequency sensor and an activator injection quantity sensor in real time, and accumulates the abnormal data continuously;
s2-2, extracting the abnormal data, and normalizing the extracted features, wherein the normalization of the extracted features of the first and second current sensors and the first and second voltage sensors is per unit, and the normalization of the extracted features of the remaining sensors is a linear function normalization, forming a plurality of characteristic parameter column vectors r ═ u (u ═ u- 1 u 2 i 1 i 2 s 1 s 2 T 1 T 2 In F 1 F 2 ) T Forming a training set and a verification set, wherein the ratio of the training set to the verification set is 5:1-2: 1;
s2-3 with r ═ u (u) 1 u 2 i 1 i 2 s 1 s 2 T 1 T 2 In F 1 F 2 ) T Is an input end of the input device,
establishing an ELM model for an output end by using a training set, adjusting ELM network parameters by back propagation by using a random gradient descent method, verifying an error function value by using a verification set, and finishing model training when the error function value is minimum and stable, wherein the ELM model is
s2-4, the second module acquires the parameters of each sensor in real time and extracts and standardizes the parameters through parameter characteristicsThen the model is substituted into the trained ELM model to obtain a predicted value
Comparing with the fault library and utilizing the mapping relation
And determining the equipment fault category.
The real-time monitoring of the energy consumption of the system by the LSTM algorithm specifically comprises the following steps:
p1 the second module collects data of the first current sensor, the first voltage sensor, the second current sensor and the second voltage sensor in real time and obtains a plurality of energy consumption time-series data C during the same period in a plurality of consecutive days i (t),i∈[1,M],M∈[30,365]M is the number of samples, and the samples are divided into a training set and a verification set, wherein the ratio of the training set to the verification set is 5:1-2: 1;
p2 constructing LSTM model, selecting initial time t 0 And a target time t g Time series data of energy consumption of time interval
Energy consumption C of initial moment i (t 0 ) Energy consumption of each time node C i (t j ) Sequence target moment energy consumption C i (t g ) Substituting the LSTM model to obtain the predicted energy consumption of each time node
P3 cross entropy loss error
Carrying out backward propagation on the time points by a gradient descent method to adjust the network parameters of each time node, verifying the cross entropy loss error variation by using a verification set, and obtaining the absolute value L i+1 -L i Finish training | → 0. Wherein N is t Is the total number of time nodes, L ij And
respectively the ith training sample j The cross entropy loss error and the energy consumption predicted value at the moment are logarithms with e as a base;
and P4, the second module continuously substitutes the energy consumption data collected in real time into the well-trained LSTM model to predict the energy consumption value at any target moment.
The third module comprises a production line monitoring module, an equipment management module, an energy consumption management module, an inventory management module, a quality management module, a production data statistical report module and an access control system, the specific operation process comprises that the upper ERP transmits production order information to the third module for processing through an Enterprise Service Bus (ESB), then the production information is transmitted to each module terminal, and the production information is interacted with the lower production equipment including tailing intelligent production equipment through an SCADA system to complete the intelligent management and control of the whole plant,
the staff enters a working area through an IC electronic card access control, checks a plan report, carries out a detection process flow set by a quality management module through a production line monitoring module, generates a report through a production data statistical report module and completes a detection report so as to determine whether early warning is needed or not, and inputs detection report data and early warning information into a database.
Advantageous effects
1. The intelligent algorithm is adopted to monitor and adjust the tailing activation process in real time and monitor equipment failure and energy consumption, thereby realizing the intelligent activation production of the tailing,
2. adopts an integrated processing mode of drying, grinding and selecting, separates the steps of magnetic separation and crushing and puts the steps into the step of powder selection, ensures the stability of the quality of the product,
3. and an intelligent production execution system is adopted to realize high-quality and high-efficiency production of the tailing activation process flow.
Drawings
FIG. 1 is a schematic diagram of a tailings activation process system based on drying, grinding and separation integrated intelligent production equipment,
FIG. 2 is a structural diagram of an intelligent drying, grinding and sorting all-in-one machine A of the invention,
figure 3a is a top view structural diagram of the vibration assisted multi-layer polishing mechanism,
FIG. 3b is a side view of the structure of the vibration-assisted multi-layer polishing device,
FIG. 4a is a schematic top view of the vibration structure of the vibration-assisted multi-layer grinding mechanism,
FIG. 4b is a side view of the inner box of the intelligent drying, grinding and polishing all-in-one machine A of the invention,
FIG. 5 is a flow chart of a method for monitoring equipment failure in real time by using an ELM algorithm,
FIG. 6 is a schematic diagram of a process for training an ELM model,
figure 7 is a schematic diagram of the process of training the LSTM model,
FIG. 8 shows an energy consumption comparison real-time line graph and an energy consumption prediction point displayed by a third module of the comprehensive utilization intelligent production execution system for tailings,
FIG. 9 is a flow chart of a tailings comprehensive utilization production execution system of a third module of the tailings comprehensive utilization intelligent production execution system,
FIG. 10 is an interface for displaying real-time information on a display screen by comprehensively utilizing all modules in the third module of the intelligent production execution system and data statistics,
the device comprises an outer frame 1, an upper box 2, a lower box 3, an elastic mechanism 4, a multi-point air inlet drying device 5, a magnetic separator 6, a vibrating screen 7, a feeding port 8, an energy-saving mill 9, a laminate 9-1, a grinding body 9-2, a via hole 9-3, a first through hole 9-4, a first connecting piece 10, an inner box 10-1, a second connecting piece 11, an outer box 11-1, a second connecting rod 12, a first connecting rod 13 and a penetrating strip 14.
Detailed Description
Example 1
Fig. 1 shows a layout mode of a tailing activation process system based on drying, grinding and sorting integrated intelligent production equipment, which comprises an intelligent drying, grinding and sorting all-in-one machine A, an intelligent activation machine B connected through a closed channel D provided with a conveyor belt, and an intelligent production execution system C for comprehensive utilization of tailings.
The structure of the intelligent 'drying, grinding and selecting' all-in-one machine A is shown in figure 2, and comprises the following components: the powder separation device comprises an outer frame 1, a grinding machine consisting of a vibration-assisted multilayer grinding mechanism and a suspension type grinding box, a multipoint air inlet drying device 5 and a circulating powder separation system, wherein the suspension type grinding box is divided into an upper box body 2 and a lower box body 3, the lower box body 3 is flexibly connected with the outer frame 1 through an elastic mechanism 4 and is suspended on the outer frame 1, independent vibration of the lower box body 3 is realized, and the lower box body 3 is excited more easily. When the amplitude is enhanced, the vibration reduction effect can be achieved, and when the amplitude is weakened, the vibration lifting effect can be achieved, so that the lower box body is more stable in the vibration process.
The vibration-assisted multilayer grinding mechanism adopts an energy-saving grinder 9 with a multilayer structure design, as shown in fig. 3a, the energy-saving grinder 9 comprises a layer plate 9-1 and a grinding body 9-2, the layer plate 9-1 grinds materials in a layered manner, and the materials are ground in a vibration manner through the vibration of the layer plate 9-1 and the grinding body 9-2. The double-headed arrows in fig. 3a depict the direction of movement of the reciprocating vibrations.
As shown in fig. 3a and 3b, the laminate 9-1 comprises 5 parallel plate layers arranged in parallel from top to bottom, each plate layer is provided with 8 through holes 9-3 with the same diameter in the plate layer except for the bottommost plate layer, and the diameter of the through holes 9-3 is reduced from the topmost plate layer to the penultimate plate layer in a half-reducing manner. The direction of vibration of the plurality of laminae 9-1 is perpendicular to the paper.
The grinding body 9-2 is 5 grinding blocks having 3 first through holes 9-4 paired with 5 of the slabs 9-1 one by one with a first gap (shown in fig. 3 b) to realize stepwise screen grinding of each slab.
The diameters of the first through holes are consistent to be 3 cm. The via diameter is at most 3mm and at most 0.5mm, the first gap size is not greater than the via diameter of the mated plies as measured in a non-vibrating state, and preferably decreases progressively from the uppermost ply up to the penultimate ply.
As shown in FIG. 3b, each laminate 9-1 is fixedly provided with a first connecting member 10 and a second connecting member 11 to four sides of the abrasive body 9-2. The lower case 3 is composed of an outer case 11-1 and an inner case 10-1 as shown in fig. 4a, and serves to transmit vibration energy to the respective grinding blocks 9-2 and the slab 9-1. The energy-saving mill 9 is positioned in a grinding cavity surrounded by the inner box body 10-1. The outer box 11-1 and the inner box 10-1 are connected to the second connecting member 11 and the first connecting member 10 through 20 second connecting rods 12 and 10 first connecting rods 13, respectively (or connected to the first connecting member 10 and the second connecting member 11, respectively, at this time, as will be understood from the following, 10 penetrating strips are provided on the remaining two sides of the inner box 10-1 in fig. 4a to allow the 10 first connecting rods 13 to penetrate through and be driven by the outer box 11-1 to vibrate transversely and reciprocally).
As shown in fig. 4a or fig. 4 b. The 20 second connecting rods 12 are connected with the second connecting piece 11 through 10 penetrating strips 14 which penetrate through two opposite side surfaces of the inner box body 10-1, so that the outer box body 11-1 vibrates to drive the 20 second connecting rods 12 to transversely vibrate in the 10 penetrating strips 14 in a reciprocating mode, and each layer plate 9-1 is driven to transversely vibrate in a reciprocating mode.
And the 10 first connecting rods 13 are connected with the two opposite side surfaces of the rest of the inner box body 10-1, which are not provided with the through strips 14, and the inner box body 10-1 vibrates to drive the 10 first connecting rods 13 to vibrate transversely and reciprocally, so as to drive each grinding body 9-2 to vibrate transversely and reciprocally.
After materials enter the intelligent drying, grinding and sorting all-in-one machine A from the feeding port 8 (figure 2), the materials firstly fall on the uppermost plate layer in vibration through 3 first through holes of the grinding block on the uppermost layer in vibration to be ground as shown in figure 3b, the fine tailing blocks firstly fall on the second grinding block in vibration below through 3 first through holes to reach the second plate layer in vibration, and so on until the fine tailing blocks cannot pass through the through hole of one plate layer and are left in the plate layer to be ground, the larger blocks are gradually ground to fine-diameter powder under the vibration grinding of the grinding block on the uppermost layer and the plate layer after passing through the first through holes and the plate layer after falling, and finally all the fine tailing blocks fall on the bottommost plate layer without the through holes to be ground finally.
The multipoint type air inlet drying device 5 inputs high-temperature air to vertical multiple points in the upper box body 2, and the high-temperature air is transmitted to each interlayer in the lower box body 3, so that the whole grinding box is filled with the high-temperature air, the materials can be fully dried in the grinding process, the moisture content of the materials is greatly reduced, the grinding effect is improved, and the production efficiency is improved.
The circulating powder selecting system is connected with the mill and comprises a magnetic separator 6 and a vibrating screen 7, the ground powder enters the powder selecting system under the action of high-temperature air, magnetic separation and iron removal and screening are sequentially carried out, the selected micro powder which accords with the granularity is collected through a bag dust collector (not shown) and enters a micro powder bin d, and the selected material after powder selection is input into the drying and grinding system again for drying and grinding treatment, so that the material is fully utilized.
Example 2
As shown in fig. 5, the real-time monitoring of the device failure by the ELM algorithm specifically includes:
s1, acquiring abnormal data of the intelligent drying, grinding and selecting integrated machine and the intelligent activating machine, and establishing a mapping relation between the fault and the characteristic parameters of the abnormal data after analysis to form a fault library;
s2, performing offline big data analysis and feature extraction on the abnormal data; and finally, calculating through an ELM extreme learning algorithm, comparing a calculation result with a fault library, and determining the fault type of the equipment.
Wherein, S1 specifically includes:
s1-1, the second module respectively collects data of a first current sensor, a first voltage sensor, a first displacement sensor, a high-temperature airflow temperature sensor, a first tension sensor, a first vibration frequency sensor and an infrared sensor, and the data of a second current sensor, a second voltage sensor, a second displacement sensor, a temperature sensor, a second tension sensor, a second vibration frequency sensor and an activator injection quantity sensor in real time, and typical abnormal data are continuously accumulated;
s1-2, extracting the characteristic of the typical abnormal data, and normalizing the extracted characteristic, wherein the normalization of the extracted characteristic of the data of the first and second current sensors and the first and second voltage sensors is per unit, and the normalization of the extracted characteristic of the data of the remaining sensors is a linear function normalization, so as to form a plurality of typical characteristic parameter column vectors R ═ u (u is u) 1c u 2c i 1c i 2c s 1c s 2c T 1c T 2c In c F 1c F 2c ) T Wherein u is 1c 、u 2c 、i 1c 、i 2c 、s 1c 、s 2c 、T 1c 、T 2c 、In c 、F 1c 、F 2c Respectively a first second voltage characteristic parameter, a first second current characteristic parameter, a first second displacement characteristic parameter, a high-temperature airflow temperature characteristic parameter, an air temperature characteristic parameter in the box body, an infrared sensing data characteristic parameter and a first second tension characteristic parameter;
s1-3, establishing mapping relation between fault and abnormal data characteristic parameter
A failure bank H will be formed.
S2 specifically includes:
s2-1, the second module collects abnormal data of a first current sensor, a first voltage sensor, a first displacement sensor, a high-temperature airflow temperature sensor, a first tension sensor, a first vibration frequency sensor and an infrared sensor, and the second current sensor, the second voltage sensor, a second displacement sensor, a temperature sensor, a second tension sensor, a second vibration frequency sensor and an activator injection quantity sensor in real time, and accumulates the abnormal data continuously;
s2-2, extracting the abnormal data, and normalizing the extracted features, wherein the normalization of the extracted features of the first and second current sensors and the first and second voltage sensors is per unit, and the normalization of the extracted features of the remaining sensors is a linear function normalization, forming a plurality of characteristic parameter column vectors r ═ u (u ═ u- 1 u 2 i 1 i 2 s 1 s 2 T 1 T 2 In F 1 F 2 ) T Forming a training set and a verification set, wherein the ratio of the training set to the verification set is 3: 1;
s2-3 is as shown in fig. 6, with r ═ u (u) 1 u 2 i 1 i 2 s 1 s 2 T 1 T 2 In F 1 F 2 ) T Is an input end of the input device,
establishing an ELM model for an output end by using a training set, adjusting ELM network parameters by back propagation by using a random gradient descent method, verifying an error function value by using a verification set, and finishing model training when the error function value is minimum and stable, wherein the ELM model is
s2-4, the second module acquires parameters of each sensor in real time, and the parameters are input into a trained ELM model after parameter feature extraction and standardization processing to obtain a predicted value
Comparing with the fault library by using the mapping relation
The device failure category is determined (see fig. 6).
Example 3
The real-time monitoring of the energy consumption of the system by the LSTM algorithm specifically comprises the following steps:
p1 the second module collects data of the first current sensor, the first voltage sensor, the second current sensor and the second voltage sensor in real time and obtains a plurality of energy consumption time-series data C during the same period in a plurality of consecutive days i (t),i∈[1,90]Number of samples 90, willThe method comprises the following steps of dividing the training set into a training set and a verification set, wherein the ratio of the training set to the verification set is 4: 1;
p2 As shown in FIG. 7, an LSTM model is constructed, and an initial time t is selected 0 And a target time t g Time series data of energy consumption of time interval
Energy consumption C of initial moment i (t 0 ) Energy consumption of each time node C i (t j ) Sequence target moment energy consumption C i (t g ) Substituting the LSTM model to obtain the predicted energy consumption of each time node
FIG. 7 contains the states of cells C downloaded by each time node to the next time node 0 、C 1 、C j-1 、C j 、
P3 cross entropy loss error
Carrying out backward propagation (indicated by a backward arrow) on the time points by a gradient descent method to adjust network parameters of each time node, and verifying the cross entropy loss error variation by using a verification set when L i+1 -L i Finish training | → 0. Wherein N is t Is the total number of time nodes, L ij And
respectively the ith training sample j The cross entropy loss error and the energy consumption predicted value at the moment are logarithms with e as a base;
and P4, the second module continuously substitutes the energy consumption data collected in real time into the well-trained LSTM model to predict the energy consumption value at any target moment. As shown in fig. 8, two predicted points of the gas and water energy consumptions from 11/9/2021 to 12/0/9/2021 to 13/0/9/2021 are circled in the energy consumption comparison real-time line graph.
Example 4
The third module of the intelligent production execution system for comprehensive utilization of tailings uses Vue.js as a front-end framework in a Windows 10 system environment, the system adopts a B/S architecture, a Visual Studio Code as a development tool and MySql as a data management library.
Fig. 9 shows a tailing integrated utilization production execution system flow chart of the third module of the tailing integrated utilization intelligent production execution system. The upper ERP transmits the information of the production order and the like to MES through an enterprise service bus ESB, the MES carries out corresponding processing, and then the production information is transmitted to each module terminal. The quality management module in the core module is responsible for making a detection process flow according to an optimal activation process and an activity index of unknown tailings formed by establishing an RNN-ELM combined model of the first module, after a plan table is obtained through entrance guard approach, the plan report is checked, the detection process flow set by the quality management module is carried out through a production line monitoring module, a report is generated and a detection report is completed through a production data statistical report module, whether early warning is needed or not is determined, and detection report data and early warning information are recorded into a database.
The equipment management module monitors the tailing intelligent production equipment to perform ELM identification on the equipment state, complete fault monitoring, perform tracking and early warning on equipment maintenance and record early warning information into a database.
On the other hand, the energy consumption management module can monitor the operation of the power distribution room and the water pump according to the LSTM prediction, so that the obtained electric parameters of the instrument and the parameters of all sensors in the tailing intelligent production equipment collected by the second module are recorded into a database through an OPC protocol.
The production tracking operation process of the employee also forms flow data to be recorded in the database. The stock management module records the operation records of a feeding system for feeding raw materials into the intelligent drying, grinding and sorting all-in-one machine A and finished product warehousing information into a database one by one to form a complete material activation process data chain from materials to products. And finally, managing the loading plans of the raw materials and the finished products, and recording the monitoring of entering vehicles, loading and leaving the warehouse and leaving the factory.
Fig. 10 shows an interface for displaying real-time information of statistical analysis of data and various modules in the third module of the comprehensive utilization intelligent production execution system for tailings on a display screen. After the tailings of the embodiment 1 use the comprehensive utilization intelligent production execution system, the product quality is obviously improved, and the abnormal conditions of equipment and a production line are obviously reduced. With the implementation depth, the system achieves the following effects:
1) the production process of enterprises is effectively managed, and through implementation of a production management information system, the enterprises establish a digital model taking tailing micro powder preparation as a main resource, so that uniform management of workshop resources is realized. By tracking the IC electronic card, task execution information is recorded and tracked, workers receive tasks, report information and feed back abnormal conditions through a field manufacturing terminal, managers monitor workshop conditions through a production real-time board, production process paperless is improved, meanwhile, average processing time of field problems is shortened by 25%, and production plan execution rate is effectively improved by 15%.
2) The system refines the work of quality inspectors, and realizes the whole-course standardization of enterprise quality management by acquiring related quality information, so that the product quality of an enterprise is improved by 10 percent on average.
3) The system provides the report forms of all aspects for statistical analysis of production information, traceability of production historical data is achieved, internal problems in the production process are found continuously through statistical analysis, therefore, the self-learning function of the production line is achieved, weak links of enterprise production are improved continuously, and production benefits of enterprises are improved.
Claims (10)
1. The tailing activation process system based on drying, grinding and selecting integrated intelligent production equipment is characterized by comprising tailing intelligent production equipment and used for carrying out the following activation processes on tailings, namely drying, grinding and selecting treatment, drying, grinding and selecting circulation treatment of materials under selecting and selecting, activation treatment of materials on selecting and
the tailings are comprehensively utilized by an intelligent production execution system, wherein the powder selection treatment comprises magnetic separation and screening which are sequentially carried out;
the tailing intelligent production equipment comprises an intelligent 'drying, grinding and selecting' all-in-one machine and an intelligent activating machine which are sequentially connected, wherein,
the intelligent drying, grinding and selecting integrated machine is provided with a first current sensor, a first voltage sensor, a first displacement sensor for monitoring the vibration displacement of a lower box body, a high-temperature airflow temperature sensor, a first tension sensor for monitoring the tension required by vibration provided for the lower box body of the intelligent drying, grinding and selecting integrated machine, a first vibration frequency sensor and an infrared sensor for monitoring the temperature in the box body of the intelligent drying, grinding and selecting integrated machine, and the drying, grinding and selecting of the tailings, the drying, grinding and selecting of the materials under the selection of the tailings and the activation of the materials on the intelligent drying, grinding and selecting integrated machine are realized through the intelligent drying, grinding and selecting integrated machine;
the intelligent activation machine is provided with a second current sensor, a second voltage sensor, a second displacement sensor for monitoring the vibration displacement of the activation chamber, a temperature sensor for detecting the temperature of the activation chamber, a second tension sensor for monitoring the tension required by the vibration of the activation chamber, a second vibration frequency sensor and an activator injection quantity sensor; the tailing comprehensive utilization intelligent production execution system comprises:
the first module is in data communication with the intelligent drying, grinding and selecting integrated machine and the intelligent activation machine, is used for establishing an RNN-ELM combined model by taking the output ends of a plurality of extreme learning ELM models as the input ends of a plurality of corresponding time sequence nodes of a Recurrent Neural Network (RNN) so as to intelligently predict the optimal activation process and activity index of the tailings with unknown composition,
the second module is in data communication with each sensor of the intelligent drying, grinding and selecting integrated machine and the intelligent activating machine, independently uses an ELM algorithm to carry out real-time monitoring on equipment faults, independently uses an LSTM algorithm to carry out real-time monitoring on the energy consumption of the system,
a third module for performing unified management on product quality management, equipment management, energy consumption management and inventory management, an
The system comprises a whole plant management and control module, a user management module and an information main display screen.
2. The system of claim 1, wherein the screening comprises rolling screening or vibratory screening.
3. The system of claim 1, wherein the intelligent "bake mill" all-in-one machine specifically comprises:
an outer frame, a mill consisting of a vibration-assisted multilayer grinding mechanism and a suspended grinding box, a multipoint air inlet drying device and a circulating powder selecting system, wherein,
the vibration-assisted multilayer grinding mechanism adopts an energy-saving grinder with a multilayer structure design, the energy-saving grinder comprises a layer plate and a grinding body, the layer plate grinds materials in a layered manner, and the materials are ground in a vibration manner through the vibration of the layer plate and the grinding body; the laminated plate comprises a plurality of parallel plate layers which are arranged in parallel from top to bottom, each plate layer is provided with a plurality of through holes with the same diameter in the plate layer except for the bottommost plate layer, the diameters of the through holes are sequentially reduced from the topmost plate layer to the penultimate plate layer, and the grinding bodies are a plurality of grinding blocks which are paired with the plurality of plate layers one by one at first gaps and provided with at least one first through hole, so that the plate layers are gradually screened and ground;
the suspension type grinding box is divided into an upper box body and a lower box body, the lower box body is used for transmitting vibration energy to each grinding block and each plate layer, and the lower box body is flexibly connected with the outer frame by adopting an elastic mechanism and is suspended on the outer frame, so that the independent vibration of the lower box body is realized, and the lower box body is more easily excited;
the multipoint type air inlet drying device inputs high-temperature air to the upper box body vertically at multiple points, and the high-temperature air is transmitted to each interlayer in the lower box body, so that the whole grinding box is filled with the high-temperature air, and materials can be fully dried in the grinding process;
the circulating powder selecting system is connected with the mill, the ground powder enters the powder selecting system under the action of high-temperature air, magnetic separation and iron removal and screening are sequentially carried out, the selected micro powder which accords with the granularity is collected by the bag dust collector and enters the micro powder bin, and the selected material after powder selection is input into the drying and grinding system again for drying and grinding.
4. The system of claim 3, wherein the lower box body is composed of an outer box body and an inner box body, the energy-saving mill is positioned in a grinding cavity surrounded by the inner box body, a second connecting piece and a first connecting piece are respectively fixedly arranged on four sides of each layer plate and the grinding body, the outer box body and the inner box body are respectively connected with the second connecting piece and the first connecting piece through a plurality of second connecting rods and a plurality of first connecting rods, the plurality of second connecting rods are connected with the second connecting pieces through a plurality of penetrating strips penetrating through two opposite sides of the inner box body, the plurality of second connecting rods are driven by the vibration of the outer box body to transversely reciprocate in the plurality of penetrating strips, so as to drive each layer plate to transversely reciprocate, and the plurality of first connecting rods are connected with two opposite sides of the inner box body which are not provided with the penetrating strips, so as to drive the plurality of first connecting rods to transversely reciprocate by the vibration of the inner box body, thereby driving each grinding body to vibrate transversely and reciprocally, or,
the outer box body and the inner box body are respectively connected with a first connecting piece and a second connecting piece through a plurality of second connecting rods and a plurality of first connecting rods, a plurality of penetrating strips are arranged on the two residual side surfaces of the inner box body to allow the first connecting rods to penetrate through and are driven by the outer box body to enable each laminate to transversely vibrate in a reciprocating manner, the two residual opposite side surfaces of the inner box body, which are not provided with the penetrating strips, are connected with the first connecting piece of each grinding body through the second connecting rods, and the inner box body vibrates to drive each grinding body to transversely vibrate in a reciprocating manner,
the intelligent activation machine comprises a material stabilizing bin, a metering device and an activation chamber, wherein the micro powder in the micro powder bin is conveyed into the material stabilizing bin by a conveying belt arranged on a sealed channel, the micro powder is weighed by the metering device and then transferred into the activation chamber, the activated functional material and water are mixed and heated in proportion, and the mixture is pneumatically conveyed to the surface of the tailing micro powder sprayed into the activation chamber through atomization.
5. The system of claim 3 or 4, wherein the reduction is a linear reduction, the first through hole diameter is consistently 1-3cm, the via hole diameter is 3-5mm at maximum and 0.2-1mm at minimum, and the first gap size is no greater than the via hole diameter of the mated plies measured in a non-vibrating state.
6. System according to claim 3 or 4, characterized in that the activation chamber has at least one vibratable grinding plate, at least one vibratable grinding tool provided with at least one second through hole paired with a second gap of the at least one vibratable grinding plate, the second gap having a size measured in a non-vibrating state not larger than 0.5mm, and an atomizing pneumatic feed spraying device.
7. The system of claim 1, wherein the real-time monitoring of the device failure by the ELM algorithm specifically comprises:
s1, acquiring abnormal data of the intelligent integrated drying, grinding and selecting machine and the intelligent activating machine, and establishing a mapping relation between the fault and the characteristic parameters of the abnormal data after analysis to form a fault library;
s2, performing offline big data analysis and feature extraction on the abnormal data; and finally, calculating through an ELM extreme learning algorithm, comparing a calculation result with a fault library, and determining the fault type of the equipment.
8. The system according to claim 7, wherein S1 specifically includes:
s1-1, the second module respectively collects data of a first current sensor, a first voltage sensor, a first displacement sensor, a high-temperature airflow temperature sensor, a first tension sensor, a first vibration frequency sensor and an infrared sensor, and the data of a second current sensor, a second voltage sensor, a second displacement sensor, a temperature sensor, a second tension sensor, a second vibration frequency sensor and an activator injection quantity sensor in real time, and typical abnormal data are continuously accumulated;
s1-2, extracting the characteristic of the typical abnormal data, and normalizing the extracted characteristic, wherein the normalization of the extracted characteristic of the data of the first and second current sensors and the first and second voltage sensors is per unit, and the normalization of the extracted characteristic of the data of the remaining sensors is a linear function normalization, forming a plurality of typical characteristic parameter column vectors R ═ u (u ═ f) 1c u 2c i 1c i 2c s 1c s 2c T 1c T 2c In c F 1c F 2c ) T Wherein u is 1c 、u 2c 、i 1c 、i 2c 、s 1c 、s 2c 、T 1c 、T 2c 、In c 、F 1c 、F 2c Respectively a first and a second voltage characteristic parameter, a first and a second current characteristic parameter, a first and a second displacement characteristic parameter, a high-temperature airflow temperature characteristic parameter, an air temperature characteristic parameter in the box body, an infrared sensing data characteristic parameter and a first and a second tension characteristic parameter;
s1-3, establishing a mapping relation f (R) between the fault and the abnormal data characteristic parameter:
a fault bank H is formed;
s2 specifically includes:
s2-1, the second module collects abnormal data of a first current sensor, a first voltage sensor, a first displacement sensor, a high-temperature airflow temperature sensor, a first tension sensor, a first vibration frequency sensor and an infrared sensor, and the second current sensor, the second voltage sensor, a second displacement sensor, a temperature sensor, a second tension sensor, a second vibration frequency sensor and an activator injection quantity sensor in real time, and accumulates the abnormal data continuously;
s2-2, extracting the characteristics of the abnormal data and standardizing the extracted characteristics, whereinThe normalization process for the extracted features of the data of the first and second current sensors and the first and second voltage sensors is a per unit process, and the normalization process for the extracted features of the data of the remaining sensors is a linear function normalization process to form a plurality of characteristic parameter column vectors r ═ u (u) 1 u 2 i 1 i 2 s 1 s 2 T 1 T 2 In F 1 F 2 ) T Forming a training set and a verification set, wherein the ratio of the training set to the verification set is 5:1-2: 1;
s2-3 with r ═ u 1 u 2 i 1 i 2 s 1 s 2 T 1 T 2 In F 1 F 2 ) T Is an input end of the input device,
establishing an ELM model for an output end by using a training set, adjusting ELM network parameters by back propagation by using a random gradient descent method, verifying an error function value by using a verification set, and finishing model training when the error function value tends to be minimum and stable, wherein the ELM model is
s2-4, the second module acquires parameters of each sensor in real time, and the parameters are input into a trained ELM model after parameter feature extraction and standardization processing to obtain a predicted value
Comparing with a fault library, and utilizing the mapping relation f (R):
and determining the equipment fault category.
9. The system according to any one of claims 1-8, wherein the real-time monitoring of the energy consumption of the system by the LSTM algorithm specifically comprises:
p1 the second module collects data of the first current sensor, the first voltage sensor, the second current sensor and the second voltage sensor in real time and obtains a plurality of energy consumption time series data C during the same period in a plurality of consecutive days i (t),i∈[1,M],M∈[30,365]M is the number of samples, and the samples are divided into a training set and a verification set, wherein the ratio of the training set to the verification set is 5:1-2: 1;
p2 constructing LSTM model, selecting initial time t 0 And a target time t g Time series data of energy consumption of time interval
Energy consumption C of initial moment i (t 0 ) Energy consumption of each time node C i (t j ) Sequence target moment energy consumption C i (t g ) Substituting the LSTM model to obtain the predicted energy consumption of each time node
P3 cross entropy loss error
Carrying out backward propagation on the time points by a gradient descent method to adjust the network parameters of each time node, verifying the cross entropy loss error variation by using a verification set, and obtaining the absolute value L i+1 -L i Training is finished when | → 0; wherein N is t Is the total number of time nodes, L ij And
respectively the ith training sample j The cross entropy loss error and the energy consumption predicted value at the moment are logarithms with e as a base;
and P4, the second module continuously substitutes the energy consumption data collected in real time into the well-trained LSTM model to predict the energy consumption value at any target moment.
10. The system according to any one of claims 1 to 8, wherein the third module comprises a production line monitoring module, an equipment management module, an energy consumption management module, an inventory management module, a quality management module, a production data statistics and report module, and the access control system comprises a specific operation process that the upper ERP transmits production order information to the third module for processing through an enterprise service bus ESB, then transmits the production information to each module terminal, and performs information interaction with the lower layer production equipment including tailing intelligent production equipment through an SCADA system to complete the intelligent management and control of the whole plant,
the staff enters a working area through an IC electronic card access control, checks a plan report, carries out a detection process flow set by a quality management module through a production line monitoring module, generates a report through a production data statistical report module and completes a detection report so as to determine whether early warning is needed or not, and records detection report data and early warning information into a database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111217407.6A CN113926571B (en) | 2021-10-19 | 2021-10-19 | Tailing activation process system based on drying, grinding and selecting integrated intelligent production equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111217407.6A CN113926571B (en) | 2021-10-19 | 2021-10-19 | Tailing activation process system based on drying, grinding and selecting integrated intelligent production equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113926571A CN113926571A (en) | 2022-01-14 |
| CN113926571B true CN113926571B (en) | 2022-09-27 |
Family
ID=79280615
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111217407.6A Active CN113926571B (en) | 2021-10-19 | 2021-10-19 | Tailing activation process system based on drying, grinding and selecting integrated intelligent production equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113926571B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115055265A (en) * | 2022-06-30 | 2022-09-16 | 中钢石家庄工程设计研究院有限公司 | Large-scale iron ore deposit underground mining selects to select fills integration system |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AUPO944297A0 (en) * | 1997-09-25 | 1997-10-16 | Advance R & D Pty Ltd | Modular and transportable processing plant and mobile mineral process evaluation unit |
| CN106007436B (en) * | 2015-01-23 | 2019-02-12 | 福建鼎盛元环保科技有限公司 | Production process of tailing micro powder |
| CN109388078A (en) * | 2017-08-02 | 2019-02-26 | 景蓓蓓 | A kind of Tailings Dam remote supervision system |
| CN107721217B (en) * | 2017-08-31 | 2020-05-19 | 重庆七星龙环保发展有限公司 | Activation kiln system and control method thereof |
| CA3008149C (en) * | 2018-06-27 | 2020-05-05 | Aly Gomaa | System and method for treating and pumping tailings |
| CN110002775B (en) * | 2019-03-20 | 2023-02-03 | 安徽海螺集团有限责任公司 | Integrated intelligent quality control system for cement production quality control |
| CN110657349B (en) * | 2019-10-08 | 2020-04-28 | 山东黄金矿业科技有限公司充填工程实验室分公司 | Mine tailing paste slurry ring pipe test system and system operation method |
-
2021
- 2021-10-19 CN CN202111217407.6A patent/CN113926571B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113926571A (en) | 2022-01-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111539583B (en) | Production process simulation optimization method based on digital twin | |
| Liang et al. | Cyber Physical System and Big Data enabled energy efficient machining optimisation | |
| Huber et al. | DMME: Data mining methodology for engineering applications–a holistic extension to the CRISP-DM model | |
| Wang et al. | Big Data enabled Intelligent Immune System for energy efficient manufacturing management | |
| CN108500736A (en) | Tool state estimating device and lathe | |
| DE102017108169B4 (en) | Production system that defines a determination value of a variable in relation to a product deviation | |
| CN113926571B (en) | Tailing activation process system based on drying, grinding and selecting integrated intelligent production equipment | |
| CN201616026U (en) | Monitoring and maintaining system for vibrating screen | |
| CN108121295A (en) | Prediction model establishing method, related prediction method and computer program product | |
| CN102867237A (en) | Intelligent production management method | |
| CN107704984A (en) | Job management system with sensor and rote learning portion | |
| CN107065795B (en) | A kind of the automatic charging method and its system of multi-varieties and small-batch pcb board | |
| Altaf et al. | Online simulation modeling of prefabricated wall panel production using RFID system | |
| CN104820901A (en) | Method for evaluating skill of clothing employees at production line based on production on-site data | |
| AU2021339237A1 (en) | System, equipment, and procedure for monitoring, predictive maintenance, and operational optimization of vibrating screeners | |
| CN114330026A (en) | Digital twin system simulation method and device | |
| CN107589731A (en) | A kind of intelligence manufacture flexible production line and its production method based on Internet of Things | |
| CN103034219A (en) | Meter assembly line capable of achieving best arrangement efficiency and control method thereof | |
| Kumbhar et al. | Bottleneck detection through data integration, process mining and factory physics-based analytics | |
| Nasseri et al. | Duality in fuzzy linear programming with symmetric trapezoidal numbers | |
| CN111736555A (en) | Global insect condition monitoring system and design method thereof | |
| CN118195284A (en) | Intelligent operation and maintenance visual monitoring method and system based on big data | |
| CN110266811B (en) | Workshop abnormal information pushing device and method based on MQTT technology | |
| CN117314202A (en) | Product carbon efficiency monitoring and optimizing method and system based on industrial Internet | |
| CN115577936A (en) | Clothing workshop data information statistical system and method based on industrial Internet of things |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: Room 303a, South venture building, Pioneer Park, torch hi tech Zone, Xiamen, Fujian, 361000 Applicant after: Xiamen Dutai New Material Technology Co.,Ltd. Address before: 361006 room 303a, south of entrepreneurship building, entrepreneurship Park, torch high tech Zone, Xiamen, Fujian Applicant before: Xiamen Ditai Environmental Protection Technology Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Tailings activation process system based on integrated intelligent production equipment of drying, grinding and selection Effective date of registration: 20231016 Granted publication date: 20220927 Pledgee: Bank of China Limited Xiamen Xiang'an sub branch Pledgor: Xiamen Dutai New Material Technology Co.,Ltd. Registration number: Y2023980061504 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |