CN112801398B - Cooling device failure prediction method and device, electronic equipment and storage medium - Google Patents
Cooling device failure prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN112801398B CN112801398B CN202110169049.XA CN202110169049A CN112801398B CN 112801398 B CN112801398 B CN 112801398B CN 202110169049 A CN202110169049 A CN 202110169049A CN 112801398 B CN112801398 B CN 112801398B
- Authority
- CN
- China
- Prior art keywords
- model
- prediction
- fault
- cooling device
- fault prediction
- 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
- 238000001816 cooling Methods 0.000 title claims abstract description 165
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000003860 storage Methods 0.000 title claims abstract description 15
- 230000004927 fusion Effects 0.000 claims abstract description 69
- 230000007246 mechanism Effects 0.000 claims abstract description 58
- 238000013178 mathematical model Methods 0.000 claims abstract description 15
- 238000001514 detection method Methods 0.000 claims description 46
- 238000012549 training Methods 0.000 claims description 35
- 238000013527 convolutional neural network Methods 0.000 claims description 8
- 238000007477 logistic regression Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 238000003066 decision tree Methods 0.000 claims description 7
- 238000003062 neural network model Methods 0.000 claims description 7
- 230000000306 recurrent effect Effects 0.000 claims description 6
- 230000008901 benefit Effects 0.000 abstract description 7
- 238000004458 analytical method Methods 0.000 abstract description 5
- 239000010720 hydraulic oil Substances 0.000 description 19
- 238000007500 overflow downdraw method Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 7
- 239000000110 cooling liquid Substances 0.000 description 5
- 239000000498 cooling water Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000036541 health Effects 0.000 description 4
- 239000003921 oil Substances 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000001050 lubricating effect Effects 0.000 description 3
- 238000005299 abrasion Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 239000002826 coolant Substances 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000005195 poor health Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000005086 pumping Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 238000005338 heat storage Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000011089 mechanical engineering Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Strategic Management (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Operations Research (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Molecular Biology (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention provides a cooling device fault prediction method, a cooling device fault prediction device, electronic equipment and a storage medium, wherein the method comprises the following steps: determining a state parameter of the cooling device; inputting the state parameters into a fault prediction fusion model to obtain a fault prediction result output by the fault prediction fusion model; the fault prediction fusion model is obtained by fusing a fault prediction mechanism model and a fault prediction data driving model, and the fault prediction mechanism model is determined based on a thermodynamic mathematical model of the cooling device. The method, the device, the electronic equipment and the storage medium provided by the invention have the advantages that the analysis capability of the state parameters of the cooling device is improved, and the accuracy and the reliability of the fault prediction of the cooling device are improved.
Description
Technical Field
The present invention relates to the field of mechanical engineering technologies, and in particular, to a cooling device failure prediction method, a cooling device failure prediction device, an electronic device, and a storage medium.
Background
In a working machine, a cooling device is used for radiating heat of a working medium in a power system, for example, in an excavator, the cooling device is used for cooling hydraulic oil in a hydraulic system, so that the hydraulic oil can maintain good lubricating performance.
In the prior art, a plurality of types of sensors are usually installed on a cooling device, fault prediction is performed by collecting state parameters of the cooling device, the types of the state parameters are more, and the accuracy of the fault prediction is poor.
Disclosure of Invention
The invention provides a cooling device fault prediction method, a cooling device fault prediction device, electronic equipment and a storage medium, which are used for improving the accuracy of cooling device fault prediction.
The invention provides a cooling device fault prediction method, which comprises the following steps:
determining a state parameter of the cooling device;
inputting the state parameters into a fault prediction fusion model to obtain a fault prediction result output by the fault prediction fusion model;
the fault prediction fusion model is obtained by fusing a fault prediction mechanism model and a fault prediction data driving model, and the fault prediction mechanism model is determined based on a thermodynamic mathematical model of the cooling device.
According to the cooling device fault prediction method provided by the invention, the fault prediction fusion model is determined based on the following steps:
determining the prediction weights of the fault prediction mechanism model and the fault prediction data driving model;
And carrying out weighted fusion on the fault prediction mechanism model and the fault prediction data driving model based on the prediction weight to obtain the fault prediction fusion model.
According to the cooling device fault prediction method provided by the invention, the state parameters are input into a fault prediction fusion model to obtain a fault prediction result output by the fault prediction fusion model, and then the method comprises the following steps:
And adjusting the prediction weights of the fault prediction mechanism model and the fault prediction data driving model based on the fault detection result of the cooling device.
According to the cooling device fault prediction method provided by the invention, the fault prediction data driving model is determined based on the following steps:
Determining a data training set; the data training set comprises sample state parameters of the cooling device and sample fault detection results corresponding to the sample state parameters;
And training an initial model based on the data training set to obtain the fault prediction data driving model.
According to the cooling device fault prediction method provided by the invention, the data training set comprises a positive sample set and a negative sample set, wherein the sample fault detection result in the positive sample set is normal, and the sample fault detection result in the negative sample set is fault.
According to the cooling device fault prediction method provided by the invention, the initial model comprises at least one of XGBoost decision tree models, convolutional neural network models, cyclic neural network models and logistic regression models.
According to the cooling device fault prediction method provided by the invention, the fault prediction fusion model corresponds to the type of the cooling device one by one.
The invention also provides a cooling device fault prediction device, comprising:
A determining unit for determining a status parameter of the cooling device;
The prediction unit is used for inputting the state parameters into a fault prediction fusion model to obtain a fault prediction result output by the fault prediction fusion model;
the fault prediction fusion model is obtained by fusing a fault prediction mechanism model and a fault prediction data driving model, and the fault prediction mechanism model is determined based on a thermodynamic mathematical model of the cooling device.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the cooling device fault prediction method as described in any one of the above when executing the program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of a cooling device failure prediction method as described in any of the above.
According to the cooling device fault prediction method, the cooling device fault prediction device, the electronic equipment and the storage medium, the fault prediction fusion model is obtained after the fault prediction mechanism model and the fault prediction data driving model of the cooling device are fused, the fault prediction result can be determined according to the state parameters of the cooling device by the obtained fault prediction fusion model, the advantages of the fault prediction mechanism model and the fault prediction data driving model can be fully utilized, the analysis capability of the state parameters of the cooling device is improved, and the accuracy and the reliability of the fault prediction of the cooling device are improved.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a cooling device failure prediction method provided by the invention;
FIG. 2 is a schematic flow chart of a method for predicting faults of an excavator cooling device;
FIG. 3 is a schematic diagram of a cooling device failure prediction apparatus according to the present invention;
Fig. 4 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Work machines are a variety of types of work machines that perform work such as cranes, excavators, dozers, and the like. The cooling device in the working machine cools the hydraulic oil in the hydraulic system, so that the hydraulic oil has good lubricating performance. The cooling device is also called a cooler, and the operating mechanism is to take away the heat generated by the hydraulic oil because of the operation through a heat exchange mode. The frequent faults of the cooling device include radiator breakage, coolant leakage, abnormal operation of a cooling water pump, unsmooth circulation of the coolant and the like. Common cooling devices can be divided into a tube type cooling device and a plate type cooling device, and also can be divided into an air-cooled cooling device and a water-cooled cooling device.
Fig. 1 is a schematic flow chart of a cooling device fault prediction method provided by the present invention, as shown in fig. 1, the method includes:
at step 110, a status parameter of the cooling device is determined.
In particular, the state parameter is used to characterize the operating state of the cooling device. The kind and number of status parameters of the cooling device are closely related to the type of construction of the cooling device. For example, a cooling device for cooling a hydraulic system is often of a water-cooled structure, and a cooling device for cooling a generator is often of an air-cooled structure.
Taking an excavator cooling device as an example, the cooling device adopts cooling liquid, and main parts of the cooling device comprise a cooling motor, a cooling water pump, a radiator, a cooling fan, a heat storage tank and various types of sensors. The sensor is used for collecting state parameters of the cooling device. The sensor comprises a temperature sensor, a pressure sensor, a liquid level sensor, a voltmeter, an ammeter, a tachometer and the like. The state parameters of the excavator cooling device may include the cooling motor operating voltage, the cooling motor operating current, the cooling motor power, the cooling water pump rotational speed, the cooling water pump outlet pressure, the cooling liquid temperature, etc.
The state parameter is closely related to the real-time operating state of the cooling device. When the cooling device is in a normal state, its state parameter fluctuates within a normal numerical range, and when the cooling device is in a failure state, its state parameter may exceed the normal numerical range.
Step 120, inputting the state parameters into the fault prediction fusion model to obtain a fault prediction result output by the fault prediction fusion model; the fault prediction fusion model is obtained by fusing a fault prediction mechanism model and a fault prediction data driving model, and the fault prediction mechanism model is determined based on a thermodynamic mathematical model of the cooling device.
In particular, the fault prediction result may be used to measure the equipment health status of the cooling device, and may also be used to indicate the type of fault that may occur in the cooling device. The types of the fault prediction results are in one-to-one correspondence with the types of the uses of the fault prediction fusion model. When the fault prediction fusion model is used for predicting the health state of the equipment, the fault prediction result can be the equipment health degree of the cooling device; when the failure prediction fusion model is used to predict the failure type, the failure prediction result may be the failure type of the cooling device.
The mechanism model, also known as the white box model. An accurate mathematical model is built based on the internal mechanisms of the object, the production process, or the delivery mechanism of the material flow. It is a mathematical model of an object or process based on mass balance equations, energy balance equations, momentum balance equations, phase balance equations, and certain physical equations, chemical reaction laws, basic laws of circuits, etc.
The failure prediction mechanism model is determined based on a thermodynamic mathematical model of the cooling device. A thermodynamic mathematical model of the cooling device can be obtained from published papers.
For example, a failure prediction mechanism model of the cooling device in the hydraulic system of the excavator can be established according to the combination of a thermodynamic mathematical model of the cooling device in the hydraulic system of the excavator and the viscosity-Wen Gongshi of Ubbelohde-Walter. The failure prediction mechanism model takes the viscosity of hydraulic oil as an output and takes the pressure and the temperature of cooling liquid in a cooling device as an input. When the viscosity of the hydraulic oil is lower than a preset threshold value, the cooling device is considered to be in fault, and when the viscosity of the hydraulic oil is higher than the preset threshold value, the cooling device is considered to work normally. The adhesive-Wen Gongshi is as follows:
loglog(v+0.7)＝A-B*log(t+273.15)
wherein v is the viscosity of hydraulic oil, t is the temperature of the hydraulic oil, A and B are viscosity coefficients, v is continuously reduced along with the rise of the temperature t of the hydraulic oil, the lubricity of the hydraulic oil is deteriorated, and the abrasion of a hydraulic system is aggravated. Wherein, can be according to the thermodynamic mathematical model of cooling device to obtain:
t＝f(P,T)
wherein T is the temperature of hydraulic oil, P is the pressure of cooling liquid in the cooling device, T is the temperature of the cooling liquid in the cooling device, and f is the thermal transfer function, and can be obtained by reference in the open literature.
The failure prediction data driving model is a model which is obtained by collecting a large number of state parameters and training by adopting a machine learning method and used for carrying out failure prediction on the cooling device.
And after the fault prediction mechanism model and the fault prediction data driving model are fused, a fault prediction fusion model can be obtained.
Here, the failure prediction data driving model may also be obtained by a model fusion method. For example, the classification may be performed according to the fault detection result of the cooling device, the sample state parameters of the same fault detection result are classified into the same class, and the initial model is trained to obtain the fault prediction sub-model corresponding to the fault detection result. According to the method, the fault predictor models corresponding to the fault detection results can be obtained through training. And then fusing the plurality of fault prediction sub-models to obtain a fault prediction data driving model.
The failure prediction mechanism model utilizes mechanism knowledge, and the relation between state parameters and failures is dependent, but the effect on multi-cause concurrent failures is limited; the failure prediction data driven model can utilize the internal relation between data and realize higher precision by utilizing large data volume, but is difficult to reach higher precision when the failure condition is complex and the sample is insufficient.
According to the cooling device fault prediction method provided by the embodiment of the invention, the fault prediction mechanism model and the fault prediction data driving model of the cooling device are fused to obtain the fault prediction fusion model, the obtained fault prediction fusion model can determine the fault prediction result according to the state parameters of the cooling device, the advantages of the fault prediction mechanism model and the fault prediction data driving model can be fully utilized, the analysis capability of the state parameters of the cooling device is improved, and the accuracy and the reliability of the fault prediction of the cooling device are improved.
Based on any of the above embodiments, the fault prediction fusion model is determined based on the steps of:
determining the prediction weights of a fault prediction mechanism model and a fault prediction data driving model;
And carrying out weighted fusion on the fault prediction mechanism model and the fault prediction data driving model based on the prediction weight to obtain a fault prediction fusion model.
Specifically, model fusion is to train a plurality of models, and the plurality of models are fused into one model according to a certain method. The model fusion method comprises a linear weighting fusion method, a cross fusion method, a waterfall fusion method, a characteristic fusion method, a prediction fusion method and the like.
A linear weighted fusion method can be adopted for the failure prediction mechanism model and the failure prediction data driving model. The predictive weights of the failure prediction mechanism model and the failure prediction data driven model may be determined first. The prediction weight is used for measuring the accuracy of the model prediction result. The more accurate the model prediction result is, the larger the prediction weight is; the worse the model prediction results, the smaller the prediction weights.
And carrying out weighted fusion on the fault prediction mechanism model and the fault prediction data driving model according to the prediction weight corresponding to each model to obtain a fault prediction fusion model.
Based on any of the above embodiments, step 120 then comprises:
and adjusting the prediction weights of the failure prediction mechanism model and the failure prediction data driving model based on the failure detection result of the cooling device.
Specifically, the fault detection result may be a real result obtained after the cooling device is detected. The prediction weights of the failure prediction mechanism model and the failure prediction data driving model can be adjusted according to the failure detection result of the cooling device.
For example, the initial value of the prediction weight of the failure prediction mechanism model and the failure prediction data driving model may be set to 0.5. When the cooling device fails, comparing the fault detection result obtained by manual detection with the fault prediction result output by the fault prediction mechanism model and the fault prediction data driving model respectively, if the fault detection result accords with any fault prediction result, increasing the prediction weight of the model corresponding to the fault prediction result, and if the fault detection result does not accord with any fault prediction result, reducing the prediction weight of the model corresponding to the fault prediction result.
According to the cooling device fault prediction method provided by the embodiment of the invention, the prediction weights of the fault prediction mechanism model and the fault prediction data driving model are adjusted, so that the advantage complementation of the two models is realized, and the precision of the fault prediction of the working machine is improved.
Based on any of the above embodiments, the failure prediction data driving model is determined based on the steps of:
Determining a data training set; the data training set comprises sample state parameters of the cooling device and sample fault detection results corresponding to the sample state parameters;
And training the initial model based on the data training set to obtain a fault prediction data driving model.
Specifically, the fault prediction data driving model can be obtained through pre-training, and specifically, the fault prediction data driving model can be obtained through the following training mode:
Firstly, a large number of sample state parameters of the cooling device are collected, and secondly, sample fault detection results corresponding to the sample state parameters are determined. The fault detection result may be the actual fault type or the equipment health degree obtained by detecting the cooling device. And then training the initial model according to each sample state parameter and a sample fault detection result corresponding to each sample state parameter so as to improve the fault prediction capacity of the initial model for the cooling device and obtain a fault prediction data driving model.
The sample state parameters of the cooling device and the sample fault detection results corresponding to the sample state parameters form a data training set.
Based on any of the above embodiments, the data training set includes a positive sample set and a negative sample set, the sample fault detection result in the positive sample set is normal, and the sample fault detection result in the negative sample set is a fault.
Specifically, positive and negative samples can be adopted to train the fault prediction data driving model, so that the fault prediction data driving model can learn the characteristic information between the normal operation state parameters of the cooling device and the characteristic information between the normal operation state parameters of the cooling device.
When constructing the data training set, the positive sample set and the negative sample set can be constructed simultaneously. The sample fault detection result in the positive sample set is normal, and the sample fault detection result in the negative sample set is fault.
For example, first, a point in time when the cooling device fails may be obtained from historical service information of the work machine.
Secondly, the returned data of one week before the time point when the cooling device breaks down is obtained to be used as a negative sample set, the original data of the rotating speed, the pumping pressure, the power, the oil temperature, the water temperature, the current, the voltage, the oil level, the working time length and the like in the returned data are extracted, the characteristic parameters of peak value, valley value, average value, mode, numerical distribution statistics and the like are extracted to be used as state parameters, and a sample fault detection result corresponding to the state parameters is set as a fault.
And thirdly, acquiring returned data of one week after the time point of the failure of the cooling device as a positive sample set, extracting original data such as rotating speed, pumping pressure, power, oil temperature, water temperature, current, voltage, oil level, working time and the like in the returned data, extracting characteristic parameters such as peak value, valley value, average value, mode, numerical distribution statistics and the like as state parameters, and setting a sample failure detection result corresponding to the state parameters to be normal.
Based on any of the above embodiments, the initial model includes at least one of a XGBoost decision tree model, a convolutional neural network model, a recurrent neural network model, and a logistic regression model.
Specifically, the fault prediction data driven model may select XGBoost a decision tree model, a convolutional neural network model, a cyclic neural network model, and a linear function model.
The XGboost (Extreme Gradient Boosting) algorithm is a model of a decision tree, commonly used for regression and classification, and is an extension of the gradient lifting machine algorithm. The principle is that a large number of CART trees (regression trees) with lower accuracy are combined to form a model with higher accuracy. The model generates a new tree to reduce errors each training iteration. The XGboost algorithm generates a new tree toward the goal of minimization using a gradient descent method based on the previous tree at each iteration. A large number of trees are generated in one iteration to achieve the intended expectations. The XGboost algorithm has the characteristics of high accuracy, difficult fitting and the like.
The convolutional neural network (Convolutional Neural Networks, CNN) model is a type of feed-forward neural network (Feedforward Neural Networks) that contains convolutional computations and has a depth structure that includes an input layer, a hidden layer, and an output layer. The input layer can process multidimensional data, the hidden layer can conduct feature extraction on the multidimensional data, and the multidimensional data is output after being classified by the output layer.
The recurrent neural network model (Recurrent Neural Network, RNN) is a type of recurrent neural network model that takes sequence data as input, recursions in the evolution direction of the sequence, and all nodes are chained.
The logistic regression model (Logistic Regression, LR) is a generalized linear regression analysis model, which uses a logistic function on the basis of linear regression. The logistic regression model has the advantage of higher training speed, and the calculated amount is only related to the number of the features when the classification is carried out. Furthermore, the logistic regression model is very interpretable, and the impact of different features on the final result can be seen from the weights of the features.
Based on any of the above embodiments, the failure prediction fusion model corresponds to the type of the cooling device one by one.
Specifically, the structure and performance of each type of cooling device are different, resulting in a completely different failure prediction method for each type of cooling device.
Therefore, when the fault prediction fusion model is established, the fault prediction fusion model can be in one-to-one correspondence with the types of the cooling devices, so that the fault prediction fusion model can be adapted to the corresponding cooling devices, the cooling devices can be more accurately predicted, and the obtained fault prediction result is more accurate.
Based on any of the above embodiments, fig. 2 is a schematic flow chart of a fault prediction method for an excavator cooling device according to the present invention, as shown in fig. 2, where the method includes:
step one, constructing a mechanism model
The high hydraulic oil temperature in the excavator can lead to the poor lubricating property of the hydraulic oil, further increase the internal abrasion and the friction between the connecting pieces, and can lead to the deterioration of the hydraulic oil and the occurrence of cracks on the connecting pieces, further lead to the phenomenon of insufficient output and the like of the excavator. The cooling device is used as a heat radiating device of the excavator, the operation state of the cooling device is clearly related to the temperature of hydraulic oil, and when the cooling device has faults, the phenomenon that the temperature of the hydraulic oil is high is often caused. According to the hydraulic oil temperature, basic parameters of the cooling device of the excavator and a thermodynamic transfer principle, a mechanism model corresponding to the cooling device of the excavator can be constructed, and the mechanism model can be used for carrying out fault prediction on the cooling device of the excavator.
Step two, constructing a data driving model
And collecting a large number of sample state parameters of the cooling device, and determining sample fault detection results corresponding to the sample state parameters so as to construct a data training set, wherein the data training set comprises a positive sample set and a negative sample set. The sample fault detection result in the positive sample set is normal, and the sample fault detection result in the negative sample set is fault. And training the initial model according to a data training set by taking XGBoost decision trees as the initial model, and continuously adjusting parameters such as tree depth, number and the like to obtain a data driving model capable of carrying out fault prediction on the cooling device of the excavator.
Step three, model fusion
And the output results of the mechanism model and the data driving model are the equipment health state scores of the cooling device of the excavator. And setting a prediction weight, and fusing the mechanism model and the data driving model to obtain a fused prediction model. And after the output of the fusion prediction model reaches the early warning condition, detecting the cooling device of the excavator on site, and adjusting the prediction weight according to the actual result obtained by detection. When the cooling device of the excavator has faults, the prediction weight of a model with a lower score (poor health state) can be correspondingly increased; when no fault exists in the cooling device of the excavator, the prediction weight of the model with low score (poor health state) can be correspondingly pulled down.
Based on any of the above embodiments, fig. 3 is a schematic structural diagram of a cooling device failure prediction device provided by the present invention, and as shown in fig. 3, the device includes:
a determining unit 310 for determining a status parameter of the cooling device;
The prediction unit 320 is configured to input the state parameter to the failure prediction fusion model, and obtain a failure prediction result output by the failure prediction fusion model;
The fault prediction fusion model is obtained by fusing a fault prediction mechanism model and a fault prediction data driving model, and the fault prediction mechanism model is determined based on a thermodynamic mathematical model of the cooling device.
Specifically, the determining unit 310 is configured to determine a state parameter of the cooling device. The prediction unit 320 is configured to input the state parameter to the failure prediction fusion model, and obtain a failure prediction result output by the failure prediction fusion model.
According to the cooling device fault prediction device provided by the embodiment of the invention, the fault prediction mechanism model and the fault prediction data driving model of the cooling device are fused to obtain the fault prediction fusion model, the obtained fault prediction fusion model can determine the fault prediction result according to the state parameters of the cooling device, the advantages of the fault prediction mechanism model and the fault prediction data driving model can be fully utilized, the analysis capability of the state parameters of the cooling device is improved, and the accuracy and the reliability of the fault prediction of the cooling device are improved.
Based on any of the above embodiments, the fault prediction fusion model is determined based on the steps of:
determining the prediction weights of a fault prediction mechanism model and a fault prediction data driving model;
And carrying out weighted fusion on the fault prediction mechanism model and the fault prediction data driving model based on the prediction weight to obtain a fault prediction fusion model.
Based on any of the above embodiments, the apparatus further comprises:
And the adjusting unit is used for adjusting the prediction weights of the fault prediction mechanism model and the fault prediction data driving model based on the fault detection result of the cooling device.
Based on any of the above embodiments, the failure prediction data driving model is determined based on the steps of:
Determining a data training set; the data training set comprises sample state parameters of the cooling device and sample fault detection results corresponding to the sample state parameters;
And training the initial model based on the data training set to obtain a fault prediction data driving model.
Based on any of the above embodiments, the data training set includes a positive sample set and a negative sample set, the sample fault detection result in the positive sample set is normal, and the sample fault detection result in the negative sample set is a fault.
Based on any of the above embodiments, the initial model includes at least one of a XGBoost decision tree model, a convolutional neural network model, a recurrent neural network model, and a logistic regression model.
Based on any of the above embodiments, the failure prediction fusion model corresponds to the type of the cooling device one by one.
Based on any of the above embodiments, fig. 4 is a schematic structural diagram of an electronic device provided by the present invention, and as shown in fig. 4, the electronic device may include: processor (Processor) 410, communication interface (Communications Interface) 420, memory (Memory) 430, and communication bus (Communications Bus) 440, wherein Processor 410, communication interface 420, memory 430 complete communication with each other through communication bus 440. The processor 410 may invoke logic commands in the memory 430 to perform the following method:
Determining a state parameter of the cooling device; inputting the state parameters into a fault prediction fusion model to obtain a fault prediction result output by the fault prediction fusion model; the fault prediction fusion model is obtained by fusing a fault prediction mechanism model and a fault prediction data driving model, and the fault prediction mechanism model is determined based on a thermodynamic mathematical model of the cooling device.
In addition, the logic commands in the memory 430 described above may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in the form of a software product stored in a storage medium, comprising several commands for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The processor in the electronic device provided by the embodiment of the invention can call the logic instruction in the memory to realize the method, and the specific implementation mode is consistent with the implementation mode of the method, and the same beneficial effects can be achieved, and the detailed description is omitted here.
Embodiments of the present invention also provide a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the methods provided by the above embodiments, for example, comprising:
Determining a state parameter of the cooling device; inputting the state parameters into a fault prediction fusion model to obtain a fault prediction result output by the fault prediction fusion model; the fault prediction fusion model is obtained by fusing a fault prediction mechanism model and a fault prediction data driving model, and the fault prediction mechanism model is determined based on a thermodynamic mathematical model of the cooling device.
When the computer program stored on the non-transitory computer readable storage medium provided by the embodiment of the present invention is executed, the above method is implemented, and the specific implementation manner of the method is consistent with the implementation manner of the foregoing method, and the same beneficial effects can be achieved, which is not repeated herein.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several commands for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A cooling device failure prediction method, comprising:
determining a state parameter of the cooling device;
Inputting the state parameters into a fault prediction fusion model to obtain a fault prediction result output by the fault prediction fusion model; the types of the fault prediction results are in one-to-one correspondence with the types of the uses of the fault prediction fusion model;
The fault prediction fusion model is obtained by fusing a fault prediction mechanism model and a fault prediction data driving model, and the fault prediction mechanism model is determined based on a thermodynamic mathematical model of the cooling device;
The failure prediction data driven model is determined based on the steps of:
Classifying based on the fault detection result of the cooling device, dividing sample state parameters of the same fault detection result into the same class, and training an initial model to obtain a fault prediction sub-model corresponding to the fault detection result;
And fusing the fault prediction sub-models corresponding to the plurality of fault detection results to obtain the fault prediction data driving model.
2. The cooling device failure prediction method according to claim 1, wherein the failure prediction fusion model is determined based on the steps of:
determining the prediction weights of the fault prediction mechanism model and the fault prediction data driving model;
And carrying out weighted fusion on the fault prediction mechanism model and the fault prediction data driving model based on the prediction weight to obtain the fault prediction fusion model.
3. The cooling device failure prediction method according to claim 2, wherein the step of inputting the state parameter into a failure prediction fusion model to obtain a failure prediction result output by the failure prediction fusion model, and then comprises the steps of:
And adjusting the prediction weights of the fault prediction mechanism model and the fault prediction data driving model based on the fault detection result of the cooling device.
4. The cooling device failure prediction method according to claim 1, wherein the failure prediction data driving model is determined based on the steps of:
Determining a data training set; the data training set comprises sample state parameters of the cooling device and sample fault detection results corresponding to the sample state parameters;
And training an initial model based on the data training set to obtain the fault prediction data driving model.
5. The cooling device failure prediction method according to claim 4, wherein the data training set includes a positive sample set and a negative sample set, the positive sample set sample failure detection result is normal, and the negative sample set sample failure detection result is failure.
6. The cooling device failure prediction method of claim 4, wherein the initial model comprises at least one of XGBoost decision tree model, convolutional neural network model, recurrent neural network model, and logistic regression model.
7. The cooling device failure prediction method according to any one of claims 1 to 6, characterized in that the failure prediction fusion model corresponds one-to-one to the type of the cooling device.
8. A cooling device failure prediction apparatus, comprising:
A determining unit for determining a status parameter of the cooling device;
The prediction unit is used for inputting the state parameters into a fault prediction fusion model to obtain a fault prediction result output by the fault prediction fusion model; the types of the fault prediction results are in one-to-one correspondence with the types of the uses of the fault prediction fusion model;
The fault prediction fusion model is obtained by fusing a fault prediction mechanism model and a fault prediction data driving model, and the fault prediction mechanism model is determined based on a thermodynamic mathematical model of the cooling device;
The failure prediction data driven model is determined based on the steps of:
Classifying based on the fault detection result of the cooling device, dividing sample state parameters of the same fault detection result into the same class, and training an initial model to obtain a fault prediction sub-model corresponding to the fault detection result;
And fusing the fault prediction sub-models corresponding to the plurality of fault detection results to obtain the fault prediction data driving model.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the cooling device failure prediction method of any one of claims 1 to 7 when the program is executed.
10. A non-transitory computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the steps of the cooling device failure prediction method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110169049.XA CN112801398B (en) | 2021-02-07 | 2021-02-07 | Cooling device failure prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110169049.XA CN112801398B (en) | 2021-02-07 | 2021-02-07 | Cooling device failure prediction method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112801398A CN112801398A (en) | 2021-05-14 |
| CN112801398B true CN112801398B (en) | 2024-04-30 |
Family
ID=75814781
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110169049.XA Active CN112801398B (en) | 2021-02-07 | 2021-02-07 | Cooling device failure prediction method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112801398B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719002A (en) * | 2016-01-18 | 2016-06-29 | 重庆大学 | Wind turbine generator state parameter abnormity identification method based on combination prediction |
| CN107589695A (en) * | 2017-09-12 | 2018-01-16 | 中国中车股份有限公司 | A kind of train groups prognostic and health management system |
| CN109709926A (en) * | 2019-01-25 | 2019-05-03 | 东华大学 | The method of adjustment of melt quality in fiber production process |
| CN110688118A (en) * | 2019-10-28 | 2020-01-14 | 郑州阿帕斯科技有限公司 | Webpage optimization method and device |
| CN110957758A (en) * | 2019-11-20 | 2020-04-03 | 国网河北省电力有限公司电力科学研究院 | Comprehensive energy efficiency evaluation and control performance optimization method based on wide-area situation awareness |
| WO2020153934A1 (en) * | 2019-01-21 | 2020-07-30 | Hewlett-Packard Development Company, L.P. | Fault prediction model training with audio data |
| CN111554353A (en) * | 2020-04-10 | 2020-08-18 | 中国恩菲工程技术有限公司 | On-line prediction method for parameters of copper smelting process of oxygen bottom-blowing furnace |
-
2021
- 2021-02-07 CN CN202110169049.XA patent/CN112801398B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719002A (en) * | 2016-01-18 | 2016-06-29 | 重庆大学 | Wind turbine generator state parameter abnormity identification method based on combination prediction |
| CN107589695A (en) * | 2017-09-12 | 2018-01-16 | 中国中车股份有限公司 | A kind of train groups prognostic and health management system |
| WO2020153934A1 (en) * | 2019-01-21 | 2020-07-30 | Hewlett-Packard Development Company, L.P. | Fault prediction model training with audio data |
| CN109709926A (en) * | 2019-01-25 | 2019-05-03 | 东华大学 | The method of adjustment of melt quality in fiber production process |
| CN110688118A (en) * | 2019-10-28 | 2020-01-14 | 郑州阿帕斯科技有限公司 | Webpage optimization method and device |
| CN110957758A (en) * | 2019-11-20 | 2020-04-03 | 国网河北省电力有限公司电力科学研究院 | Comprehensive energy efficiency evaluation and control performance optimization method based on wide-area situation awareness |
| CN111554353A (en) * | 2020-04-10 | 2020-08-18 | 中国恩菲工程技术有限公司 | On-line prediction method for parameters of copper smelting process of oxygen bottom-blowing furnace |
Non-Patent Citations (2)
| Title |
|---|
| 基于组合模型的汽车轮毂轴承故障预测研究;张凯;中国优秀硕士学位论文全文数据库(第1期);原文第15-35页 * |
| 大型电站锅炉燃烧可视化监测与闭环控制系统研究;姜志峰;能源工程(第5期);原文第23-25页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112801398A (en) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110007652B (en) | Hydroelectric generating set degradation trend interval prediction method and system | |
| CN112785091B (en) | Method for carrying out fault prediction and health management on oil field electric submersible pump | |
| Liu et al. | Research on fault diagnosis of wind turbine based on SCADA data | |
| CN107944090B (en) | Gas turbine system performance prediction method based on key component failure model | |
| KR102103006B1 (en) | Method and Apparatus for Operating Optimal of Equipment based on Machine Learning Model | |
| CN111784061B (en) | Training method, device and equipment for power grid engineering cost prediction model | |
| CN113742993A (en) | Method, device, equipment and storage medium for predicting life loss of dry-type transformer | |
| CN117490338A (en) | Ship cooling unit control method and system based on deep learning | |
| CN112801398B (en) | Cooling device failure prediction method and device, electronic equipment and storage medium | |
| CN115204658A (en) | Method and system for assessing health state of oil supply equipment of hydroelectric generating set | |
| CN114548539A (en) | Method, device, equipment and medium for predicting turbine energy consumption of compressor of circulating water system | |
| CN113348473B (en) | Management artificial intelligence module installation foundation | |
| CN113529845A (en) | Excavator fault diagnosis method, excavator, electronic device and readable storage medium | |
| CN107956731A (en) | The Forecasting Methodology that the change of Air Cooling Fan for Power Station frequency influences unit back pressuce parameter | |
| CN110967188B (en) | Rolling bearing residual life prediction method and system based on iterative correlation vector machine | |
| Xie et al. | The application of echo state network and recurrent multilayer perceptron in rotating machinery fault prognosis | |
| CN106845144A (en) | A kind of trend prediction method excavated based on industrial big data | |
| CN114943281B (en) | Intelligent decision-making method and system for heat pipe cooling reactor | |
| CN115639799B (en) | Tire vulcanizer hydraulic system fault diagnosis method based on improved BP-HMM hybrid model | |
| CN114997333B (en) | Fault diagnosis method and device for wind driven generator | |
| CN117218602A (en) | Structural health monitoring data abnormality diagnosis method and system | |
| CN116596120A (en) | Variable working condition degradation trend prediction method and device for pumped storage unit | |
| Li et al. | Remaining life prediction of cores based on data-driven and physical modeling methods | |
| CN115935814A (en) | Transformer fault prediction method based on ARIMA-SVM model | |
| CN115795397A (en) | Gearbox temperature prediction method based on 1DCNN-LSTM and BiLSTM parallel network |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230103 Address after: 314506 room 116, building 4, No. 288, development avenue, Tongxiang Economic Development Zone, Tongxiang City, Jiaxing City, Zhejiang Province Applicant after: Shengjing Intelligent Technology (Jiaxing) Co.,Ltd. Address before: 102206 5th floor, building 6, 8 Beiqing Road, Changping District, Beijing Applicant before: SANY HEAVY INDUSTRY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |