CN108920863A - A kind of method for building up of robot servo system energy consumption estimation model - Google Patents
A kind of method for building up of robot servo system energy consumption estimation model Download PDFInfo
- Publication number
- CN108920863A CN108920863A CN201810802066.0A CN201810802066A CN108920863A CN 108920863 A CN108920863 A CN 108920863A CN 201810802066 A CN201810802066 A CN 201810802066A CN 108920863 A CN108920863 A CN 108920863A
- Authority
- CN
- China
- Prior art keywords
- energy consumption
- model
- sample data
- consumption estimation
- training sample
- 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.)
- Granted
Links
- 238000005265 energy consumption Methods 0.000 title claims abstract description 172
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012549 training Methods 0.000 claims abstract description 85
- 238000012360 testing method Methods 0.000 claims description 89
- 230000002159 abnormal effect Effects 0.000 claims description 13
- 238000010606 normalization Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 4
- 238000004804 winding Methods 0.000 claims description 4
- 238000005457 optimization Methods 0.000 abstract description 2
- 238000005259 measurement Methods 0.000 abstract 2
- 238000004364 calculation method Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 8
- 238000011160 research Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Feedback Control In General (AREA)
Abstract
The present invention provides a kind of method for building up of robot servo system energy consumption estimation model, including：Robot servo System History operating parameter is obtained, the historical operating parameter is pre-processed to obtain model training sample data and model measurement sample data；Energy consumption estimation model is established based on support vector regression algorithm and using the model training sample data；The model measurement sample data is inputted in the energy consumption estimation model and obtains estimation power consumption values, and obtains the error rate of the estimation power consumption values and actual consumption value；Energy consumption estimation model is modified according to the error rate, estimates model to obtain the energy consumption of optimization.Technical solution provided by the invention solve the problems, such as it is existing to robot servo system loss analyze it is relatively complicated complexity.
Description
Technical Field
The invention relates to the technical field of communication, in particular to a method for establishing an energy consumption estimation model of a robot servo system.
Background
As the multi-joint robot is increasingly widely used in industry, reducing the energy consumption of the multi-joint robot becomes one of the main research directions. The energy consumption problem of the multi-joint robot is mainly researched from two aspects, namely the research on an energy consumption model of a robot system and the research on a control strategy of the robot.
At present, the energy consumption calculation of a multi-joint robot servo system is also in a stage of establishing an energy consumption model according to a loss mechanism through analyzing the loss of the servo system; however, because the energy consumption influence factors of the multi-joint robot servo system are numerous, an accurate energy consumption calculation model is difficult to establish; the existing methods for solving the problem of energy consumption calculation mostly build energy consumption models through statistical modeling methods, such as decision trees, neural networks and other methods, and such modeling methods are mostly used for analyzing the electricity consumption conditions of numerical control machine systems, ship systems and buildings. Based on the above problems, the analysis of the servo system loss is complicated, which results in more time and energy consumption for people.
Disclosure of Invention
The embodiment of the invention provides a method for establishing an energy consumption estimation model of a robot servo system, which aims to solve the problem that the existing robot servo system is complicated in loss analysis.
In order to achieve the above object, the present invention provides a method for establishing an energy consumption estimation model of a robot servo system, comprising:
acquiring historical operating parameters of a robot servo system, and preprocessing the historical operating parameters to acquire model training sample data and model test sample data;
establishing an energy consumption estimation model based on a support vector regression algorithm and by adopting the model training sample data;
inputting the model test sample data into the energy consumption estimation model to obtain an estimated energy consumption value, and obtaining an error rate of the estimated energy consumption value and an actual energy consumption value;
and correcting the energy consumption estimation model according to the error rate to obtain an optimized energy consumption estimation model.
Optionally, the step of obtaining historical operating parameters of the robot servo system and preprocessing the historical operating parameters to obtain model training sample data and model test sample data includes:
acquiring historical operating parameters of a robot servo system, and carrying out normalization processing on the historical operating parameters; wherein the historical operating parameters include historical output parameters and historical input parameters, the historical input parameters including one or more of the following: the servo motor comprises a servo motor stator phase current, a phase voltage, a winding phase resistance, a servo motor rotor rotating speed and friction torque of each rotating shaft of a servo transmission joint;
and taking the historical operating parameters of the preset ratio in the historical operating parameters after the normalization processing as model training sample data, and taking the remaining historical operating parameters as model test sample data.
Optionally, the step of establishing an energy consumption estimation model based on a support vector regression algorithm and using the model training sample data includes:
and establishing an energy consumption estimation function by adopting the model training sample data based on a radial Gaussian kernel function to obtain the energy consumption estimation model.
Optionally, the step of inputting the model test sample data into the energy consumption estimation model to obtain an estimated energy consumption value, and obtaining an error rate between the estimated energy consumption value and an actual energy consumption value includes:
equally dividing the model test sample data into preset groups of test set data, and respectively inputting the preset groups of test set data into the energy consumption estimation model to obtain preset groups of estimated energy consumption values;
and calculating the error rate between the estimated energy consumption value estimated by each group of the test set data and the actual energy consumption value of the corresponding model test sample data.
Optionally, the step of modifying the energy consumption estimation model according to the error rate to obtain an optimized energy consumption estimation model further includes:
determining whether an error rate between an estimated energy consumption value estimated by each group of the test set data and an actual energy consumption value of corresponding model test sample data is less than a preset threshold value;
when the error rate corresponding to each group of the test set data is smaller than the preset value, determining that the step is based on a support vector regression algorithm and adopting the model training sample data to establish an energy consumption estimation model obtained in the energy consumption estimation model as a final energy consumption estimation model;
when the error rate corresponding to one or more groups of the test set data is larger than the preset value, updating the one or more groups of the test set data into the model training sample data;
and returning to the step of establishing the energy consumption estimation model based on the support vector regression algorithm and by adopting the model training sample data so as to obtain the optimized energy consumption estimation model.
Optionally, after the step of modifying the energy consumption estimation model according to the error rate to obtain an optimized energy consumption estimation model, the method further includes:
and periodically acquiring current operating parameters of the robot servo system, and updating the current operating parameters into model training sample data and model test sample data so as to optimize the energy consumption estimation model.
Optionally, the step of updating the current operating parameter into the model training sample data and the model test sample data includes:
clustering the model training sample data in the historical operating data sample by adopting a k-means algorithm;
taking data, which is in the model training sample data and has a distance with a sample central point larger than a preset distance, as abnormal data, and removing the abnormal data to update the model training sample data;
and respectively adding the current operating parameters into the model training sample data and the model test sample data according to a preset proportion to form new model training sample data and new model test sample data.
According to the technical scheme provided by the invention, historical operating parameters of a robot servo system are obtained, and the historical operating parameters are preprocessed to obtain model training sample data and model test sample data; establishing an energy consumption estimation model based on a support vector regression algorithm and by adopting the model training sample data; inputting the model test sample data into the energy consumption estimation model to obtain an estimated energy consumption value, and obtaining an error rate of the estimated energy consumption value and an actual energy consumption value; and correcting the energy consumption estimation model according to the error rate to obtain an optimized energy consumption estimation model. Therefore, complex energy consumption analysis calculation of the robot servo system can be avoided, a robot servo system energy consumption estimation model is established based on historical operating parameters through a support vector regression algorithm, people do not need to consume a large amount of time and energy to analyze the energy consumption, follow-up energy consumption estimation of the robot servo system is facilitated, and important guiding significance is provided for design and use of the robot.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive exercise.
FIG. 1 is a schematic flow chart diagram illustrating an embodiment of a method for establishing an energy consumption estimation model of a robot servo system according to the present invention;
fig. 2 is a schematic flow chart of a method for establishing an energy consumption estimation model of a robot servo system according to another embodiment of the invention.
Fig. 3 is a schematic flow chart of a method for establishing an energy consumption estimation model of a robot servo system according to another embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, fig. 1 is a flowchart illustrating an embodiment of a method for establishing an energy consumption estimation model of a robot servo system according to the present invention. As shown in fig. 1, the method for establishing the energy consumption estimation model of the robot servo system includes the following steps:
step 101, obtaining historical operating parameters of a robot servo system, and preprocessing the historical operating parameters to obtain model training sample data and model test sample data.
In the embodiment of the invention, the robot servo system is a multi-joint robot servo system. The historical operating parameters of the robot servo system can comprise: stator phase current I and phase voltage U of servo motor, winding phase resistance R, rotor speed omega of servo motor, and friction torque tau of each rotating shaft of servo transmission jointfi(t), etc. It should be noted that the historical operating parameters are obtained by measuring the servo system of the articulated robot by the corresponding standard measuring instrument under the same load condition.
The preprocessing may be to perform normalization processing and screening on the historical operating parameters to obtain model training sample data and model test sample data. Specifically, the step 101 includes:
acquiring historical operating parameters of a robot servo system, and carrying out normalization processing on the historical operating parameters;
and taking the historical operating parameters of the preset ratio in the historical operating parameters after the normalization processing as model training sample data, and taking the remaining historical operating parameters as model test sample data.
For example, in a specific example, n sets of historical operating parameters { (x) are extracted from a multi-joint robot servo system parameter historical operating database1，y1),…,(xn，yn) In which xiFor inputting parameter data, yiThe output parameter is the energy consumption of the servo system of the multi-joint robot. And adopting a linear function to carry out normalization processing on the extracted historical operating parameter data, wherein the formula is as follows:
wherein x ismaxFor maximum value of input parameter data, xminAs a minimum value of the input parameter data, xiFor inputting parameter data, XiIs normalized data.
And (3) forming a sample set by the historical operating parameters after the normalization processing, and dividing the sample set into two types: model training sample data and model test sample data. In an embodiment of the present invention, the preset ratio may be 70%. That is, 70% of the normalized historical operating parameters are used as model training sample data, and the remaining 30% are used as model test sample data. Of course, the preset ratio can be set according to the needs, and is not limited in particular here.
And 102, establishing an energy consumption estimation model by adopting the model training sample data based on a support vector regression algorithm.
In the embodiment of the invention, a radial basis Gaussian kernel function is adopted as a kernel function of a support vector regression algorithm. Specifically, the step 102 includes:
and establishing an energy consumption estimation function by adopting the model training sample data based on a radial Gaussian kernel function to obtain the energy consumption estimation model.
Specifically, the method comprises the following steps:
(1) selecting a radial basis gaussian kernel as a kernel for support vector regression:
(2) determining an error penalty parameter C:
wherein,for the mean, σ, of the output values of model training sample datayThe standard deviation of the model training sample data output values is used.
(3) Assigning initial values to the kernel parameter sigma, constructing and solving the optimization problem of the epsilon-insensitive loss function to obtain the optimal solution
(4) Constructing an energy consumption estimation function to obtain an energy consumption estimation model:
wherein b is a hyperplane parameter, and the calculation formula is as follows:
wherein, yiAnd outputting a data value for the model training sample data, wherein epsilon is calculated by adopting the following formula:
wherein, σ is the standard deviation of the noise of the input data of the model training sample data, and l is the number of the input data of the model training sample data.
(5) Determining a suitable nuclear parameter σ: given a set of σi(i-1, 2, …, n) for each σiPerforming steps (3) and (4), calculating respective classification error degrees, and selecting the sigma with the minimum classification error degreeiAs the final gaussian kernel parameter σ.
The method comprises the following specific steps: dividing all model training sample data into k groups, taking the (k-1) group as a training group, and using the rest group for testing; repeating the steps, testing all sample groups, and taking the average value of the obtained classification error degrees as the final parameter sigmaiAn estimated classification error degree.
The calculation formula of the classification error degree Δ e is as follows:
wherein, Δ eiThe error value of the estimated value and the actual value in the test group is shown, and n is the number of the test group data in each group.
Step 103, inputting the model test sample data into the energy consumption estimation model to obtain an estimated energy consumption value, and obtaining an error rate of the estimated energy consumption value and an actual energy consumption value.
It is to be understood that, after the energy consumption estimation model is built according to the model training sample data, the model test sample data may be input into the energy consumption estimation model to test the energy consumption estimation model. Specifically, the step 103 includes:
equally dividing the model test sample data into preset groups of test set data, and respectively inputting the preset groups of test set data into the energy consumption estimation model to obtain preset groups of estimated energy consumption values;
and calculating the error rate between the estimated energy consumption value estimated by each group of the test set data and the actual energy consumption value of the corresponding model test sample data.
Specifically, the model test sample data may be equally divided into n test set data, each test set includes m groups of samples, and the n groups of test set data are respectively input into the energy consumption estimation model to obtain a test result, that is, an estimated energy consumption value estimated by each group of test set data.
Performing error rate P on the estimation result of each test set and the actual energy consumption value of the corresponding model test sample dataeIs calculated to obtain an error rate P based on the calculationeEvaluating the performance of the energy consumption estimation model, error rate PeThe evaluation formula of (2) is as follows:
wherein, ypjAs a result of the energy consumption estimation, yjAnd m is the number of samples in each group of test set for the actual energy consumption value.
And 104, correcting the energy consumption estimation model according to the error rate to obtain an optimized energy consumption estimation model.
Referring to fig. 2, specifically, the step 104 may include:
step 1041, determining whether an error rate between an estimated energy consumption value estimated by each group of the test set data and an actual energy consumption value of the corresponding model test sample data is less than a preset threshold;
1042, when the error rate corresponding to each group of test set data is smaller than the preset value, determining that an energy consumption estimation model obtained in the step of establishing the energy consumption estimation model based on a support vector regression algorithm by using the model training sample data is the energy consumption estimation model;
and step 104, entering the step S104 when the error rate corresponding to one or more groups of the test set data is larger than the preset value.
Further, the step 104 may include:
step 1043, when the error rate corresponding to one or more groups of the test set data is larger than the preset value, updating the one or more groups of the test set data to the model training sample data;
returning to the step 102.
In the embodiment of the present invention, the preset threshold may be 5%, that is, the error rate P is determinedeAnd adding more than 5% of the historical operating parameters of the test set into the model training sample data to form new model training sample data so as to finish updating the model training sample data.
Further, the energy consumption estimation model is optimized according to the updated model training sample data to obtain an optimized energy consumption estimation model. For example, according to the updated model training sample data, the energy consumption estimation model establishing step in step S102 is re-entered to modify the energy consumption estimation model, and the modified energy consumption estimation model is used as an optimized energy consumption estimation model, thereby completing establishment of the optimized energy consumption estimation model.
And when the error rate between each estimated energy consumption value and the actual energy consumption value of the corresponding model test sample data is less than 5%, the energy consumption estimation model is considered to be acceptable.
According to the technical scheme provided by the invention, historical operating parameters of a robot servo system are obtained, and the historical operating parameters are preprocessed to obtain model training sample data and model test sample data; establishing an energy consumption estimation model based on a support vector regression algorithm and by adopting the model training sample data; inputting the model test sample data into the energy consumption estimation model to obtain an estimated energy consumption value, and obtaining an error rate of the estimated energy consumption value and an actual energy consumption value; and correcting the energy consumption estimation model according to the error rate to obtain an optimized energy consumption estimation model. And further, complicated energy consumption analysis calculation on the robot servo system can be avoided, a robot servo system energy consumption estimation model is established based on historical operating parameters through a support vector regression algorithm, a large amount of time and energy are not needed to be consumed by people for energy consumption analysis, follow-up energy consumption estimation on the robot servo system is facilitated, and important guiding significance is provided for design and use of the robot.
Referring to fig. 3, fig. 3 is a flowchart illustrating a method for establishing an energy consumption estimation model of a robot servo system according to another embodiment of the present invention. As shown in fig. 2, the method for establishing the energy consumption estimation model of the robot servo system includes the following steps:
step 101, acquiring historical operating parameters of a robot servo system, and preprocessing the historical operating parameters to acquire model training sample data and model test sample data;
102, establishing an energy consumption estimation model based on a support vector regression algorithm and by adopting the model training sample data;
103, inputting the model test sample data into the energy consumption estimation model to obtain an estimated energy consumption value, and obtaining an error rate of the estimated energy consumption value and an actual energy consumption value;
step 104, correcting the energy consumption estimation model according to the error rate to obtain an optimized energy consumption estimation model;
and 105, periodically acquiring current operating parameters of the robot servo system, and updating the current operating parameters to model training sample data and model test sample data so as to optimize the energy consumption estimation model.
It should be noted that, for specific implementation and beneficial effects of the steps 101 to 104, reference may be made to the above embodiments, and details are not described in this embodiment.
In the embodiment of the invention, after the establishment of the energy consumption estimation model of the robot servo system is completed, the established energy consumption estimation model can be corrected and optimized by periodically acquiring the current operating parameters of the robot servo system, so that the accuracy of the energy consumption analysis of the robot servo system is ensured.
In this embodiment of the present invention, the step 105 includes:
clustering the model training sample data in the historical operating data sample by adopting a k-means algorithm;
taking data, which is in the model training sample data and has a distance with a sample central point larger than a preset distance, as abnormal data, and removing the abnormal data to update the model training sample data;
and respectively adding the current operating parameters into the model training sample data and the model test sample data according to a preset proportion to form new model training sample data and new model test sample data.
Specifically, a collection time limit may be set manually, or an automatic collection time limit may be set for the robot servo system, so as to periodically collect the current operation parameters of the robot servo system. The current operating parameter may be a packetComprises the following steps: stator phase current I and phase voltage U of servo motor, winding phase resistance R, rotor speed omega of servo motor, and friction torque tau of each rotating shaft of servo transmission jointfi(t), etc.
Clustering the model training sample data in the historical operating data sample by adopting a k-means algorithm, finding a sample central point, taking data, which is in the model training sample data and has a distance with the sample central point larger than a preset distance, as abnormal data, and removing the abnormal data so as to update the model training sample data. For example, data far away from the sample center point by 5% may be determined as abnormal data, and the abnormal data may be removed, so as to complete the update of the model training sample data. It should be noted that, the distance between the sample data features in the model training sample data may be calculated by using an euclidean distance calculation formula.
And then, respectively adding the current operating parameters into the model training sample data and the model test sample data according to a preset proportion to form new model training sample data and new model test sample data. For example, n groups of operation parameter data of the multi-joint robot servo system are collected periodically, the collected new operation parameter data are classified, the (n/2) group is used as a training learning sample and added into model training sample data, and the (n/2) group is used as an estimation comparison sample and added into model test sample data. And after the current operation parameters are respectively added into the model test sample data according to a preset proportion, the newly determined model test sample data only comprises the newly acquired current operation parameters.
It is of course understood that the ratio can be set as desired and is not particularly limited.
In an embodiment, the energy consumption estimation model is optimized by determining new model training sample data and new model test sample data from the updated historical operating data samples, and adding the updated model training sample data to the historical training sample data to perform the estimation of the energy consumption estimation model again.
For example, clustering the model training sample data in the historical operating data sample by adopting a k-means algorithm, taking the data, which is in the model training sample data and has a distance with a sample central point larger than a preset distance, as abnormal data, and removing the abnormal data; and periodically acquiring 100 groups of non-abnormal current operating parameters of the robot servo system, classifying the acquired current operating parameters, adding 50 groups of the current operating parameters into model training sample data as training learning samples, adding the other 50 groups of the current operating parameters into model test sample data as prediction comparison samples, repeating the step 102, and revising the energy consumption estimation model.
Grouping 50 groups of prediction comparison samples in the new model test sample data, wherein each group of 10 groups of data has 5 data sets, inputting each group of data into the modified energy consumption estimation model to obtain a prediction result, and repeating the step 103 to calculate an error rate, for example, when the error rate is less than 5% in advance, determining that the energy consumption estimation model is an acceptable model; and when the error rate is greater than the preset 5%, repeating the step 104, and correcting the energy consumption estimation model to obtain an optimized energy consumption estimation model.
According to the technical scheme provided by the invention, after the energy consumption estimation model is obtained, the current operation parameters of the robot servo system are periodically acquired, and the current operation parameters are updated to model training sample data so as to optimize the energy consumption estimation model. Therefore, the energy consumption estimation model can be corrected and optimized according to the operation data of the robot servo system, so that the accuracy of the energy consumption estimation model in estimating data can be ensured.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (8)
1. A method for establishing an energy consumption estimation model of a robot servo system is characterized by comprising the following steps:
acquiring historical operating parameters of a robot servo system, and preprocessing the historical operating parameters to acquire model training sample data and model test sample data;
establishing an energy consumption estimation model based on a support vector regression algorithm and by adopting the model training sample data;
inputting the model test sample data into the energy consumption estimation model to obtain an estimated energy consumption value, and obtaining an error rate of the estimated energy consumption value and an actual energy consumption value;
and correcting the energy consumption estimation model according to the error rate to obtain an optimized energy consumption estimation model.
2. The method of claim 1, wherein the steps of obtaining historical operating parameters of the robot servo system, and preprocessing the historical operating parameters to obtain model training sample data and model test sample data comprise:
acquiring historical operating parameters of a robot servo system, and carrying out normalization processing on the historical operating parameters;
and taking the historical operating parameters of the preset ratio in the historical operating parameters after the normalization processing as model training sample data, and taking the remaining historical operating parameters as model test sample data.
3. The method of claim 1, wherein the historical operating parameters include historical output parameters and historical input parameters including one or more of: the servo motor stator phase current, the phase voltage, the winding phase resistance, the servo motor rotor speed and the friction torque of each rotating shaft of the servo transmission joint.
4. The method of claim 1, wherein the step of building an energy consumption estimation model based on a support vector regression algorithm and using the model training sample data comprises:
and establishing an energy consumption estimation function by adopting the model training sample data based on a radial Gaussian kernel function to obtain the energy consumption estimation model.
5. The method of claim 1, wherein said step of inputting said model test sample data into said energy consumption estimation model to obtain an estimated energy consumption value and obtaining an error rate between said estimated energy consumption value and an actual energy consumption value comprises:
equally dividing the model test sample data into preset groups of test set data, and respectively inputting the preset groups of test set data into the energy consumption estimation model to obtain preset groups of estimated energy consumption values;
and calculating the error rate between the estimated energy consumption value estimated by each group of the test set data and the actual energy consumption value of the corresponding model test sample data.
6. The method of claim 5, wherein said step of modifying said energy consumption estimation model based on said error rate to obtain an optimized energy consumption estimation model comprises:
determining whether an error rate between an estimated energy consumption value estimated by each group of the test set data and an actual energy consumption value of corresponding model test sample data is less than a preset threshold value;
when the error rate corresponding to each group of the test set data is smaller than the preset value, determining that the step is based on a support vector regression algorithm and adopting the model training sample data to establish an energy consumption estimation model obtained in the energy consumption estimation model as a final energy consumption estimation model;
when the error rate corresponding to one or more groups of the test set data is larger than the preset value, updating the one or more groups of the test set data into the model training sample data;
and returning to the step of establishing the energy consumption estimation model based on the support vector regression algorithm and by adopting the model training sample data so as to obtain the optimized energy consumption estimation model.
7. The method according to any of claims 1-6, wherein said step of modifying said energy consumption estimation model according to said error rate to obtain an optimized energy consumption estimation model is followed by the further step of:
and periodically acquiring current operating parameters of the robot servo system, and updating the current operating parameters into model training sample data and model test sample data so as to optimize the energy consumption estimation model.
8. The method of claim 7, wherein the step of updating the current operating parameters into the model training sample data and model test sample data comprises:
clustering the model training sample data in the historical operating data sample by adopting a k-means algorithm;
taking data, which is in the model training sample data and has a distance with a sample central point larger than a preset distance, as abnormal data, and removing the abnormal data to update the model training sample data;
and respectively adding the current operating parameters into the model training sample data and the model test sample data according to a preset proportion to form new model training sample data and new model test sample data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810802066.0A CN108920863B (en) | 2018-07-20 | 2018-07-20 | Method for establishing energy consumption estimation model of robot servo system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810802066.0A CN108920863B (en) | 2018-07-20 | 2018-07-20 | Method for establishing energy consumption estimation model of robot servo system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108920863A true CN108920863A (en) | 2018-11-30 |
| CN108920863B CN108920863B (en) | 2021-02-09 |
Family
ID=64416656
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810802066.0A Active CN108920863B (en) | 2018-07-20 | 2018-07-20 | Method for establishing energy consumption estimation model of robot servo system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108920863B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110263618A (en) * | 2019-04-30 | 2019-09-20 | 阿里巴巴集团控股有限公司 | The alternative manner and device of one seed nucleus body model |
| CN110936382A (en) * | 2019-12-18 | 2020-03-31 | 哈尔滨工业大学 | Data-driven industrial robot energy consumption optimization method |
| CN111638642A (en) * | 2019-03-01 | 2020-09-08 | 北京国双科技有限公司 | Method and device for determining running state of three-phase separator |
| CN111666709A (en) * | 2020-04-26 | 2020-09-15 | 中国石油天然气股份有限公司 | Pipe conveying parameter adjusting method and device for gas conveying pipeline |
| CN111813624A (en) * | 2020-06-29 | 2020-10-23 | 中国平安人寿保险股份有限公司 | Robot execution time length estimation method based on time length analysis and related equipment thereof |
| CN111830350A (en) * | 2020-07-23 | 2020-10-27 | 珠海格力电器股份有限公司 | Energy consumption metering method and device and electric appliance |
| CN112749035A (en) * | 2019-10-31 | 2021-05-04 | 阿里巴巴集团控股有限公司 | Anomaly detection method, device and computer readable medium |
| CN112925320A (en) * | 2021-01-25 | 2021-06-08 | 河南大学 | Biped robot gait energy consumption assessment method based on centroid model |
| CN113752300A (en) * | 2021-09-22 | 2021-12-07 | 重庆工商大学 | Industrial robot energy consumption prediction method |
| CN116165680A (en) * | 2023-02-14 | 2023-05-26 | 中科三清科技有限公司 | Inversion model updating method and device, storage medium and chip |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102176701A (en) * | 2011-02-18 | 2011-09-07 | 哈尔滨工业大学 | Active learning based network data anomaly detection method |
| CN103150610A (en) * | 2013-02-28 | 2013-06-12 | 哈尔滨工业大学 | Fuzzy information granulation and support vector machine-based heating load prediction method |
| CN103544544A (en) * | 2013-10-29 | 2014-01-29 | 广东工业大学 | Energy consumption forecasting method and device |
| US20140095415A1 (en) * | 2012-09-28 | 2014-04-03 | Electronics And Telecommunications Research Institute | Apparatus and method for forecasting energy consumption |
| CN105224960A (en) * | 2015-11-04 | 2016-01-06 | 江南大学 | Based on the corn seed classification hyperspectral imagery model of cognition update method of clustering algorithm |
| CN105511445A (en) * | 2015-12-01 | 2016-04-20 | 沈阳化工大学 | Multi-modal process fault detection method based on local neighbor standardization matrix |
-
2018
- 2018-07-20 CN CN201810802066.0A patent/CN108920863B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102176701A (en) * | 2011-02-18 | 2011-09-07 | 哈尔滨工业大学 | Active learning based network data anomaly detection method |
| US20140095415A1 (en) * | 2012-09-28 | 2014-04-03 | Electronics And Telecommunications Research Institute | Apparatus and method for forecasting energy consumption |
| CN103150610A (en) * | 2013-02-28 | 2013-06-12 | 哈尔滨工业大学 | Fuzzy information granulation and support vector machine-based heating load prediction method |
| CN103544544A (en) * | 2013-10-29 | 2014-01-29 | 广东工业大学 | Energy consumption forecasting method and device |
| CN105224960A (en) * | 2015-11-04 | 2016-01-06 | 江南大学 | Based on the corn seed classification hyperspectral imagery model of cognition update method of clustering algorithm |
| CN105511445A (en) * | 2015-12-01 | 2016-04-20 | 沈阳化工大学 | Multi-modal process fault detection method based on local neighbor standardization matrix |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111638642A (en) * | 2019-03-01 | 2020-09-08 | 北京国双科技有限公司 | Method and device for determining running state of three-phase separator |
| CN110263618A (en) * | 2019-04-30 | 2019-09-20 | 阿里巴巴集团控股有限公司 | The alternative manner and device of one seed nucleus body model |
| CN110263618B (en) * | 2019-04-30 | 2023-10-20 | 创新先进技术有限公司 | Iteration method and device of nuclear body model |
| CN112749035A (en) * | 2019-10-31 | 2021-05-04 | 阿里巴巴集团控股有限公司 | Anomaly detection method, device and computer readable medium |
| CN112749035B (en) * | 2019-10-31 | 2024-06-11 | 阿里巴巴集团控股有限公司 | Abnormality detection method, abnormality detection device, and computer-readable medium |
| CN110936382A (en) * | 2019-12-18 | 2020-03-31 | 哈尔滨工业大学 | Data-driven industrial robot energy consumption optimization method |
| CN110936382B (en) * | 2019-12-18 | 2020-08-25 | 哈尔滨工业大学 | Data-driven industrial robot energy consumption optimization method |
| CN111666709B (en) * | 2020-04-26 | 2022-11-04 | 中国石油天然气股份有限公司 | Pipe conveying parameter adjusting method and device for gas conveying pipeline |
| CN111666709A (en) * | 2020-04-26 | 2020-09-15 | 中国石油天然气股份有限公司 | Pipe conveying parameter adjusting method and device for gas conveying pipeline |
| CN111813624A (en) * | 2020-06-29 | 2020-10-23 | 中国平安人寿保险股份有限公司 | Robot execution time length estimation method based on time length analysis and related equipment thereof |
| CN111813624B (en) * | 2020-06-29 | 2022-08-12 | 中国平安人寿保险股份有限公司 | Robot execution time length estimation method based on time length analysis and related equipment thereof |
| CN111830350A (en) * | 2020-07-23 | 2020-10-27 | 珠海格力电器股份有限公司 | Energy consumption metering method and device and electric appliance |
| CN112925320A (en) * | 2021-01-25 | 2021-06-08 | 河南大学 | Biped robot gait energy consumption assessment method based on centroid model |
| CN113752300A (en) * | 2021-09-22 | 2021-12-07 | 重庆工商大学 | Industrial robot energy consumption prediction method |
| CN116165680A (en) * | 2023-02-14 | 2023-05-26 | 中科三清科技有限公司 | Inversion model updating method and device, storage medium and chip |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108920863B (en) | 2021-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108920863B (en) | Method for establishing energy consumption estimation model of robot servo system | |
| CN109297689B (en) | Large-scale hydraulic machinery intelligent diagnosis method introducing weight factors | |
| CN109214708B (en) | Electric power system risk assessment method based on cross entropy theory optimization support vector machine | |
| CN103473480B (en) | Based on the online monitoring data bearing calibration improving universal gravitation support vector machine | |
| CN109409425B (en) | Fault type identification method based on neighbor component analysis | |
| CN108074015B (en) | Ultra-short-term prediction method and system for wind power | |
| CN106980910B (en) | Medium-and-long-term power load measuring and calculating system and method | |
| CN107403196B (en) | Method for predicting concentration of butane at bottom of debutanizer by instant learning modeling based on spectral clustering analysis | |
| CN116448419A (en) | Zero sample bearing fault diagnosis method based on depth model high-dimensional parameter multi-target efficient optimization | |
| Chou et al. | Physically consistent soft-sensor development using sequence-to-sequence neural networks | |
| CN116485049B (en) | Electric energy metering error prediction and optimization system based on artificial intelligence | |
| CN112465034B (en) | Method and system for establishing T-S fuzzy model based on hydraulic generator | |
| CN114583767B (en) | Data-driven wind power plant frequency modulation response characteristic modeling method and system | |
| CN114329347B (en) | Method and device for predicting metering error of electric energy meter and storage medium | |
| CN114861522A (en) | Precision manufacturing quality monitoring method and device based on artificial intelligence meta-learning technology | |
| CN113703506A (en) | Building material production workshop environment control and regulation method and system | |
| CN117587403A (en) | Aluminum alloy template surface treatment method and system based on chemical conversion | |
| CN116306292A (en) | Hydropower station physical field digital twin model construction method based on convolutional neural network | |
| CN115358473A (en) | Power load prediction method and prediction system based on deep learning | |
| CN105787507A (en) | LS-SVMs online learning method based on budget support vector set | |
| CN116085245A (en) | Online compressor performance prediction method and system based on OS-ELM | |
| CN114115150A (en) | Data-based heat pump system online modeling method and device | |
| CN115881882A (en) | Method for controlling electrode coating surface density and related product | |
| CN116595883B (en) | Real-time online system state correction method for numerical reactor | |
| CN115576288B (en) | Production control method and system for filter material |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |