CN118092199B - Prediction method for dynamic response time of steering engine - Google Patents
Prediction method for dynamic response time of steering engine Download PDFInfo
- Publication number
- CN118092199B CN118092199B CN202410520660.6A CN202410520660A CN118092199B CN 118092199 B CN118092199 B CN 118092199B CN 202410520660 A CN202410520660 A CN 202410520660A CN 118092199 B CN118092199 B CN 118092199B
- Authority
- CN
- China
- Prior art keywords
- steering engine
- response time
- data
- optimization
- constructing
- 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
- 230000004044 response Effects 0.000 title claims abstract description 149
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 59
- 238000012360 testing method Methods 0.000 claims abstract description 36
- 238000004458 analytical method Methods 0.000 claims abstract description 18
- 238000012417 linear regression Methods 0.000 claims abstract description 17
- 230000007613 environmental effect Effects 0.000 claims abstract description 13
- 238000005457 optimization Methods 0.000 claims description 80
- 230000002159 abnormal effect Effects 0.000 claims description 33
- 238000012544 monitoring process Methods 0.000 claims description 30
- 238000012545 processing Methods 0.000 claims description 15
- 238000007781 pre-processing Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 10
- 238000010606 normalization Methods 0.000 claims description 9
- 230000004043 responsiveness Effects 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000007689 inspection Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 8
- 238000009826 distribution Methods 0.000 description 7
- 238000005315 distribution function Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 230000001186 cumulative effect Effects 0.000 description 6
- 230000009471 action Effects 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 238000002790 cross-validation Methods 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 239000003638 chemical reducing agent Substances 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 230000002035 prolonged effect Effects 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000000844 transformation Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 241000282414 Homo sapiens Species 0.000 description 1
- 238000001276 Kolmogorov–Smirnov test Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000010687 lubricating oil Substances 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000036632 reaction speed Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/27—Regression, e.g. linear or logistic regression
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Medical Informatics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Automation & Control Theory (AREA)
- Combined Controls Of Internal Combustion Engines (AREA)
Abstract
The invention discloses a prediction method of steering engine dynamic response time, which relates to the technical field of steering engine control, and comprises the steps of constructing a steering engine operation prediction model by optimized motion data, testing the steering engine operation prediction model, constructing a precision coefficient by test data, and optimizing the constructed steering engine operation prediction model if the precision coefficient is lower than expected; predicting the response time of the steering engine by using the optimized steering engine operation prediction model, constructing prediction responsivity by using prediction data, carrying out linear regression analysis on the dynamic response time if the prediction responsivity exceeds the expectation, constructing the influence degree of environmental conditions on the response of the steering engine, optimizing the working environmental conditions or control algorithm of the steering engine, and predicting and acquiring the dynamic response time of the steering engine by using the optimized steering engine operation prediction model. And analyzing the reason of the slow response, and pertinently adjusting the working environment condition of the steering engine or a corresponding control algorithm to improve the problem of the slow response of the steering engine.
Description
Technical Field
The invention relates to the technical field of steering engine control, in particular to a method for predicting dynamic response time of a steering engine.
Background
The robot steering engine is a motor control device and is mainly used for controlling the movement of a robot joint to realize the movement control and the gesture control of a robot. The robot is one of key parts of the robot, and has obvious influence on the performance and the motion precision of the robot. The steering engine generally comprises a motor, a speed reducer, an encoder, a control circuit, a shell and the like. The working principle is that the motor is driven to rotate through a control signal, and then the steering wheel is driven to rotate after the motor is decelerated through the speed reducer, so that the joint motion of the robot is realized. Meanwhile, the encoder can monitor the position of the steering wheel in real time and feed back the position information to the control circuit so as to carry out accurate control.
In robot application, the steering engine plays a role in not only control precision and response speed, but also wide applicability. For example, in the automobile manufacturing process, a robot steering engine can control equipment such as a mechanical arm, a welding tool and the like, so that accurate action control and efficient production efficiency are realized; in the manufacturing process of the electronic product, the mechanical arm and the surface mounting equipment on the assembly line can be controlled, and the production efficiency and the product quality are improved; in the medical field, the robot steering engine can be used for controlling the movement and operation of surgical instruments, so that an accurate surgical effect is realized.
In the Chinese patent of the invention with the application publication number of CN109696825A, the invention is applicable to the technical field of robots, and provides a steering engine control response method and device and a steering engine, wherein the steering engine control response method comprises the following steps: acquiring an action instruction sent by a main control unit; acquiring instruction control parameters of a steering engine corresponding to the action instruction, and determining a plurality of working time periods with corresponding relation with the instruction control parameters; acquiring the current running time of a steering engine; and determining the working time period of the running time, and executing the response to the action instruction based on the instruction control parameter corresponding to the working time period.
Combining the contents of the above applications and prior art:
Considering the influence of the response time of the steering engine on the corresponding equipment, before the steering engine is started, the response time of the steering engine is usually required to be calculated or predicted, the subsequent work tasks are conveniently arranged according to the response time, if the response time of the steering engine is too long, the robot is possibly difficult to work normally, the response time of the steering engine is predicted through a machine learning model or linear regression before the steering engine is started, and corresponding prediction results are obtained, but because the response time of the steering engine is influenced by more factors when the steering engine actually works, for example, when the steering engine is high in load and has a certain degree of abnormality in the working temperature, the influence on the response time of the steering engine is possibly nonlinear, and the influence is possibly accumulated, under a complex working condition, if the response time of the steering engine is still predicted through a linear regression or a simple machine learning model, a certain deviation between a predicted value and an actual value is possibly caused, and in the existing prediction method, when the response time of the steering engine is slow, the steering engine is usually only early-warned, and corresponding processing is not carried out, so that after the response time of the steering engine is still slow.
Therefore, the invention provides a method for predicting the dynamic response time of the steering engine.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a steering engine dynamic response time prediction method, which is characterized in that a steering engine operation prediction model is tested, an accuracy coefficient is built by test data, and if the accuracy coefficient is lower than expected, the built steering engine operation prediction model is optimized; predicting the response time of the steering engine by using the optimized steering engine operation prediction model, constructing prediction responsivity by using prediction data, carrying out linear regression analysis on the dynamic response time if the prediction responsivity exceeds the expectation, constructing the influence degree of environmental conditions on the response of the steering engine, optimizing the working environmental conditions or control algorithm of the steering engine, and predicting and acquiring the dynamic response time of the steering engine by using the optimized steering engine operation prediction model. And the reasons of too slow response are analyzed, and the working environment conditions of the steering engine or the corresponding control algorithm are adjusted in a targeted manner, so that the problems in the background technology are solved.
In order to achieve the above purpose, the invention is realized by the following technical scheme: a method for predicting dynamic response time of a steering engine monitors operation conditions of the steering engine, and generates condition coefficients of steering engine operation according to steering engine operation condition data obtained through monitoringIf the condition coefficient exceeds the expected value, an early warning instruction is sent to the outside;
Collecting and preprocessing the motion data of the steering engine, constructing a corresponding motion data column, screening out abnormal values in the motion data column, marking the abnormal values in the motion data column, generating a replacement value to replace the abnormal values, and completing the optimization of the motion data;
Constructing a steering engine operation prediction model according to the optimized motion data, testing the steering engine operation prediction model, and constructing an accuracy coefficient according to the test data If the precision coefficientOptimizing the constructed steering engine operation prediction model below the expected value; wherein the precision coefficientThe acquisition mode of (a) is as follows: rate of recallMean square errorPerforming linear normalization processing, and mapping corresponding data values to intervalsIn, according to the following formula:
Weight coefficient: ， And (2) and ，Wherein,The number of tests; For the recall at the ith test, Is used as a qualified standard value of the test paper,For the mean square error at the ith test,Is a qualified standard value of the mean square error;
predicting steering engine response time by using the optimized steering engine operation prediction model, and constructing prediction responsiveness by using prediction data If the responsivity is predictedPerforming linear regression analysis on the dynamic response time beyond expectation, and acquiring a corresponding regression equation;
And constructing influence of environmental conditions on steering engine response by using a linear regression equation, sending a corresponding optimization instruction to the outside according to the relation between the influence and a corresponding influence threshold, optimizing the working environmental conditions or control algorithm of the steering engine, and predicting and acquiring the dynamic response time of the steering engine by using an optimized steering engine operation prediction model after optimization.
Further, a monitoring period comprising a plurality of monitoring nodes is set, the operation conditions of the steering engine are monitored on the monitoring nodes, after operation condition data on the plurality of monitoring nodes are continuously obtained, a steering engine operation condition data set is constructed in a summarizing mode, and a condition coefficient of steering engine operation is generated by the steering engine operation condition data。
Further, the condition coefficientThe acquisition mode of (a) is as follows: for work loadOperating temperaturePerforming linear normalization processing to map corresponding data values to intervalsIn, according to the following formula:
Weight coefficient: ， And is also provided with ；，The number of the nodes is monitored; for the operating temperature of the ith monitoring node, As an average value of the operating temperature,Is a qualified reference value of the working temperature; For the workload of the ith monitoring node, For the mean value of the workload on each monitoring node,Is a qualified reference value for the workload.
Further, after receiving the early warning instruction, collecting motion data of the steering engine under various conditions, and collecting the collected data to obtain a steering engine running state data set; preprocessing steering engine running state data in the steering engine running state data set, and screening out abnormal values after finishing data preprocessing.
Further, generating a replacement value, and replacing the abnormal value in the motion data column by the replacement value to obtain an optimized motion data column; wherein, the motion data column is subjected to function fitting, and after K-S normal inspection, a corresponding fitting function is obtained, and according to the position of the abnormal value, the fitting value given by the fitting functionAnd giving interpolation by spline interpolationThereby constructing a replacement valueThe mode is as follows:
Weight coefficient: ， And is also provided with 。
Further, part of optimized data is obtained from the steering engine running state data set to serve as sample data, an initial model is built by a deep neural network, the initial model is trained, and a trained steering engine running prediction model is obtained; after several tests, constructing corresponding precision coefficient from the test dataIf the precision coefficientAnd (3) optimizing the constructed steering engine operation prediction model below the precision threshold.
Further, the dynamic response time of the steering engine is predicted by using the optimized steering engine operation prediction model, and after a plurality of predictions, the response time is predictedConstructing a predictive responsivity comprising: response time toPerforming linear normalization processing according to the following modes:
Wherein, N, n is the number of predictions,For the response time at the ith prediction,Weight coefficient, which is a qualified standard value of response time:， And (2) and 。
Further, if the obtained prediction responsivityThe responsiveness threshold value is not exceeded, and the average value of the prediction data of a plurality of times is taken as a final prediction value; and if the time is exceeded, the workload, the working temperature and the dynamic response time on each time node are obtained one by one along the time axis, the workload, the working temperature and the dynamic response time are taken as independent variables, and the linear regression analysis is carried out by taking the independent variables as the independent variables to obtain a corresponding regression equation.
Further, obtaining regression coefficients corresponding to the independent variables from a regression equation, determining corresponding weight coefficients for the independent variables by using a hierarchical analysis method by taking the regression coefficients as influence factors, and obtaining influence degree of environmental conditions on steering engine response by combining weight coefficient weighted average on a plurality of influence factors; if the obtained influence exceeds the influence threshold, a first instruction is sent to the outside for optimization; if the first optimization instruction does not exceed the second optimization instruction, a second optimization instruction is sent to the outside.
Further, after receiving the first optimization instruction, identifying current working condition data of the steering engine to obtain a plurality of condition optimization features; constructing a condition optimization model by an optimization algorithm, optimizing the condition optimization characteristics, and obtaining optimized steering engine working conditions;
After receiving the second optimization instruction, acquiring a current control algorithm of the steering engine, and identifying a plurality of preset parameters of the acquisition control algorithm; carrying out feature recognition on preset parameters to obtain corresponding optimized features; selecting steering engine control algorithm optimization as a target word, and constructing a control algorithm optimization knowledge graph after deep search; matching an optimization scheme for the current control algorithm according to the correspondence between the optimization features and the optimization scheme of the control algorithm; and executing the optimization scheme to optimize a control algorithm of the steering engine.
The invention provides a method for predicting the dynamic response time of a steering engine, which has the following beneficial effects:
1. According to the condition coefficient Judging whether the current working condition of the steering engine is abnormal or not, if so, adjusting the current working condition of the steering engine is needed, and adjusting the load and the working temperature of the steering engine in time by sending out early warning, so that the steering engine is prevented from being damaged due to the fact that the steering engine is in the abnormal working condition for a long time, and the service life of the steering engine is prolonged.
2. After preliminary pretreatment is carried out on the motion data, optimization is carried out again, and abnormal data are screened on the basis of combining the change trend of the data, so that the data quality can be further improved; and the final replacement value is obtained through the combination of function fitting and spline interpolation, the abnormal value is replaced by the replacement value, the reliability of the replacement value can be improved after the working condition of the steering engine is considered, and the risk of generating larger errors during the calculation and prediction of the response time of the steering engine by the abnormal data is reduced.
3. Analyzing and obtaining precision coefficientBy a precision coefficientAnd evaluating the accuracy and the reliability of the prediction model, training and optimizing by reselecting sample data under the condition of insufficient reliability, improving the reliability and the accuracy of the prediction model, and obtaining more reliable results when the prediction model is used for predicting and calculating the response time of the steering engine.
4. From predicted responsivityJudging the reliability of the current response time, if the reliability exceeds the expected reliability, indicating that the prediction process of the current steering engine response time is finished, if the reliability is still unreliable, judging which factors have a larger influence on the steering engine response, and confirming the influence degree of each influence factor on the steering engine response through linear regression analysis, wherein the influence degree is used as a reference when the steering engine work needs to be optimized.
5. After receiving the first or second optimization instruction, judging and confirming the reason causing the slow response time of the steering engine, respectively constructing an optimization model and a knowledge graph, if the reason causing the slow response of the steering engine is the working condition of the steering engine, optimizing each current working condition by using an optimization algorithm, otherwise, if the result is caused by the control algorithm of the steering engine, optimizing the knowledge graph by the constructed control algorithm, and providing a corresponding optimization scheme for the control algorithm; therefore, when the response time of the steering engine is predicted and obtained and the response time is confirmed to be slower, the response problem of the steering engine is pertinently improved, and the response of the steering engine is quickened.
6. When the response time of the steering engine is required to be predicted and calculated, judging whether the working environment condition of the steering engine is abnormal, if so, processing the steering engine in a targeted manner, accelerating the response of the steering engine, analyzing the reliability of the prediction model after constructing the prediction model, and if the reliability is insufficient, improving the problem through targeted optimization, and finishing the prediction of the response time of the steering engine; after the steering engine response time prediction is completed, whether the steering engine response time is qualified or not is judged, and when the steering engine response is too slow, the reason of too slow response is analyzed, so that the steering engine working environment condition or a corresponding control algorithm can be adjusted in a targeted manner, and the problem of slow steering engine response is improved.
Drawings
Fig. 1 is a flow chart of a method for predicting the dynamic response time of a steering engine.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but 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.
Referring to fig. 1, the invention provides a method for predicting dynamic response time of a steering engine, which comprises the following steps:
step one, monitoring the operation condition of a steering engine, and generating a condition coefficient of the operation of the steering engine according to the steering engine operation condition data obtained by monitoring If the condition coefficient exceeds the expected value, an early warning instruction is sent to the outside;
The first step comprises the following steps:
Step 101, when a steering engine is in an operating state, setting a monitoring period comprising a plurality of monitoring nodes, and monitoring the operating conditions of the steering engine on each monitoring node, wherein the monitoring period comprises the current working load and the current working temperature of the steering engine; after continuously acquiring the operation condition data on a plurality of monitoring nodes, summarizing to construct a steering engine operation condition data set;
102, generating a condition coefficient of steering engine operation from steering engine operation condition data The mode is as follows: for work loadOperating temperaturePerforming linear normalization processing to map corresponding data values to intervalsIn, according to the following formula:
Weight coefficient: ， And is also provided with ；，The number of the nodes is monitored; for the operating temperature of the ith monitoring node, As an average value of the operating temperature,Is a qualified reference value of the working temperature; For the workload of the ith monitoring node, For the mean value of the workload on each monitoring node,Is a qualified reference value for the workload;
presetting a condition threshold according to historical data and management expectation of safe operation of the steering engine, and if the obtained condition coefficient is When the condition threshold value is exceeded, namely the condition coefficient exceeds the expected value, the current operation environment of the steering engine possibly affects the operation state of the steering engine, and at the moment, an early warning instruction is sent to the outside;
If the load that the steering engine needs to drive changes, its response time may be affected. A heavier load may result in longer response times and a lighter load may result in shorter response times. High temperatures may reduce the efficiency of the motor, lengthening the response time. Low temperatures may increase the viscosity of the lubricating oil and may also affect response time.
In use, the contents of steps 101 and 102 are combined:
When the steering engine is in an operating state, the current working condition of the steering engine is monitored, and a condition coefficient is constructed by corresponding monitoring data According to the condition coefficientJudging whether the current working condition of the steering engine is abnormal or not, if so, adjusting the current working condition of the steering engine is needed, and at the moment, adjusting the load and the working temperature of the steering engine in time by sending out early warning, so that the steering engine can be prevented from being damaged due to the fact that the steering engine is in the abnormal working condition for a long time, and the service life of the steering engine is prolonged.
Combining the contents of the above applications and prior art:
Considering the influence of the response time of the steering engine on the corresponding equipment, before the steering engine is started, the response time of the steering engine is usually required to be calculated or predicted, the subsequent work tasks are conveniently arranged according to the response time, if the response time of the steering engine is too long, the robot is possibly difficult to work normally, the response time of the steering engine is predicted through a machine learning model or linear regression before the steering engine is started, and corresponding prediction results are obtained, but because the response time of the steering engine is influenced by more factors when the steering engine actually works, for example, when the steering engine is high in load and has a certain degree of abnormality in the working temperature, the influence on the response time of the steering engine is possibly nonlinear, and the influence is possibly accumulated, under a complex working condition, if the response time of the steering engine is still predicted through a linear regression or a simple machine learning model, a certain deviation between a predicted value and an actual value is possibly caused, and in the existing prediction method, when the response time of the steering engine is slow, the steering engine is usually only early-warned, and corresponding processing is not carried out, so that after the response time of the steering engine is still slow.
Step two, collecting and preprocessing the motion data of the steering engine, constructing a corresponding motion data column, screening out abnormal values in the motion data column, marking the abnormal values in the motion data column, generating a replacement value to replace the abnormal values, and completing the optimization of the motion data;
the second step comprises the following steps:
Step 201, after receiving an early warning instruction, collecting motion data of a steering engine under various conditions, including angles, speeds, moments, currents and the like, wherein during collection, the motion data can be collected through a sensor or a built-in data collection system; collecting collected data, and then obtaining a steering engine running state data set;
step 202, preprocessing steering engine running state data in a steering engine running state data set to improve the accuracy of an algorithm, wherein the preprocessing comprises the following steps of; identifying, acquiring and filling missing values, performing standardized processing on data, and screening abnormal values after finishing data preprocessing, wherein:
according to the time of data acquisition, the preprocessed motion data are arranged along a time axis to obtain a corresponding motion data column; an upper operation threshold is constructed in advance according to historical data and management expectation of the operation state of the steering engine Lower operation thresholdForm a conventional threshold of movementThe following may be specifically referred to:
Wherein n is motion data Is set in the number of (3),As a mean value of the running coefficients,I-th data in the motion data column; if it isDataNot at the normal thresholdIn, marking the corresponding motion data as an outlier;
When the method is used, abnormal values are screened by constructing a threshold range and arranging the motion data, the motion data can be optimized again after preliminary pretreatment, the abnormal data is screened on the basis of combining the change trend of the data, and the data quality can be further improved;
step 203, generating a replacement value to replace the abnormal value, and completing data optimization in the following manner: performing function fitting on the motion data sequence, obtaining a corresponding fitting function after K-S normal inspection, and according to the position of the abnormal value, giving a fitting value by the fitting function And giving interpolation by spline interpolationThereby constructing a replacement valueThe mode is as follows:
Weight coefficient: ， And is also provided with ; The weight coefficient may be consistent with the previous value;
replacing the abnormal value in the motion data column by the replacement value to obtain an optimized motion data column;
in use, the contents of steps 201 to 203 are combined:
On the basis of screening abnormal data from a motion data column, acquiring a final replacement value through the synthesis of function fitting and spline interpolation, replacing the abnormal value with the replacement value, improving the reliability of the replacement value after considering the working condition of a steering engine, and reducing the risk of generating larger errors when calculating and predicting the response time of the steering engine by the abnormal data; conversely, if it is difficult to secure the reliability of the replacement value, it is also difficult to secure the reliability of the data optimization.
Thirdly, constructing a steering engine operation prediction model according to the optimized motion data, testing the steering engine operation prediction model, and constructing an accuracy coefficient according to the test dataIf the precision coefficientOptimizing the constructed steering engine operation prediction model below the expected value;
the third step comprises the following steps:
Step 301, obtaining part of optimized data from a steering engine running state data set as sample data, dividing the sample data into a training set, a test set and a verification set, constructing an initial model by a deep neural network, training the initial model by using the training set, and obtaining a trained steering engine running prediction model;
Step 302, testing the trained steering engine operation prediction model on a test set, obtaining corresponding test data after a plurality of tests, and constructing corresponding precision coefficients by the test data, wherein the precision of the steering engine operation prediction model is evaluated by the test data in the following manner: rate of recall Mean square errorPerforming linear normalization processing, and mapping corresponding data values to intervalsIn, according to the following formula:
Weight coefficient: ， And (2) and ，Wherein,The number of tests; For the recall at the ith test, Is used as a qualified standard value of the test paper,For the mean square error at the ith test,Is a qualified standard value of the mean square error; the weight coefficient is obtained by a reference analytic hierarchy process;
Step 303, presetting an accuracy threshold according to historical data and management expectations of model accuracy; if the precision coefficient If the accuracy threshold value is lower than the accuracy threshold value, the steering engine operation prediction model is difficult to give data with higher reliability, the model needs to be further optimized, and at the moment, the data is reselected to train the model; optimizing the constructed steering engine operation prediction model, adjusting parameters of the model to improve prediction accuracy and the like, and verifying on a verification set;
in use, the contents of steps 301 to 303 are combined:
When the response time of the steering engine is predicted and calculated by constructing a steering engine operation prediction model, after the reliability of the preliminarily trained prediction model is tested, the accuracy coefficient is obtained by analysis By a precision coefficientThe accuracy and the reliability of the prediction model are evaluated, under the condition that the reliability is insufficient, sample data are reselected for training and optimizing, and the prediction model is subjected to targeted compensation and compensation, so that the reliability and the accuracy of the prediction model can be improved, and when the prediction model is used for predicting and calculating the response time of a steering engine, the obtained result is more reliable.
Predicting steering engine response time by using the optimized steering engine operation prediction model, and constructing prediction responsiveness by prediction dataIf the responsivity is predictedPerforming linear regression analysis on the dynamic response time beyond expectation, and acquiring a corresponding regression equation;
The fourth step comprises the following steps:
step 401, predicting the dynamic response time of the steering engine by using the optimized steering engine operation prediction model, and respectively obtaining corresponding prediction response time after a plurality of predictions, wherein the prediction response time is used for Constructing a predictive responsivity, wherein, for response timePerforming linear normalization processing according to the following modes:
Wherein, N, n is the number of predictions,For the response time at the ith prediction,Weight coefficient, which is a qualified standard value of response time:， And (2) and ; Wherein the weight coefficientAndThe values of (2) and (d) appear aboveAndCan be kept consistent;
Step 402, presetting a responsiveness threshold according to historical data and management expectations of steering engine responses; if the obtained prediction responsivity If the response threshold is not exceeded, the steering engine reaction speed is higher, and the expectation can be met; taking the average value of the predicted data of a plurality of times as a final predicted value;
If the expectation is not met, the workload, the working temperature and the dynamic response time on each time node are obtained one by one along a time axis, the workload, the working temperature and the dynamic response time are taken as independent variables, and linear regression analysis is carried out by taking the independent variables as the dependent variables to obtain a corresponding regression equation;
in use, the contents of steps 401 and 402 are combined:
after the steering engine response time is calculated, the predicted responsiveness is built through the predicted response time which is continuously acquired for many times From predicted responsivityJudging the reliability of the current response time, if the reliability exceeds the expected reliability, the current prediction process of the steering engine response time is finished, if the reliability is still unreliable, judging factors which have great influence on the steering engine response are needed, at this time, confirming the influence degree of each influence factor on the steering engine response through linear regression analysis, and taking the influence degree as a reference when the steering engine work needs to be optimized.
Fifthly, constructing influence degree of environmental conditions on response of the steering engine by using a linear regression equation, sending corresponding optimization instructions to the outside according to the relation between the influence degree and a corresponding influence degree threshold value, optimizing the working environmental conditions or control algorithm of the steering engine, and predicting and obtaining dynamic response time of the steering engine by using an optimized steering engine operation prediction model after optimization;
the fifth step comprises the following steps:
Step 501, obtaining regression coefficients corresponding to independent variables from a regression equation, determining corresponding weight coefficients for the independent variables by using a hierarchical analysis method by taking the regression coefficients as influence factors, and obtaining influence degree of environmental conditions on steering engine response by combining weight coefficient weighted average on a plurality of influence factors; pre-constructing an influence threshold according to historical data and management expectation of steering engine working conditions; if the acquired influence degree exceeds the influence degree threshold, the steering engine response is relatively greatly influenced by the environment, and at the moment, a first instruction is sent to the outside for optimization; if the current response speed of the steering engine is not exceeded, the fact that the current response speed of the steering engine is low is possibly caused by the fact that the control algorithm is insufficient is indicated, and at the moment, a second optimization instruction is sent to the outside;
When the system is used, when the response time of the steering engine is unreasonable, under the cooperation of the analytic hierarchy process, the influence degree is obtained through calculation of regression coefficients corresponding to the independent variables, the influence degree applied to the response of the steering engine by the influence degree is comprehensively evaluated, and the reason that the response time of the steering engine is slower can be judged according to the evaluation result.
Step 502, after receiving a first optimization instruction, identifying current working condition data of the steering engine, such as working environment, load size, temperature, voltage and the like of the steering engine, taking corresponding influence factors as references, and obtaining a plurality of condition optimization features after setting an optimization standard; constructing a condition optimization model by an optimization algorithm, taking shortening the response time of the steering engine as an optimization target, optimizing the condition optimization characteristics, and obtaining the optimized working condition of the steering engine;
Step 503, after receiving the second optimizing instruction, obtaining a current control algorithm of the steering engine, and identifying a plurality of preset parameters of the obtained control algorithm; performing feature recognition on preset parameters, and acquiring corresponding optimized features after setting an optimization standard; selecting steering engine control algorithm optimization as a target word, and constructing a control algorithm optimization knowledge graph after deep search;
Matching an optimization scheme for the current control algorithm according to the correspondence between the optimization features and the optimization scheme of the control algorithm; and executing the optimization scheme to optimize a control algorithm of the steering engine.
In use, the contents of steps 501 to 503 are combined:
After receiving the first or second optimization instruction, judging and confirming the reason causing the slow response time of the steering engine, respectively constructing an optimization model and a knowledge graph, if the reason causing the slow response of the steering engine is the working condition of the steering engine, optimizing each current working condition by using an optimization algorithm, otherwise, if the result is caused by the control algorithm of the steering engine, optimizing the knowledge graph by the constructed control algorithm, and providing a corresponding optimization scheme for the control algorithm; therefore, when the response time of the steering engine is predicted and obtained and the response time is confirmed to be slower, the response problem of the steering engine is pertinently improved, and the response of the steering engine is quickened.
When the response time of the steering engine is required to be predicted and calculated, judging whether the working environment condition of the steering engine is abnormal or not, if so, processing the steering engine pertinently, accelerating the response of the steering engine, analyzing the reliability of the prediction model after constructing the prediction model, and if the reliability is insufficient, improving the problem through pertinently optimizing, and finishing the prediction of the response time of the steering engine; further, after the steering engine response time prediction is completed, whether the steering engine response time is qualified or not is judged, and when the steering engine response is too slow, the reason of the too slow response is analyzed, so that the steering engine working environment condition or a corresponding control algorithm can be adjusted in a targeted manner, and the problem of slow steering engine response is improved.
It should be noted that;
When the model needs to be optimized, various strategies can be adopted to optimize the performance of the model. The method comprises the following steps: characteristic engineering:
Feature selection: the features most relevant to the task are selected, removing noise or extraneous features. Feature extraction: feature quantity is reduced by a dimension reduction method (such as PCA, t-SNE and the like), and important information is reserved at the same time. Feature conversion: nonlinear transformations (e.g., polynomial features, logarithmic transformations, etc.) are used to create new, possibly more useful features.
Model selection: different types of models, such as linear models, decision trees, random forests, neural networks, etc., are tried to see which model is better suited for the current task and data. For complex datasets, an integration method (e.g., bagging, boosting) or Stacking method (Stacking) may help to improve performance.
Parameter adjustment: a grid Search (GRID SEARCH) or a Random Search (Random Search) is used to find the optimal parameter combination of the model. For deep learning models, optimization techniques such as learning rate decay, momentum, etc. may be used to adjust the training process.
Regularization: the use of L1 or L2 regularization to prevent model overfitting helps reduce model complexity and avoids fitting noise in the training data.
Processing unbalanced data: if there is a class imbalance problem in the dataset, an oversampling or undersampling technique may be used to balance the number of samples of each class. Cost-sensitive learning (Cost-SENSITIVE LEARNING) may also be used in an attempt to assign different weights to different classes of errors.
Cross-validation: model performance was assessed using K-fold cross-validation and the model with the best average performance was selected. Cross-validation can also be used to select optimal hyper-parameters, avoiding over-fitting or under-fitting.
Model fusion: the prediction results of the multiple models are combined, such as by voting (Voting) or weighted averaging (Blending/Stacking), to improve the accuracy of the final prediction.
Error analysis: the analytical model predicts errors on which samples and attempts to find the cause. This can help locate the problem and optimize it specifically. Based on the results of the error analysis, features, models, or training strategies may be adjusted.
Data enhancement: for data types such as images or texts, more training samples can be generated by using a data enhancement technology, so that the generalization capability of the model is improved.
It should be noted that:
The construction of the control algorithm optimization knowledge graph is a complex and systematic process, and mainly relates to the structural and graphical representation of knowledge in the field of control algorithms. The following basic steps are used for constructing a control algorithm to optimize a knowledge graph:
explicit goal and scope: first, the target and the range of the knowledge graph need to be defined, and the type of the control algorithm, the optimization method, the application field and the like to be covered are determined.
Collecting and controlling data in the field of algorithm optimization: this includes collecting relevant data of the principles, features, application scenarios, optimization methods, etc. of various control algorithms. Such data may originate from academic papers, technical documents, expert interviews, and the like.
Data cleaning and pretreatment: and cleaning and preprocessing the collected data, removing redundant and erroneous information, and ensuring the accuracy and consistency of the data.
Defining entities and relationships: in the field of control algorithm optimization, entities may include various control algorithms, optimization methods, parameters, performance indicators, etc., and relationships describe interactions and links between these entities. For example, a relationship of "algorithm a use optimization method B" may be defined.
Knowledge extraction and entity linking: entities and relationships are extracted from the text data using techniques such as natural language processing and linked to corresponding knowledge bases.
And (3) constructing a knowledge graph: and constructing a knowledge graph based on the extracted entities and the relationships. This typically involves using techniques such as graph databases to store and query the graph data.
Querying and reasoning of the atlas: after the knowledge graph construction is completed, various query and reasoning operations can be supported, such as searching for an optimization method of a certain control algorithm, comparing the performances of different algorithms, and the like.
Updating and maintaining the map: with the development of the control algorithm field and the generation of new knowledge, the knowledge graph needs to be continuously updated and maintained, so that the timeliness and the accuracy of the knowledge graph are ensured.
In the construction process, the following points should be noted:
Ensuring the data quality: the accuracy and integrity of the data is the key to constructing a high quality knowledge graph. Selection of appropriate tools and techniques: and selecting a proper graph database, a natural language processing tool and the like according to actual requirements. Considering scalability and maintainability: reasonable architecture and interfaces are designed so that knowledge maps can be conveniently expanded and maintained in the future. By constructing the control algorithm optimization knowledge graph, the systematic management and the efficient utilization of the knowledge in the field of the control algorithm can be realized, and powerful support is provided for algorithm research, optimization and application.
It should be noted that;
The analytic hierarchy process is a decision analysis method, which decomposes elements related to decision into levels of targets, criteria, schemes, etc., and then performs qualitative and quantitative analysis on the basis of the analysis. The core of the method is that the complex decision problem is decomposed into a series of layering factors, and a layering hierarchy is established, so that the judgment of human beings is converted into the comparison of importance between a plurality of factors, and the qualitative judgment which is difficult to quantify is converted into the comparison of operational importance.
The K-S normal test, the Kolmogorov-Smirnov test, is a non-parametric test method for testing whether a sample is from a particular theoretical distribution, particularly a normal distribution. The main principle is to compare the maximum absolute difference between the Cumulative Distribution Function (CDF) of the samples and the CDF of the theoretical distribution.
The specific steps of the K-S normal test are as follows:
A cumulative distribution function Fn (x) of the given sample data is calculated. This is typically accomplished by sorting the sample data from small to large and then computing a cumulative distribution function according to a formula.
A cumulative distribution function F0 (x) of the assumed normal distribution is calculated. This can be done by using the formula of a normal distribution function. The maximum absolute difference Dn between the sample cumulative distribution function Fn (x) and the cumulative distribution function F0 (x) assuming normal distribution is calculated.
The calculated maximum absolute difference Dn is compared with a known probability value dα. If Dn is less than Dα, then the sample may be considered to fit a normal distribution; otherwise, it indicates that the sample does not conform to the normal distribution.
The K-S normal test has the advantage that it is not affected by extreme values and sample skew, and can give accurate results even with a small sample size. Furthermore, it can also be used to detect simple non-normal hypotheses and thus has a wide range of applications.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A prediction method of the dynamic response time of a steering engine is characterized by comprising the following steps: comprising the steps of (a) a step of,
Monitoring the operation condition of the steering engine, and generating a condition coefficient of the steering engine operation according to the steering engine operation condition data obtained by monitoringIf the condition coefficient exceeds the expected value, an early warning instruction is sent to the outside;
Collecting and preprocessing the motion data of the steering engine, constructing a corresponding motion data column, screening out abnormal values in the motion data column, marking the abnormal values in the motion data column, generating a replacement value to replace the abnormal values, and completing the optimization of the motion data;
Constructing a steering engine operation prediction model according to the optimized motion data, testing the steering engine operation prediction model, and constructing an accuracy coefficient according to the test data If the precision coefficientOptimizing the constructed steering engine operation prediction model below the expected value; wherein the precision coefficientThe acquisition mode of (a) is as follows: rate of recallMean square errorPerforming linear normalization processing, and mapping corresponding data values to intervalsIn, according to the following formula:
Weight coefficient: ， And (2) and ，Wherein,The number of tests; For the recall at the ith test, Is used as a qualified standard value of the test paper,For the mean square error at the ith test,Is a qualified standard value of the mean square error;
predicting steering engine response time by using the optimized steering engine operation prediction model, and constructing prediction responsiveness by using prediction data If the responsivity is predictedPerforming linear regression analysis on the dynamic response time beyond expectation, and acquiring a corresponding regression equation;
And constructing influence of environmental conditions on steering engine response by using a linear regression equation, sending a corresponding optimization instruction to the outside according to the relation between the influence and a corresponding influence threshold, optimizing the working environmental conditions or control algorithm of the steering engine, and predicting and acquiring the dynamic response time of the steering engine by using an optimized steering engine operation prediction model after optimization.
2. The method for predicting the dynamic response time of the steering engine according to claim 1, wherein the method comprises the following steps:
Setting a monitoring period comprising a plurality of monitoring nodes, monitoring the operation conditions of the steering engine on the monitoring nodes, continuously acquiring the operation condition data on the monitoring nodes, summarizing to construct a steering engine operation condition data set, and generating a condition coefficient of steering engine operation by the steering engine operation condition data 。
3. The method for predicting the dynamic response time of the steering engine according to claim 2, wherein the method comprises the following steps:
Condition coefficient The acquisition mode of (a) is as follows: for work loadOperating temperaturePerforming linear normalization processing to map corresponding data values to intervalsIn, according to the following formula:
Weight coefficient: ， And is also provided with ；，The number of the nodes is monitored; for the operating temperature of the ith monitoring node, As an average value of the operating temperature,Is a qualified reference value of the working temperature; For the workload of the ith monitoring node, For the mean value of the workload on each monitoring node,Is a qualified reference value for the workload.
4. The method for predicting the dynamic response time of the steering engine according to claim 1, wherein the method comprises the following steps:
After receiving the early warning instruction, collecting motion data of the steering engine under various conditions, and collecting the collected data to obtain a steering engine running state data set; preprocessing steering engine running state data in the steering engine running state data set, and screening out abnormal values after finishing data preprocessing.
5. The method for predicting the dynamic response time of the steering engine according to claim 4, wherein the method comprises the following steps:
generating a replacement value, and replacing the abnormal value in the motion data column by the replacement value to obtain an optimized motion data column; wherein, the motion data column is subjected to function fitting, and after K-S normal inspection, a corresponding fitting function is obtained, and according to the position of the abnormal value, the fitting value given by the fitting function And giving interpolation by spline interpolationThereby constructing a replacement valueThe mode is as follows:
Weight coefficient: ， And is also provided with 。
6. The method for predicting the dynamic response time of the steering engine according to claim 1, wherein the method comprises the following steps:
the method comprises the steps of obtaining part of optimized data from a steering engine running state data set as sample data, constructing an initial model by a deep neural network, training the initial model, and obtaining a trained steering engine running prediction model; after several tests, constructing corresponding precision coefficient from the test data If the precision coefficientAnd (3) optimizing the constructed steering engine operation prediction model below the precision threshold.
7. The method for predicting the dynamic response time of the steering engine according to claim 6, wherein the method comprises the following steps:
Predicting the dynamic response time of the steering engine by using the optimized steering engine operation prediction model, and predicting the response time after a plurality of times of prediction Constructing a predictive responsivity comprising: response time toPerforming linear normalization processing according to the following modes:
Wherein, N, n is the number of predictions,For the response time at the ith prediction,Weight coefficient, which is a qualified standard value of response time:， And (2) and 。
8. The method for predicting the dynamic response time of the steering engine according to claim 7, wherein the method comprises the following steps:
If the obtained prediction responsivity The responsiveness threshold value is not exceeded, and the average value of the prediction data of a plurality of times is taken as a final prediction value; and if the time is exceeded, the workload, the working temperature and the dynamic response time on each time node are obtained one by one along the time axis, the workload, the working temperature and the dynamic response time are taken as independent variables, and the linear regression analysis is carried out by taking the independent variables as the independent variables to obtain a corresponding regression equation.
9. The method for predicting the dynamic response time of the steering engine according to claim 8, wherein the method comprises the following steps:
Acquiring regression coefficients corresponding to the independent variables from a regression equation, determining corresponding weight coefficients for the independent variables by using a hierarchical analysis method by taking the regression coefficients as influence factors, and then carrying out weighted average on a plurality of influence factors by combining the weight coefficients to acquire influence degree of environmental conditions on steering engine response; if the obtained influence exceeds the influence threshold, a first instruction is sent to the outside for optimization; if the first optimization instruction does not exceed the second optimization instruction, a second optimization instruction is sent to the outside.
10. The method for predicting the dynamic response time of the steering engine according to claim 9, wherein the method comprises the following steps:
after receiving a first optimization instruction, identifying current working condition data of a steering engine to obtain a plurality of condition optimization features; constructing a condition optimization model by an optimization algorithm, optimizing the condition optimization characteristics, and obtaining optimized steering engine working conditions;
after receiving the second optimization instruction, acquiring a current control algorithm of the steering engine, and identifying a plurality of preset parameters of the acquisition control algorithm; carrying out feature recognition on preset parameters to obtain corresponding optimized features;
selecting steering engine control algorithm optimization as a target word, and constructing a control algorithm optimization knowledge graph after deep search; matching an optimization scheme for the current control algorithm according to the correspondence between the optimization features and the optimization scheme of the control algorithm; and executing the optimization scheme to optimize a control algorithm of the steering engine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410520660.6A CN118092199B (en) | 2024-04-28 | 2024-04-28 | Prediction method for dynamic response time of steering engine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410520660.6A CN118092199B (en) | 2024-04-28 | 2024-04-28 | Prediction method for dynamic response time of steering engine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118092199A CN118092199A (en) | 2024-05-28 |
| CN118092199B true CN118092199B (en) | 2024-07-19 |
Family
ID=91165511
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410520660.6A Active CN118092199B (en) | 2024-04-28 | 2024-04-28 | Prediction method for dynamic response time of steering engine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118092199B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118411161A (en) * | 2024-07-03 | 2024-07-30 | 南通伟达电气成套设备有限公司 | Electromechanical equipment operation and maintenance method and system based on data analysis |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107844658A (en) * | 2017-11-16 | 2018-03-27 | 中国民航大学 | The steering wheel trend prediction method that Method Using Relevance Vector Machine merges with autoregressive moving average |
| CN112256026A (en) * | 2020-10-14 | 2021-01-22 | 中国船舶重工集团公司第七0七研究所九江分部 | Ship course model predictive control algorithm design method under multi-constraint condition |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115729200B (en) * | 2021-08-26 | 2024-05-31 | 哈尔滨工业大学 | Unmanned aerial vehicle steering engine fault detection model construction method and device, and unmanned aerial vehicle steering engine fault detection method and device |
| FR3140181A1 (en) * | 2022-09-27 | 2024-03-29 | Centre National De La Recherche Scientifique | Method for controlling a mechatronic system |
-
2024
- 2024-04-28 CN CN202410520660.6A patent/CN118092199B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107844658A (en) * | 2017-11-16 | 2018-03-27 | 中国民航大学 | The steering wheel trend prediction method that Method Using Relevance Vector Machine merges with autoregressive moving average |
| CN112256026A (en) * | 2020-10-14 | 2021-01-22 | 中国船舶重工集团公司第七0七研究所九江分部 | Ship course model predictive control algorithm design method under multi-constraint condition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118092199A (en) | 2024-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118092199B (en) | Prediction method for dynamic response time of steering engine | |
| Don et al. | Dynamic process fault detection and diagnosis based on a combined approach of hidden Markov and Bayesian network model | |
| Patange et al. | A machine learning approach for vibration-based multipoint tool insert health prediction on vertical machining centre (VMC) | |
| US10921759B2 (en) | Computer system and method for monitoring key performance indicators (KPIs) online using time series pattern model | |
| Tiddens et al. | The adoption of prognostic technologies in maintenance decision making: a multiple case study | |
| JPH06170696A (en) | System and method for using real time expert system for diagnosing tool life and predicted tool wear | |
| Alfarizi et al. | Optimized random forest model for remaining useful life prediction of experimental bearings | |
| CN118154174B (en) | Intelligent operation and maintenance cloud platform for industrial equipment | |
| CN116448419A (en) | Zero sample bearing fault diagnosis method based on depth model high-dimensional parameter multi-target efficient optimization | |
| CN112328703B (en) | Numerical control equipment health state diagnosis device and method based on incremental learning | |
| Nasser et al. | A hybrid of convolutional neural network and long short-term memory network approach to predictive maintenance | |
| CN112699597A (en) | Nuclear power starting water pump rolling bearing fault detection method and system | |
| CN114118982A (en) | Cloud computing and edge computing combined production control system | |
| Hu et al. | Health monitoring of automotive suspensions: a lstm network approach | |
| Li et al. | Fault forecasting of a machining center tool magazine based on health assessment | |
| CN116894211A (en) | System for generating human perceptible interpretive output, method and computer program for monitoring anomaly identification | |
| Liu | Cyber-physical system augmented prognostics and health management for fleet-based systems | |
| Li et al. | A novel approach for tool condition monitoring based on transfer learning of deep neural networks using time–frequency images | |
| US20240177541A1 (en) | Adaptive training of data driven analysis to characterize faults in physical systems | |
| Chakroun et al. | A predictive maintenance model for health assessment of an assembly robot based on machine learning in the context of smart plant | |
| CN117826606A (en) | Variable optimization method and control equipment of absorption stabilization system | |
| CN117743964A (en) | Method, system and medium for predicting residual service life of equipment based on graph data | |
| CN110110401B (en) | Safety tree model-based electric vehicle safety design optimization method | |
| Martins et al. | Predictive Maintenance of Mining Machines Applying Advanced Data Analysis | |
| CN118012479B (en) | Digital twin real-time updating method, medium and system based on pre-estimation mode |
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 |