CN111832171A - Railway signal relay performance state prediction method based on mathematical model - Google Patents
Railway signal relay performance state prediction method based on mathematical model Download PDFInfo
- Publication number
- CN111832171A CN111832171A CN202010670335.XA CN202010670335A CN111832171A CN 111832171 A CN111832171 A CN 111832171A CN 202010670335 A CN202010670335 A CN 202010670335A CN 111832171 A CN111832171 A CN 111832171A
- Authority
- CN
- China
- Prior art keywords
- model
- prediction
- unit
- prediction unit
- data
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000013178 mathematical model Methods 0.000 title claims abstract description 13
- 230000008569 process Effects 0.000 claims abstract description 19
- 238000005457 optimization Methods 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000005259 measurement Methods 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 230000003313 weakening effect Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims 1
- 230000006870 function Effects 0.000 description 6
- 238000012360 testing method Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 4
- 230000015556 catabolic process Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000035882 stress Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006353 environmental stress Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/10—Numerical modelling
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)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a railway signal relay performance state prediction method based on a mathematical model, which comprises the following steps: establishing a prediction unit model by referring to relevant experimental data of the railway signal relay at constant temperature and carrying out optimization processing; repeatedly utilizing the prediction unit model to carry out iterative operation to obtain a unit chain model; arranging a single time sequence formed by related parameter data of the railway signal relay at a constant temperature; establishing a sliding window prediction model, substituting a single time sequence into a unit chain model, comparing the final output value with the measured value to calculate an error, and determining the maximum allowable error; and carrying out logic judgment, updating the input data set, discarding the farthest data, and continuing the processes of prediction, repeated updating, adjustment and prediction until the total output quantity requirement is met. According to the method, the unit model is established, the unit parameters are determined, the unit models are connected end to obtain the unit chain model, so that the conventional prediction of the performance parameter sequence is realized, the prediction result is corrected in real time, and the prediction precision is improved.
Description
Technical Field
The invention relates to a relay performance state prediction technology, in particular to a railway signal relay performance state prediction method based on a mathematical model.
Background
The normal operation of the railway signal relay has important significance for ensuring the safe and reliable operation of railway transportation, so that the performance state of the relay is predicted and judged to be particularly important. Through a constant temperature stress life test, various performance parameter data which gradually change along with the increase of the operation times can be measured and recorded at the same time, and many scholars at home and abroad study the performance state or the operation life of the railway signal relay by analyzing the performance parameter data of the relay. Since the railway signal relay generally has a long service life, it is difficult to obtain data of failure, which brings great difficulty to the analysis of the performance state of the railway signal relay based on the performance parameter data.
There are some conventional methods for performing predictive analysis and reliability evaluation on relay performance parameters by using mathematical theory, and fig. 1 shows a prior art for studying the degradation of relay performance.
Fig. 2 is a curve fitting model. The curve fitting is performed by fitting a specific performance parameter sequence, firstly judging fitting modes such as polynomial fitting, power function fitting and logarithmic function fitting, determining fitting coefficients by solving the minimum value of fitting errors to obtain a fitting curve, and extending the fitting curve to obtain a prediction result.
Fig. 3 is a neural network model. Firstly, determining a network structure, performing learning training on sample data by adjusting the number of input nodes, the number of hidden layers, a node function and the number of output nodes, and continuously changing the weight and the threshold of the node function to enable the network output to be continuously close to the expected output, thereby obtaining an optimal network model. After the model is determined, the non-sample data can be predicted by combining the homogeneous data to be predicted and the trained network model.
The similar method only simply substitutes data into a model in the prediction process of the parameter sequence to obtain a final result, and the prediction result only depends on the method and cannot obtain higher prediction precision.
Therefore, how to improve the prediction accuracy of the performance parameter sequence becomes one of the problems to be solved urgently by those skilled in the art.
Disclosure of Invention
Aiming at the defects that the prediction of the relay performance state cannot obtain higher prediction precision and the like in the prior art, the invention aims to provide the railway signal relay performance state prediction method based on the mathematical model, which can correct the prediction result in real time and improve the prediction precision.
In order to solve the technical problems, the invention adopts the technical scheme that:
the invention provides a railway signal relay performance state prediction method based on a mathematical model, which comprises the following steps:
s1) establishing a prediction unit model by referring to the relevant experimental data of the railway signal relay at constant temperature and carrying out optimization processing to obtain an optimized prediction unit model;
s2) repeatedly using the optimized prediction unit model to perform iterative operation, and performing optimization processing on the result of each iterative operation to obtain a unit chain model;
s3) arranging a single time sequence formed by related parameter data of the railway signal relay at a constant temperature;
s4), establishing a sliding window prediction model, substituting data formed by a single time sequence into a unit chain model as an input data set, comparing the final output value with an actual measurement value to calculate an error, and determining a maximum allowable error;
s5), then updating the input data set, abandoning the farthest data, continuing the processes of prediction, repeated updating, adjustment and prediction until the requirement of the output total amount is met, wherein the input data set is updated in a sliding way in the prediction process, and the total amount is kept unchanged.
The step S1) of building a prediction unit model and performing optimization processing includes:
s101) weight ω of initialization unitσ、ωi、ωc、ωoAnd corresponding intercept dσ、di、dc、do;
S102) combining the current input data x of the prediction unit modeltAnd last output data yt-1Calculating a rejection rate ft=σ(ωσ[ht-1,xt]+dσ) Wherein ω isσAnd dσRespectively weight and intercept, sigma is a nonlinear function, h is a transfer coefficient, t is a unit series, xtInputting data;
s103) obtaining the current weakening coefficient i of the prediction unit modelt=σ(ωi[ht-1,xt]+di) Predicting current state characteristics of a cell modelAn intermediate amount of Ct′=tanh(ωc[yt-1,xt]+dc) Wherein ω iscIs a weight, yt-1For the output of the previous stage, dcIs the intercept;
s104) abandoning and reserving the input data through the unit model, and predicting the last state characteristic C of the unit modelt-1Obtaining a prediction Unit model Current feature CtThe calculation formula is as follows: ct=ftCt-1+itCt′;ftTo discard rate, itIs a characteristic coefficient, Ct' is the characteristic intermediate quantity of the current state of the prediction unit model;
s105) attenuation coefficient o of output gatet=σ(ωo[ht-1,xt]+do) Finally, the output y of the prediction unit model is obtainedt=ottanh(Ct);doIs intercept, ωoAre weights.
The step of obtaining the cell chain model in step S2) is:
repeatedly using the prediction unit model to calculate to obtain a group of data, and outputting data y of any prediction unit modeltInput data x as next prediction unit modelt+1And simultaneously recalculating the rejection rate of the next prediction unit model, and thus, obtaining a unit chain model by connecting the head and the tail of data obtained by calculating the prediction unit model for multiple times.
Step S4), the step of establishing the sliding window prediction model is:
s401) determining the maximum allowable error E of the unit chain model according to the input total t and the output total n of the unit chain model;
s402) converting the input data set X ═ X1,x2,…,xtSubstituting the model into the unit chain model for calculation;
s403) outputting the last time unit chain modelAnd measured value yt+1Making a comparison calculation to determine the error
S404) carrying out logic judgment, ifDiscarding the farthest data Y of the input data set0And introducing the latest measured data yt+1To adjust the input data set to continue the next prediction;
otherwise ifAdjusting the weight and intercept in the model parameters until the output value meets the requirement;
s405) updating an input data set, namely introducing the latest measured data, discarding the farthest data, and continuing to predict;
s406) repeating the updating, adjusting and predicting processes until the total output quantity requirement is met, wherein the input data set is updated in a sliding mode in the predicting process, and the total quantity is kept unchanged.
The invention has the following beneficial effects and advantages:
1. the method comprises the steps of establishing a unit model, determining unit parameters, and obtaining a unit chain model by connecting the unit models end to realize conventional prediction of a performance parameter sequence.
2. According to the invention, the input data set and the unit model are updated at any time by establishing the sliding window prediction model, so that the prediction result is corrected in real time, and the prediction precision is improved.
Drawings
FIG. 1 is a schematic diagram of a prior art structure for predicting the performance status of a railway signal relay;
FIG. 2 is a graph fitted model structure;
FIG. 3 is a predictive model of a neural network;
FIG. 4 is a flow chart of a mathematical model-based method for predicting the performance state of a railway signal relay according to the present invention;
FIG. 5 is a flow chart of the sliding window model of the present invention.
Detailed Description
The invention is further elucidated with reference to the accompanying drawings.
As shown in fig. 4, the invention provides a method for predicting the performance state of a railway signal relay based on a mathematical model, which comprises the following steps:
s1) establishing a prediction unit model by referring to the relevant experimental data of the railway signal relay at constant temperature and carrying out optimization processing to obtain an optimized prediction unit model;
s2) repeatedly using the optimized prediction unit model to perform iterative operation, and performing optimization processing on the result of each iterative operation to obtain a unit chain model;
s3) arranging a single time sequence formed by related parameter data of the railway signal relay at a constant temperature;
s4), establishing a sliding window prediction model, substituting data formed by a single time sequence into a unit chain model as an input data set, comparing the final output value with an actual measurement value to calculate an error, and determining a maximum allowable error;
s5), then updating the input data set, abandoning the farthest data, continuing the processes of prediction, repeated updating, adjustment and prediction until the requirement of the output total amount is met, wherein the input data set is updated in a sliding way in the prediction process, and the total amount is kept unchanged.
The step S1) of establishing the prediction unit model and performing the optimization process includes:
s101) weight ω of initialization unitσ、ωi、ωc、ωoAnd corresponding intercept dσ、di、dc、do;
S102) combining the current input data x of the prediction unit modeltAnd last output data yt-1Calculating a rejection rate ft=σ(ωσ[ht-1,xt]+dσ) Wherein ω isσAnd dσRespectively weight and intercept, sigma is a nonlinear function, h is a transfer coefficient, t is a unit series, xtInputting data;
s103) obtaining the current weakening system of the prediction unit modelNumber it=σ(ωi[ht-1,xt]+di) And C is the characteristic intermediate quantity of the current state of the prediction unit modelt′=tanh(ωc[yt-1,xt]+dc) Wherein ω iscIs a weight, yt-1For the output of the previous stage, dcIs the intercept;
s104) abandoning and reserving the input data through the unit model, and predicting the last state characteristic C of the unit modelt-1Obtaining a prediction Unit model Current feature CtThe calculation formula is as follows: ct=ftCt-1+itCt′;ftTo discard rate, itIs a characteristic coefficient, Ct' is the characteristic intermediate quantity of the current state of the prediction unit model;
s105) attenuation coefficient o of output gatet=σ(ωo[ht-1,xt]+do) Finally, the output y of the prediction unit model is obtainedt=ottanh(Ct);doIs intercept, ωoAre weights.
The step of obtaining the cell chain model in step S2) is:
repeatedly using the prediction unit model to calculate to obtain a group of data, and outputting data y of any prediction unit modeltInput data x as next prediction unit modelt+1And simultaneously recalculating the rejection rate of the next prediction unit model, and thus, repeating the process for a plurality of times [ the data obtained by calculating the prediction unit model are connected end to obtain the unit chain model.
Step S4), the step of establishing the sliding window prediction model is shown in fig. 5, and specifically includes:
s401) determining the maximum allowable error E of the unit chain model according to the input total t and the output total n of the unit chain model;
s402) converting the input data set X ═ X1,x2,…,xtSubstituting the model into the unit chain model for calculation;
s403) outputting the last time unit chain modelAnd measured value yt+1Making a comparison calculation to determine the error
S404) carrying out logic judgment, ifDiscarding the farthest data Y of the input data set0And introducing the latest measured data yt+1To adjust the input data set to continue the next prediction;
otherwise ifAdjusting the weight and intercept in the model parameters until the output value meets the requirement;
s405) updating an input data set, namely introducing the latest measured data, discarding the farthest data, and continuing to predict;
s406) repeating the updating, adjusting and predicting processes until the total output quantity requirement is met, wherein the input data set is updated in a sliding mode in the predicting process, and the total quantity is kept unchanged.
The method of the invention performs operation test on the relay through a constant temperature stress test, and measures and records various performance parameters of the relay in the test process, including time parameters, electrical parameters, mechanical parameters and the like. The change of the performance parameters can reflect the degradation condition of the relay, and the service life of the relay can be obtained by predicting the performance parameters, so that the problem of low prediction precision in the prior art is solved. The method of the invention has the advantages of unlimited environmental temperature, unlimited types of performance parameters and unlimited operation times of the relay, namely the length of a single time sequence.
In this embodiment, an operation test is performed on a relay by taking 40 ℃ as an environmental stress of a constant temperature life test, and a railway signal relay performance state prediction method based on a mathematical model includes the following steps:
the performance parameter data of a railway signal relay (such as a JWXC-1700 type relay) sample at constant 40 ℃ is collated, and an overtravel time parameter is selected for analysis;
establishing a prediction unit model: initializing each weight and intercept parameter of a prediction unit model, and constructing a prediction unit model structure;
data obtained by calculating through a prediction unit model are connected end to end for multiple times to obtain a unit chain model;
determining a sliding window prediction model, wherein the sliding window prediction steps are briefly described as follows:
determining an input total amount t and an output total amount n;
determining a maximum allowable error E;
converting an input data set X to X1,x2,…,xtSubstituting the unit chain model and calculating;
will be the output of the last unitAnd measured value yt+1Making a comparison calculation to determine the error
Logic discrimination: if it isDirectly updating the input data set in a sliding manner, otherwise, adjusting the model parameters until the output value meets the requirement, and then updating the input data set in a sliding manner;
and repeating the logic judging step until the predicting process is finished.
TABLE 1 time series results from sliding window model prediction
The above results show that: the invention can effectively predict the performance state of the railway signal relay. Compared with traditional methods such as curve fitting models, neural network models and prediction analysis of relay performance parameters by using mathematical theories, the prediction method is more scientific, faster, more accurate and more reliable through computer prediction.
The above description is only an 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 performed by the present specification and drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (4)
1. A railway signal relay performance state prediction method based on a mathematical model is characterized by comprising the following steps:
s1) establishing a prediction unit model by referring to the relevant experimental data of the railway signal relay at constant temperature and carrying out optimization processing to obtain an optimized prediction unit model;
s2) repeatedly using the optimized prediction unit model to perform iterative operation, and performing optimization processing on the result of each iterative operation to obtain a unit chain model;
s3) arranging a single time sequence formed by related parameter data of the railway signal relay at a constant temperature;
s4), establishing a sliding window prediction model, substituting data formed by a single time sequence into a unit chain model as an input data set, comparing the final output value with an actual measurement value to calculate an error, and determining a maximum allowable error;
s5), then updating the input data set, abandoning the farthest data, continuing the processes of prediction, repeated updating, adjustment and prediction until the requirement of the output total amount is met, wherein the input data set is updated in a sliding way in the prediction process, and the total amount is kept unchanged.
2. The method for predicting the performance state of the railway signal relay based on the mathematical model as claimed in claim 1, wherein the step of building a prediction unit model and performing optimization processing in step S1) comprises the steps of:
s101) weight ω of initialization unitσ、ωi、ωc、ωoAnd corresponding intercept dσ、di、dc、do;
S102) combining the current input data x of the prediction unit modeltAnd last output data yt-1Calculating a rejection rate ft=σ(ωσ[ht-1,xt]+dσ) Wherein ω isσAnd doRespectively weight and intercept, sigma is a nonlinear function, h is a transfer coefficient, t is a unit series, xtInputting data;
s103) obtaining the current weakening coefficient i of the prediction unit modelt=σ(ωi[ht-1,xt]+di) And C is the characteristic intermediate quantity of the current state of the prediction unit modelt′=tanh(ωc[yt-1,xt]+dc) Wherein ω iscIs a weight, yt-1For the output of the previous stage, dcIs the intercept;
s104) abandoning and reserving the input data through the unit model, and predicting the last state characteristic C of the unit modelt-1Obtaining a prediction Unit model Current feature CtThe calculation formula is as follows: ct=ftCt-1+itCt′;ftTo discard rate, itIs a characteristic coefficient, Ct' is the characteristic intermediate quantity of the current state of the prediction unit model;
s105) attenuation coefficient o of output gatet=σ(ωo[ht-1,xt]+do) Finally, the output y of the prediction unit model is obtainedt=ottanh(Ct);doIs intercept, ωoAre weights.
3. The method for predicting the performance state of the railway signal relay based on the mathematical model according to claim 1, wherein the step of obtaining the cell chain model in the step S2) is:
repeatedly using the prediction unit model to calculate to obtain a group of data, and outputting data y of any prediction unit modeltInput data x as next prediction unit modelt+1While recalculating the rejection rate of the next prediction unit model, whichAnd sampling data obtained by calculating the prediction unit model for multiple times, and connecting the data end to obtain a unit chain model.
4. The method for predicting the performance state of the railway signal relay based on the mathematical model according to claim 1, wherein the step of establishing the sliding window prediction model in the step S4) is as follows:
s401) determining the maximum allowable error E of the unit chain model according to the input total t and the output total n of the unit chain model;
s402) converting the input data set X ═ X1,x2,…,xtSubstituting the model into the unit chain model for calculation;
s403) outputting the last time unit chain modelAnd measured value yt+1Making a comparison calculation to determine the error
S404) carrying out logic judgment, ifDiscarding the farthest data Y of the input data set0And introducing the latest measured data yt-1To adjust the input data set to continue the next prediction;
otherwise ifAdjusting the weight and intercept in the model parameters until the output value meets the requirement;
s405) updating an input data set, namely introducing the latest measured data, discarding the farthest data, and continuing to predict;
s406) repeating the updating, adjusting and predicting processes until the total output quantity requirement is met, wherein the input data set is updated in a sliding mode in the predicting process, and the total quantity is kept unchanged.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010670335.XA CN111832171A (en) | 2020-07-13 | 2020-07-13 | Railway signal relay performance state prediction method based on mathematical model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010670335.XA CN111832171A (en) | 2020-07-13 | 2020-07-13 | Railway signal relay performance state prediction method based on mathematical model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111832171A true CN111832171A (en) | 2020-10-27 |
Family
ID=72922737
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010670335.XA Pending CN111832171A (en) | 2020-07-13 | 2020-07-13 | Railway signal relay performance state prediction method based on mathematical model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111832171A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130278269A1 (en) * | 2010-10-05 | 2013-10-24 | Samsung Sdi Co., Ltd | Method for Predicting the Usability of a Relay or a Contactor |
CN109508818A (en) * | 2018-10-23 | 2019-03-22 | 保定正德电力技术有限公司 | A kind of online NOx prediction technique based on LSSVM |
CN109934130A (en) * | 2019-02-28 | 2019-06-25 | 中国空间技术研究院 | The in-orbit real-time fault diagnosis method of satellite failure and system based on deep learning |
CN110334881A (en) * | 2019-07-17 | 2019-10-15 | 深圳大学 | A kind of Financial Time Series Forecasting method based on length memory network and depth data cleaning, device and server |
-
2020
- 2020-07-13 CN CN202010670335.XA patent/CN111832171A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130278269A1 (en) * | 2010-10-05 | 2013-10-24 | Samsung Sdi Co., Ltd | Method for Predicting the Usability of a Relay or a Contactor |
CN109508818A (en) * | 2018-10-23 | 2019-03-22 | 保定正德电力技术有限公司 | A kind of online NOx prediction technique based on LSSVM |
CN109934130A (en) * | 2019-02-28 | 2019-06-25 | 中国空间技术研究院 | The in-orbit real-time fault diagnosis method of satellite failure and system based on deep learning |
CN110334881A (en) * | 2019-07-17 | 2019-10-15 | 深圳大学 | A kind of Financial Time Series Forecasting method based on length memory network and depth data cleaning, device and server |
Non-Patent Citations (1)
Title |
---|
李其干: ""继电器在线参数检测系统及质量预测方法研究"", 《中国优秀硕士学位论文全文数据库(工程科技II辑)》, no. 2019, pages 67 - 79 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109816095B (en) | Network flow prediction method based on improved gated cyclic neural network | |
CN112903758B (en) | Intelligent temperature and humidity self-calibration method for overlapped blocks of gas sensor | |
CN110824914B (en) | Intelligent wastewater treatment monitoring method based on PCA-LSTM network | |
CN110083125B (en) | Machine tool thermal error modeling method based on deep learning | |
CN112926265A (en) | Atmospheric porous probe measurement calibration method based on genetic algorithm optimization neural network | |
CN114386686B (en) | Improved LSTM-based watershed water quality short-term prediction method | |
CN115308558A (en) | Method and device for predicting service life of CMOS (complementary Metal oxide semiconductor) device, electronic equipment and medium | |
CN116930609A (en) | Electric energy metering error analysis method based on ResNet-LSTM model | |
CN115982141A (en) | Characteristic optimization method for time series data prediction | |
CN114611803A (en) | Switch device service life prediction method based on degradation characteristics | |
CN116842853B (en) | Missile aerodynamic characteristic prediction model construction method for uncertainty quantization | |
CN112905436B (en) | Quality evaluation prediction method for complex software | |
CN113159395A (en) | Deep learning-based sewage treatment plant water inflow prediction method and system | |
CN111832171A (en) | Railway signal relay performance state prediction method based on mathematical model | |
CN112016241A (en) | Soft measurement modeling method based on mLASSO-MLP model | |
CN110648023A (en) | Method for establishing data prediction model based on quadratic exponential smoothing improved GM (1,1) | |
CN114924489A (en) | Model autonomous learning method suitable for process industry prediction control | |
CN115238874A (en) | Quantization factor searching method and device, computer equipment and storage medium | |
CN110321650B (en) | Structural reliability analysis method based on novel test design and weight response surface | |
CN114970674A (en) | Time sequence data concept drift adaptation method based on relevance alignment | |
CN110889190B (en) | Performance degradation modeling data volume optimization method facing prediction precision requirement | |
CN113408183A (en) | Vehicle base short-term composite prediction method based on prediction model | |
CN113255207A (en) | Iterative multi-output-Markov chain-based multi-step prediction method for argon fraction variable of air separation system | |
CN109284320A (en) | Automatic returning diagnostic method in big data platform | |
CN113836619B (en) | Method for fitting arbitrary creep coefficient curve |
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 |