CN109884893B - Multi-process variable dynamic time lag estimation method - Google Patents
Multi-process variable dynamic time lag estimation method Download PDFInfo
- Publication number
- CN109884893B CN109884893B CN201910152331.XA CN201910152331A CN109884893B CN 109884893 B CN109884893 B CN 109884893B CN 201910152331 A CN201910152331 A CN 201910152331A CN 109884893 B CN109884893 B CN 109884893B
- Authority
- CN
- China
- Prior art keywords
- time
- data
- lag parameter
- relative
- data set
- 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
- 238000000034 method Methods 0.000 title claims abstract description 69
- 238000012549 training Methods 0.000 claims abstract description 8
- 238000004519 manufacturing process Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 description 8
- 238000005259 measurement Methods 0.000 description 7
- 238000007637 random forest analysis Methods 0.000 description 7
- GRYLNZFGIOXLOG-UHFFFAOYSA-N Nitric acid Chemical compound O[N+]([O-])=O GRYLNZFGIOXLOG-UHFFFAOYSA-N 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 5
- 229910017604 nitric acid Inorganic materials 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000001364 causal effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a multi-process variable interval dynamic time lag estimation method, which comprises the following steps of firstly, dividing data in a database into input data and output data, wherein the input data is auxiliary variable, and the output data is main variable; then, a time-lag parameter set of the input data relative to the intermediate variable and a time-lag parameter set of the input data relative to the output data and the intermediate variable relative to the output data are obtained for a training data set in a database; and finally, taking the time-lag parameter set of the input data relative to the intermediate variable as input, and the time-lag parameter set of the input data relative to the output data and the time-lag parameter set of the intermediate variable relative to the output data as output to obtain a multi-process variable-variable dynamic time-lag parameter prediction model, solving the time-lag parameter of the input relative to the output on line, and substituting the time-lag parameter into the prediction model to obtain the input and output time-lag parameter in real time.
Description
Technical Field
The invention belongs to the technical field of automation control in the chemical industry, and particularly relates to a multi-process variable dynamic time lag estimation method.
Background
Most production process flows in factories are formed by cascading a plurality of working procedures, one or even a few hours are usually needed from raw material input to finished product output, and time lag information between each working procedure dynamically changes along with factors such as external environment, states of the previous working procedure and the next working procedure and is not invariable. The existing documents of process variable prediction methods using time-lag parameters are all methods based on fixed time-lag parameters, because the time-lag parameters between the front and rear processes in the actual plant are dynamic values changing along with the running state of the plant, but not fixed values, if the time-lag information between input and output is not considered, or the time-lag information is simply considered as a fixed value, and then the reconstructed data is used for training a soft measurement model, the causal relationship between the input and output is likely to change, so that the model training is not accurate enough, and the prediction performance of the soft measurement model is reduced.
Disclosure of Invention
The invention aims to provide a multi-process variable dynamic time lag estimation method, which solves the problems of inaccurate online soft measurement and reduced prediction performance in the prior art.
The technical scheme adopted by the invention is that the method for estimating the dynamic time lag among the multiple process variables is implemented according to the following steps:
and 4, solving a time-lag parameter of the input data relative to the intermediate variable in the actual production data set on line by using a Pearson correlation coefficient method, substituting the time-lag parameter into a trained time-lag parameter prediction model in real time, and predicting to obtain the time-lag parameter of the input data relative to the output data and the time-lag parameter of the intermediate variable relative to the output data.
The present invention is also characterized in that,
the method for solving the time-lag parameter in the step 2 is any one of a Pearson correlation coefficient method, a fuzzy curve method and a generalized quadratic correlation method.
The step 2 is implemented according to the following steps:
step 2.1, assuming that the database dataAll has n groups of data, defining the initial value of the timer time as 1;
step 2.2, setting the upper limit of the time lag parameter as delay _ max, and taking the data of the first time to the second time + m-1 from the database dataAll to define as a data set D _ time, wherein when m is greater than delay _ max and time is 1, taking the data of the first time to the second time from the database dataAll to store in the data set D _1, at this time, D _ time is D _1, when time is h, taking the data of the total m of the data of h to h + m-1 from the database dataAll to store in the data set D _ h, and D _ time is D _ h; d _ time (i, j) is the ith row and jth column data in the data set D _ time, where i is 1, 2, 3 · m, j is 1, 2, 3, and the 1, 2, 3 columns of data in the data set D _ time correspond to input data, intermediate variable, and output data, respectively;
step 2.3, extending the data set D _ time in step 2.2 to data sets D1, D2, D3: the data set D1 is input data and intermediate variables, the data set D2 is input data and output data, and the data set D3 is intermediate variables and output data;
step 2.4, solving the optimal time lag parameters of the data sets D1, D2 and D3 by using a Pearson correlation coefficient method and respectively storing the optimal time lag parameters into D _ delayIn (time), D _ delayOut1(time) and D _ delayOut2 (time);
step 2.5, updating the data set D _ time by making time +1, executing downward if time + m-1 is n, and returning to step 2.2 if time + m-1 is n, and repeating the execution;
and 2.6, obtaining an optimal time lag parameter set D _ delayIn of the input data relative to the intermediate variable, an optimal time lag parameter set D _ delayOut1 of the input data relative to the output data, and an optimal time lag parameter set D _ delayOut2 of the intermediate variable relative to the output data.
Delay _ max in step 2.2 is 30.
Step 2.4 is specifically carried out according to the following steps:
step 2.4.1, expand the data set D1 into delay _ max +1 group data set: when the delay time is w, w is a delay time value of the second column data of the data set D1 relative to the first column data, the group of data set is named as D1_ w, the first column data of D1_ w is row 1 to row m-w of the first column data of D1, and the second column data of D1_ w is row w +1 to row m of the second column data of D1;
step 2.4.2, a correlation coefficient value P (w) of the data set corresponding to each delay time value w is obtained by using a Pearson correlation coefficient method, the delay time corresponding to the maximum value in the obtained correlation coefficient value set P is taken as the optimal time-lag parameter of the expanded data set D1 of the obtained data set D _ time and is stored in D _ delayIn (time), and the optimal time-lag parameters of the expanded data sets D2 and D3 of the data set D _ time are obtained by the same method as D1 and are respectively stored in D _ delayOut1(time) and D _ delayOut2 (time).
The pearson correlation coefficient formula in step 2.4.2 is as follows:
wherein p (w) is the correlation coefficient value of the data set D1_ w with the obtained delay time w, N is the length of the data set D1_ w, x represents the 1 st column data in the corresponding data set D1_ w requiring the corresponding correlation coefficient value after D1 expansion, and y represents the 2 nd column data in the corresponding data set D1_ w requiring the corresponding correlation coefficient value after D1 expansion.
Step 4 is specifically implemented according to the following steps:
step 4.1, when input data and intermediate variable data reach m groups through factory online collection, establishing a data set D _ timeTest, wherein the D _ timeTest takes currently collected current 1 st to m groups of data, the D _ timeTest is expanded into a data set D1 according to step 2.3, and a correlation coefficient method is used for synchronizing step 2.4 to obtain an optimal time lag parameter D _ delayIn _ test (1) of the input variable relative to the output variable;
step 4.2, after that, each group of data is updated, D _ timeTest is updated once, assuming that the current u-th group of data is acquired, and u > -m, D _ timeTest takes the current u-m +1 to u-th groups of data acquired currently, and the optimal time lag parameter D _ delayIn _ test (u-m +1) of the input variable relative to the output variable at the current time is obtained in the synchronization step 2.3;
and 4.3, substituting the optimal time lag parameter of the input variable relative to the intermediate variable obtained on line into the time lag parameter prediction model established in the step 3 in real time, and predicting the time lag parameter of the input data relative to the output data and the time lag parameter of the intermediate variable relative to the output data on line.
The method has the advantages that according to the delay time relationship inside input data and the delay time relationship between the input data and the output data, a dynamic time-lag parameter prediction model which can obtain time-lag parameters between input and output only depending on the input data is trained, and then the input data and the required delay time between the input and the output are substituted into a soft measurement model to carry out online estimation on the output data. Compared with the traditional soft measurement model using the off-line trained constant value time-lag parameter, the output result of the soft measurement model using the dynamic time-lag parameter prediction model to predict the time-lag parameter on line is more reliable and accurate.
Drawings
FIG. 1 is a diagram illustrating the present invention of retrieving 40 sets of data from a database to form a data set D _ time;
FIG. 2 is an illustration of the present invention expanding a data set D _ time to a data set D1, D2, D3;
FIG. 3 is a diagram illustrating the calculation of the optimal time lag parameter of the data set D1 according to the present invention, taking the data set D1 as an example;
FIG. 4 is a diagram of the effect of prediction of the monitoring data of nitric acid preparation process of Shaandrum corporation by using a random forest model based on fixed time lag parameters according to the invention;
FIG. 5 is a diagram of the effect of prediction of the monitoring data of nitric acid production process of Shaandrum corporation by using a random forest model based on-line solving time lag parameters according to the invention;
fig. 6 is a flow chart of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
The invention discloses a method for estimating dynamic time lag among multiple process variables, which is implemented by the following steps as shown in a flow chart shown in figure 6:
the method for solving the time-lag parameter in the step 2 is any one of a Pearson correlation coefficient method, a fuzzy curve method and a generalized quadratic correlation method.
The step 2 is implemented according to the following steps:
step 2.1, assuming that the database dataAll has n groups of data, defining the initial value of the timer time as 1;
step 2.2, setting the upper limit of the time lag parameter as delay _ max, and taking the first to the mth group of data from the database dataAll to define as a data set D _ time, where m > delay _ max, as shown in fig. 1, when time is 1, taking the first to the mth group of data from the database dataAll to store in the data set D _1, at this time, D _ time is D _1, when time is h, taking the total of m group of data from h to h + m-1 from the database dataAll to store in the data set D _ h, D _ time is D _ h, and D _ time (i, j) is the jth column data in the ith row of the data set D _ time, where i is 1, 2, 3. m, j is 1, 2, 3, and the columns of the data set D _ time are respectively corresponding to input data, intermediate data, and output data;
step 2.3, extending the data set D _ time in step 2.2 into data sets D1, D2 and D3, wherein the extended form is shown in FIG. 2: the data set D1 is input data and intermediate variables, the data set D2 is input data and output data, and the data set D3 is intermediate variables and output data;
step 2.4, solving the optimal time lag parameters of the data sets D1, D2 and D3 by using a Pearson correlation coefficient method and respectively storing the optimal time lag parameters into D _ delayIn (time), D _ delayOut1(time) and D _ delayOut2 (time);
step 2.5, updating the data set D _ time by making time +1, executing downward if time + m-1 is n, and returning to step 2.2 if time + m-1 is n, and repeating the execution;
and 2.6, obtaining an optimal time lag parameter set D _ delayIn of the input data relative to the intermediate variable, an optimal time lag parameter set D _ delayOut1 of the input data relative to the output data, and an optimal time lag parameter set D _ delayOut2 of the intermediate variable relative to the output data.
Delay _ max in step 2.2 is 30.
Step 2.4 is specifically carried out according to the following steps:
step 2.4.1, expand the data set D1 into delay _ max +1 group data set: when the delay time is w, w is a delay time value of the second column data of the data set D1 relative to the first column data, the group of data set is named as D1_ w, the first column data of D1_ w is row 1 to row m-w of the first column data of D1, and the second column data of D1_ w is row w +1 to row m of the second column data of D1;
step 2.4.2, calculating a correlation coefficient value P (w) of the data set corresponding to each delay time value w by using a Pearson correlation coefficient method, taking the delay time corresponding to the maximum value in the calculated correlation coefficient value set P as the optimal time-lag parameter of the expanded data set D1 of the data set D _ time, and storing the optimal time-lag parameter into D _ delayIn (time), as shown in FIG. 3, the optimal time-lag parameter calculation methods of the data sets D2 and D3 after the data set D _ time is expanded are the same as D1, and respectively store the optimal time-lag parameter into D _ delayOut1(time),
D _ delayOut2 (time).
The pearson correlation coefficient formula in step 2.4.2 is as follows:
wherein p (w) is the correlation coefficient value of the data set D1_ w with the obtained delay time w, N is the length of the data set D1_ w, x represents the 1 st column data in the corresponding data set D1_ w requiring the corresponding correlation coefficient value after D1 expansion, and y represents the 2 nd column data in the corresponding data set D1_ w requiring the corresponding correlation coefficient value after D1 expansion.
And 3, taking time-lag parameter data sets of input data relative to an intermediate variable, input data relative to output data and the intermediate variable relative to the output data as training samples, randomly returning to take out one sample data, carrying out n1 times of sampling to generate n1 training sets, taking a time-lag parameter set of the input data relative to the intermediate variable as input, and taking a time-lag parameter set of the input data relative to the output data and a time-lag parameter set of the intermediate variable relative to the output data as output, and thus obtaining the multi-process variable-interval dynamic time-lag parameter prediction model.
The delay time prediction model can be a BP neural network model, a random forest model, a support vector regression model and the like. And selecting a proper model according to the prediction precision requirement of the user. Since the BP neural network model, the random forest model, the support vector regression model and the like are all known prediction models, detailed description is not performed on the model. The invention selects a random forest model in a time-lag parameter prediction model.
Step 4, solving a time-lag parameter of input data relative to an intermediate variable in an actual production data set on line by using a Pearson correlation coefficient method, substituting the time-lag parameter into a trained time-lag parameter prediction model in real time, predicting to obtain the time-lag parameter of the input data relative to output data and the time-lag parameter of the intermediate variable relative to the output data, and specifically implementing according to the following steps:
step 4.1, when input data and intermediate variable data reach m groups through factory online collection, establishing a data set D _ timeTest, wherein the D _ timeTest takes currently collected current 1 st to m groups of data, the D _ timeTest is expanded into a data set D1 according to step 2.3, and an optimal time lag parameter D _ delayIn _ test (1) of an input variable relative to an output variable is obtained by using a Pearson correlation coefficient method and a synchronization step 2.4;
step 4.2, after that, each group of data is updated, D _ timeTest is updated once, assuming that the current u-th group of data is acquired, and u > -m, D _ timeTest takes the current u-m +1 to u-th groups of data acquired currently, and the optimal time lag parameter D _ delayIn _ test (u-m +1) of the input variable relative to the output variable at the current time is obtained in the synchronization step 2.3;
and 4.3, substituting the optimal time lag parameter of the input variable relative to the intermediate variable obtained on line into the time lag parameter prediction model established in the step 3 in real time, and predicting the time lag parameter of the input data relative to the output data and the time lag parameter of the intermediate variable relative to the output data on line.
Taking monitoring data in the nitric acid production process of Shaandrum as an example, compared with a soft measurement model established by using a traditional time-lag parameter estimation method, the model prediction precision is greatly improved, and the prediction result is compared with the graphs shown in FIGS. 4 and 5, wherein FIG. 4 is an effect graph of the prediction of the monitoring data of the nitric acid production process of Shaandrum by using a random forest model with fixed time-lag parameters, FIG. 5 is an effect graph of the prediction of the monitoring data of the nitric acid production process of Shaandrum by using a random forest model based on-line time-lag parameters, namely an on-line time-lag parameter estimation method, and the effects are shown in the following table 1:
TABLE 1 comparison of results of the conventional Process and the Process herein
| Mean absolute error | Mean error | R2 | RMSE | |
| Conventional methods | 19.59％ | 4.615 | 0.08 | 6.7 |
| The patented method | 4.86％ | 1.137 | 0.90 | 2.13 |
As can be seen from Table 1, compared with the traditional method, the average relative error is reduced from 19.59% to 4.86%, the improvement effect is as high as 75.19%, the average error is reduced from 4.615 to 1.137, and the improvement effect is as high as 75.36%, namely the method disclosed by the patent has a good effect on the prediction of the key variables of the complex industrial process with time lag.
Claims (7)
1. A multi-process variable dynamic time lag estimation method is characterized by comprising the following steps:
step 1, dividing data in a database into input data and output data, wherein the input data is auxiliary variable, and the output data is main variable;
step 2, solving a time-lag parameter set of input data relative to an intermediate variable and a time-lag parameter set of the input data relative to output data and the intermediate variable relative to the output data for a training data set in a database;
step 3, taking a time-lag parameter set of the input data relative to the intermediate variable as input, and taking a time-lag parameter set of the input data relative to the output data and the intermediate variable relative to the output data as output, so as to obtain a multi-process variable dynamic time-lag parameter prediction model;
and 4, solving a time-lag parameter of the input data relative to the intermediate variable in the actual production data set on line by using a Pearson correlation coefficient method, substituting the time-lag parameter into a trained time-lag parameter prediction model in real time, and predicting to obtain the time-lag parameter of the input data relative to the output data and the time-lag parameter of the intermediate variable relative to the output data.
2. The method for estimating dynamic time lag between multi-process variables according to claim 1, wherein the method for solving the time lag parameter in step 2 is any one of Pearson's correlation coefficient method, fuzzy curve method and generalized quadratic correlation method.
3. The method as claimed in claim 2, wherein the step 2 is implemented by the following steps:
step 2.1, assuming that the database dataAll has n groups of data, defining the initial value of the timer time as 1;
step 2.2, setting the upper limit of the time lag parameter as delay _ max, and taking the data from the first time to the second time + m-1 from the database dataAll to define a data set D _ time, where m > delay _ max, and when time is 1, taking the data from the first time to the second time from the database dataAll to store in a data set D _1, where D _ time is D _1, and when time is h, taking the data from the database dataAll to h + m-1 for m groups of data to store in a data set D _ h, D _ time is D _ h, and D _ time (i, j) is the ith row j of the data set D _ time, where i is 1, 2, 3. m, j is 1, 2, 3, and columns 1, 2, 3 of the data set D _ time correspond to input data, intermediate variables, and output data, respectively;
step 2.3, extending the data set D _ time in step 2.2 to data sets D1, D2, D3: the data set D1 is input data and intermediate variables, the data set D2 is input data and output data, and the data set D3 is intermediate variables and output data;
step 2.4, solving the optimal time lag parameters of the data sets D1, D2 and D3 by using a Pearson correlation coefficient method and respectively storing the optimal time lag parameters into D _ delayIn (time), D _ delayOut1(time) and D _ delayOut2 (time);
step 2.5, updating the data set D _ time by making time +1, executing downward if time + m-1 is n, and returning to step 2.2 if time + m-1 is n, and repeating the execution;
and 2.6, obtaining an optimal time lag parameter set D _ delayIn of the input data relative to the intermediate variable, an optimal time lag parameter set D _ delayOut1 of the input data relative to the output data, and an optimal time lag parameter set D _ delayOut2 of the intermediate variable relative to the output data.
4. A method according to claim 3, characterized in that in step 2.2, delay _ max is 30 and m is 40.
5. A method according to claim 3, wherein said step 2.4 is performed according to the following steps:
step 2.4.1, expand the data set D1 into delay _ max +1 group data set: when the delay time is w, w is a delay time value of the second column data of the data set D1 relative to the first column data, the group of data set is named as D1_ w, the first column data of D1_ w is row 1 to row m-w of the first column data of D1, and the second column data of D1_ w is row w +1 to row m of the second column data of D1;
step 2.4.2, a correlation coefficient value P (w) of the data set corresponding to each delay time value w is obtained by using a Pearson correlation coefficient method, the delay time corresponding to the maximum value in the obtained correlation coefficient value set P is taken as the optimal time-lag parameter of the expanded data set D1 of the obtained data set D _ time and is stored in D _ delayIn (time), and the optimal time-lag parameters of the expanded data sets D2 and D3 of the data set D _ time are obtained by the same method as D1 and are respectively stored in D _ delayOut1(time) and D _ delayOut2 (time).
6. The method of claim 5, wherein the Pearson correlation coefficient in step 2.4.2 is expressed as follows:
wherein p (w) is the correlation coefficient value of the data set D1_ w with the obtained delay time w, N is the length of the data set D1_ w, x represents the 1 st column data in the corresponding data set D1_ w requiring the corresponding correlation coefficient value after D1 expansion, and y represents the 2 nd column data in the corresponding data set D1_ w requiring the corresponding correlation coefficient value after D1 expansion.
7. The method as claimed in claim 6, wherein the step 4 is implemented by the following steps:
step 4.1, when input data and intermediate variable data reach m groups through factory online collection, establishing a data set D _ timeTest, wherein the D _ timeTest takes currently collected current 1 st to m groups of data, the D _ timeTest is expanded into a data set D1 according to step 2.3, and a correlation coefficient method is used for synchronizing step 2.4 to obtain an optimal time lag parameter D _ delayIn _ test (1) of the input variable relative to the output variable;
step 4.2, after that, each group of data is updated, D _ timeTest is updated once, assuming that the current u-th group of data is acquired, and u > -m, D _ timeTest takes the current u-m +1 to u-th groups of data acquired currently, and the optimal time lag parameter D _ delayIn _ test (u-m +1) of the input variable relative to the output variable at the current time is obtained in the synchronization step 2.3;
and 4.3, substituting the optimal time lag parameter of the input variable relative to the intermediate variable obtained on line into the time lag parameter prediction model established in the step 3 in real time, and predicting the time lag parameter of the input data relative to the output data and the time lag parameter of the intermediate variable relative to the output data on line.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910152331.XA CN109884893B (en) | 2019-02-28 | 2019-02-28 | Multi-process variable dynamic time lag estimation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910152331.XA CN109884893B (en) | 2019-02-28 | 2019-02-28 | Multi-process variable dynamic time lag estimation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109884893A CN109884893A (en) | 2019-06-14 |
| CN109884893B true CN109884893B (en) | 2021-09-10 |
Family
ID=66930040
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910152331.XA Active CN109884893B (en) | 2019-02-28 | 2019-02-28 | Multi-process variable dynamic time lag estimation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109884893B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112295255B (en) * | 2020-10-24 | 2021-11-02 | 四川泸天化创新研究院有限公司 | Intelligent control system and control method for methanol rectification device |
Family Cites Families (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4612621A (en) * | 1983-03-17 | 1986-09-16 | The Babcock & Wilcox Company | Distributed system for optimizing the performance of a plurality of multi-stage steam turbines using function blocks |
| US5144549A (en) * | 1990-06-29 | 1992-09-01 | Massachusetts Institute Of Technology | Time delay controlled processes |
| JPH07114580A (en) * | 1993-10-18 | 1995-05-02 | Fujitsu Ltd | Delay time analysis system for logical device |
| US9110452B2 (en) * | 2011-09-19 | 2015-08-18 | Fisher-Rosemount Systems, Inc. | Inferential process modeling, quality prediction and fault detection using multi-stage data segregation |
| US8958675B2 (en) * | 2012-07-12 | 2015-02-17 | Corning Incorporated | Multimode optical fiber systems with adjustable chromatic modal dispersion compensation |
| CN103279030B (en) * | 2013-03-07 | 2016-05-18 | 清华大学 | Dynamic soft measuring modeling method and device based on Bayesian frame |
| CN104503229A (en) * | 2014-11-24 | 2015-04-08 | 北京邮电大学 | Wave integral bilateral teleoperation control method based on LS-SVM (least square support vector machine) delay predication |
| CN105205224B (en) * | 2015-08-28 | 2018-10-30 | 江南大学 | Time difference Gaussian process based on fuzzy curve analysis returns soft-measuring modeling method |
| CN105429720B (en) * | 2015-11-25 | 2017-09-29 | 桂林航天工业学院 | The Time Delay Estimation Based reconstructed based on EMD |
| CN105846939B (en) * | 2016-03-24 | 2018-01-12 | 成都博思微科技有限公司 | A kind of accurate System and method for for keeping multimode synchronous |
| US9941895B2 (en) * | 2016-08-01 | 2018-04-10 | Kopin Corporation | Time delay in digitally oversampled sensor systems, apparatuses, and methods |
| CN107273633B (en) * | 2017-06-29 | 2018-03-27 | 中南大学 | Varying delay method of estimation and flow time lag method of estimation is hydrocracked between multiple operation |
| CN107942667B (en) * | 2017-11-29 | 2020-08-07 | 辽宁石油化工大学 | Injection molding process hybrid 2D tracking control method based on time-varying time lag and interference |
| CN108647808B (en) * | 2018-04-11 | 2022-03-29 | 济南大学 | Production parameter optimization prediction method, device, equipment and storage medium |
| CN108614533A (en) * | 2018-05-28 | 2018-10-02 | 江南大学 | A kind of neural network modeling approach estimated based on NARX models and time lag |
-
2019
- 2019-02-28 CN CN201910152331.XA patent/CN109884893B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109884893A (en) | 2019-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109992921B (en) | On-line soft measurement method and system for thermal efficiency of boiler of coal-fired power plant | |
| CN108764568B (en) | Data prediction model tuning method and device based on LSTM network | |
| CN108304679A (en) | A kind of adaptive reliability analysis method | |
| Varghese et al. | Statistical tolerance analysis using FRPDF and numerical convolution | |
| CN108564326A (en) | Prediction technique and device, computer-readable medium, the logistics system of order | |
| CN108667069B (en) | Short-term wind power prediction method based on partial least squares regression | |
| CN106099932B (en) | Day-ahead planning power flow analysis method considering uncertainty time-space correlation | |
| CN107730097B (en) | Bus load prediction method and device and computing equipment | |
| CN105652845A (en) | Fermentation process fault monitoring method based on just-in-time learning local model | |
| CN104050547A (en) | Non-linear optimization decision-making method of planning schemes for oilfield development | |
| CN109884893B (en) | Multi-process variable dynamic time lag estimation method | |
| CN104330972A (en) | Comprehensive prediction iterative learning control method based on model adaptation | |
| CN109542070B (en) | Dynamic process monitoring method based on dual-target optimization algorithm | |
| CN111612242A (en) | Motor state parameter prediction method based on LSTM deep learning model | |
| CN108198268A (en) | A kind of production equipment data scaling method | |
| CN112612781A (en) | Data correction method, device, equipment and medium | |
| CN108256274B (en) | Power system state identification method based on search attractor error algorithm | |
| Havinga et al. | Estimating product-to-product variations in metal forming using force measurements | |
| JP2013137774A (en) | Method for predicting state of process and system for predicting state of process | |
| CN117473833A (en) | Method and device for predicting rolling force of rod and wire based on finite element and neural network | |
| Muqtadir et al. | The Implementation of Grey Forecasting Model for Forecast Result’s Food Crop Agricultural | |
| CN108134687B (en) | Gray model local area network peak flow prediction method based on Markov chain | |
| CN110773580A (en) | Method and system for calculating rolling force based on XGboost algorithm | |
| CN113783715B (en) | Opportunistic network topology prediction method adopting causal convolutional neural network | |
| CN107967395A (en) | A kind of nonlinear time_varying system Fast Identification Method based on the expansion of beta wavelet basis functions |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240307 Address after: 1002, Building A, Zhiyun Industrial Park, No. 13 Huaxing Road, Henglang Community, Dalang Street, Longhua District, Shenzhen City, Guangdong Province, 518110 Patentee after: Shenzhen Wanzhida Technology Co.,Ltd. Country or region after: China Address before: 710048 No. 5 Jinhua South Road, Shaanxi, Xi'an Patentee before: XI'AN University OF TECHNOLOGY Country or region before: China |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240415 Address after: Room 216, 2nd Floor, Unit B, Building 4, Chang'an University Science and Technology Park, South Section 4555 Guangyuntan Avenue, Chanba Ecological Zone, Xi'an City, Shaanxi Province, 710000 Patentee after: Zhixiao Technology (Xi'an) Co.,Ltd. Country or region after: China Address before: 1002, Building A, Zhiyun Industrial Park, No. 13 Huaxing Road, Henglang Community, Dalang Street, Longhua District, Shenzhen City, Guangdong Province, 518110 Patentee before: Shenzhen Wanzhida Technology Co.,Ltd. Country or region before: China |
|
| TR01 | Transfer of patent right |