Event-Triggered Relearning Modeling Method for Stochastic System with Non-Stationary Variable Operating Conditions
This study presents a novel event-triggered relearning framework for neural network modeling, designed to improve prediction precision in dynamic stochastic complex industrial systems under non-stationary and variable conditions. Firstly, a sliding window algorithm combined with entropy is applied t...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/12/5/667 |
| _version_ | 1797264152254742528 |
|---|---|
| author | Jiyan Liu Yong Zhang Yuyang Zhou Jing Chen |
| author_facet | Jiyan Liu Yong Zhang Yuyang Zhou Jing Chen |
| author_sort | Jiyan Liu |
| collection | DOAJ |
| description | This study presents a novel event-triggered relearning framework for neural network modeling, designed to improve prediction precision in dynamic stochastic complex industrial systems under non-stationary and variable conditions. Firstly, a sliding window algorithm combined with entropy is applied to divide the input and output datasets across different operational conditions, establishing clear data boundaries. Following this, the prediction errors derived from the neural network under different operational states are harnessed to define a set of event-triggered relearning criteria. Once these conditions are triggered, the relevant dataset is used to recalibrate the model to the specific operational condition and predict the data under this operating condition. When the predicted data fall within the training input range of a pre-trained model, we switch to that model for immediate prediction. Compared with the conventional BP neural network model and random vector functional-link network, the proposed model can produce a better estimation accuracy and reduce computation costs. Finally, the effectiveness of our proposed method is validated through numerical simulation tests using nonlinear Hammerstein models with Gaussian noise, reflecting complex stochastic industrial processes. |
| first_indexed | 2024-04-25T00:24:21Z |
| format | Article |
| id | doaj.art-6294ea7f02db4e4bb1ad36c81d5ccfd6 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-04-25T00:24:21Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-6294ea7f02db4e4bb1ad36c81d5ccfd62024-03-12T16:49:53ZengMDPI AGMathematics2227-73902024-02-0112566710.3390/math12050667Event-Triggered Relearning Modeling Method for Stochastic System with Non-Stationary Variable Operating ConditionsJiyan Liu0Yong Zhang1Yuyang Zhou2Jing Chen3School of Automation and Electrical Engineering, Inner Mongolia University of Science and Technology, Baotou 014010, ChinaSchool of Automation and Electrical Engineering, Inner Mongolia University of Science and Technology, Baotou 014010, ChinaSchool of Computing Engineering and Built Environment, Edinburgh Napier University, Edinburgh EH10 5DT, UKSchool of Science, Jiangnan University, Wuxi 214122, ChinaThis study presents a novel event-triggered relearning framework for neural network modeling, designed to improve prediction precision in dynamic stochastic complex industrial systems under non-stationary and variable conditions. Firstly, a sliding window algorithm combined with entropy is applied to divide the input and output datasets across different operational conditions, establishing clear data boundaries. Following this, the prediction errors derived from the neural network under different operational states are harnessed to define a set of event-triggered relearning criteria. Once these conditions are triggered, the relevant dataset is used to recalibrate the model to the specific operational condition and predict the data under this operating condition. When the predicted data fall within the training input range of a pre-trained model, we switch to that model for immediate prediction. Compared with the conventional BP neural network model and random vector functional-link network, the proposed model can produce a better estimation accuracy and reduce computation costs. Finally, the effectiveness of our proposed method is validated through numerical simulation tests using nonlinear Hammerstein models with Gaussian noise, reflecting complex stochastic industrial processes.https://www.mdpi.com/2227-7390/12/5/667stochastic processesnon-stationary and variable conditionsevent-triggered conditionssliding window algorithminformation entropy |
| spellingShingle | Jiyan Liu Yong Zhang Yuyang Zhou Jing Chen Event-Triggered Relearning Modeling Method for Stochastic System with Non-Stationary Variable Operating Conditions Mathematics stochastic processes non-stationary and variable conditions event-triggered conditions sliding window algorithm information entropy |
| title | Event-Triggered Relearning Modeling Method for Stochastic System with Non-Stationary Variable Operating Conditions |
| title_full | Event-Triggered Relearning Modeling Method for Stochastic System with Non-Stationary Variable Operating Conditions |
| title_fullStr | Event-Triggered Relearning Modeling Method for Stochastic System with Non-Stationary Variable Operating Conditions |
| title_full_unstemmed | Event-Triggered Relearning Modeling Method for Stochastic System with Non-Stationary Variable Operating Conditions |
| title_short | Event-Triggered Relearning Modeling Method for Stochastic System with Non-Stationary Variable Operating Conditions |
| title_sort | event triggered relearning modeling method for stochastic system with non stationary variable operating conditions |
| topic | stochastic processes non-stationary and variable conditions event-triggered conditions sliding window algorithm information entropy |
| url | https://www.mdpi.com/2227-7390/12/5/667 |
| work_keys_str_mv | AT jiyanliu eventtriggeredrelearningmodelingmethodforstochasticsystemwithnonstationaryvariableoperatingconditions AT yongzhang eventtriggeredrelearningmodelingmethodforstochasticsystemwithnonstationaryvariableoperatingconditions AT yuyangzhou eventtriggeredrelearningmodelingmethodforstochasticsystemwithnonstationaryvariableoperatingconditions AT jingchen eventtriggeredrelearningmodelingmethodforstochasticsystemwithnonstationaryvariableoperatingconditions |