Pub Date : 2025-02-01DOI: 10.1016/j.isatra.2024.12.011
Wei Zhu, Yulan Liang, Chao Pei
This paper investigates event-triggered affine formation control of second-order multi-agent systems with directed communication graph. An approach based on complex-valued Laplacian is used as a means of avoiding the use of global information. Two event-triggered strategies are proposed, both of which are capable of achieving global convergence and forming the desired formation without Zeno-behavior, while also optimizing the utilization of resources. Moreover, the dynamic event-triggered strategy conserves more communication resources by comparison. Finally, a numerical simulation example is provided to demonstrate the effectiveness of the obtained conclusions.
{"title":"Event-triggered affine formation of second-order multi-agent systems via complex-valued Laplacian","authors":"Wei Zhu, Yulan Liang, Chao Pei","doi":"10.1016/j.isatra.2024.12.011","DOIUrl":"10.1016/j.isatra.2024.12.011","url":null,"abstract":"<div><div>This paper investigates event-triggered affine formation control of second-order multi-agent systems with directed communication graph. An approach based on complex-valued Laplacian is used as a means of avoiding the use of global information. Two event-triggered strategies are proposed, both of which are capable of achieving global convergence and forming the desired formation without Zeno-behavior, while also optimizing the utilization of resources. Moreover, the dynamic event-triggered strategy conserves more communication resources by comparison. Finally, a numerical simulation example is provided to demonstrate the effectiveness of the obtained conclusions.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"157 ","pages":"Pages 68-77"},"PeriodicalIF":6.3,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142866894","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.isatra.2024.12.014
Wenyang Hu, Tianyang Wang, Fulei Chu
The conventional widely-used health monitoring methods for rotating machines have shortcomings such as the reliance on the selection of the preset parameters. Also, the strong noise interference caused by factors such as transmission path hinders the practical application of many fault feature extraction methods. To overcome these gaps, the digital twin notion is introduced and a new digital twin architecture called the Ramanujan Digital Twin (RDT) is designed. The Ramanujan Periodic Transform (RPT) model is employed to isolate the potential fault feature. For each frame in the whole life cycle of the rotating machine, the high-fidelity simulation model is constructed. Once the high-fidelity simulation-induced virtual sample is obtained, the RPT will be used to provide guidance information about the potential fault. With this information, the potential fault feature can be extracted without preset parameter selection. A health indicator (HI) can be constructed to perform multiple service end tasks including health monitoring and early fault prediction. Two case studies are carried out and the results show the proposed method can not only extract the potential fault feature more effectively with less noise interference but also monitor and predict the potential early fault earlier than fault log.
{"title":"A faulty simulation model guided Ramanujan Digital twin architecture for rotating machine health monitoring","authors":"Wenyang Hu, Tianyang Wang, Fulei Chu","doi":"10.1016/j.isatra.2024.12.014","DOIUrl":"10.1016/j.isatra.2024.12.014","url":null,"abstract":"<div><div>The conventional widely-used health monitoring methods for rotating machines have shortcomings such as the reliance on the selection of the preset parameters. Also, the strong noise interference caused by factors such as transmission path hinders the practical application of many fault feature extraction methods. To overcome these gaps, the digital twin notion is introduced and a new digital twin architecture called the Ramanujan Digital Twin (RDT) is designed. The Ramanujan Periodic Transform (RPT) model is employed to isolate the potential fault feature. For each frame in the whole life cycle of the rotating machine, the high-fidelity simulation model is constructed. Once the high-fidelity simulation-induced virtual sample is obtained, the RPT will be used to provide guidance information about the potential fault. With this information, the potential fault feature can be extracted without preset parameter selection. A health indicator (HI) can be constructed to perform multiple service end tasks including health monitoring and early fault prediction. Two case studies are carried out and the results show the proposed method can not only extract the potential fault feature more effectively with less noise interference but also monitor and predict the potential early fault earlier than fault log.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"157 ","pages":"Pages 408-418"},"PeriodicalIF":6.3,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142901452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.isatra.2024.12.021
Zhaoxin Wang, Jianchang Liu
This study constructs virtual vector triangles in multidimensional space to address cooperative control issue in time-varying nonlinear multi-agent systems. The distributed adaptive virtual point and its dynamic equations are designed, with this virtual point, the leader, and the follower being respectively defined as the vertices of the virtual vector triangle. The virtual vector edges, decomposed by vectors into coordinate axis components, are organized to form a closed virtual vector triangle by connecting the three vertices with directed vector arrows that are oriented from the tail to the head. Specifically, these virtual vector edges are fictitious vector line segments connecting two vertices and used to compute the relative Euclidean distances between each vertex in multidimensional space. Based on the established virtual vector triangles, which are placed in multidimensional space, and the novel spatial coordinate transformation method, the cooperative regulation problem of the time-varying nonlinear multi-agent system is transformed into a mathematical problem of compressing the virtual vector triangles with exponential magnitude. The created distributed compression control protocol asymptotically shrinks the magnitude of the virtual vector triangles by exponential oscillatory decay towards the same dynamic point aligned with the motion trajectory of the leader or the leader, where the states of the time-varying nonlinear multi-agent systems achieve asymptotic convergence consensus. The reliable stability of the asymptotic compression convergence process of the virtual vector triangles was verified by establishing a Lyapunov function and relying on the Lyapunov stability theory. Finally, the example of time-varying nonlinear multi-agent systems are presented for simulation experiments to further validate the effectiveness and feasibility of the proposed control protocol in addressing the cooperative regulation issue.
{"title":"Cooperative regulation based on virtual vector triangles asymptotically compressed in multidimensional space for time-varying nonlinear multi-agent systems","authors":"Zhaoxin Wang, Jianchang Liu","doi":"10.1016/j.isatra.2024.12.021","DOIUrl":"10.1016/j.isatra.2024.12.021","url":null,"abstract":"<div><div>This study constructs virtual vector triangles in multidimensional space to address cooperative control issue in time-varying nonlinear multi-agent systems. The distributed adaptive virtual point and its dynamic equations are designed, with this virtual point, the leader, and the follower being respectively defined as the vertices of the virtual vector triangle. The virtual vector edges, decomposed by vectors into coordinate axis components, are organized to form a closed virtual vector triangle by connecting the three vertices with directed vector arrows that are oriented from the tail to the head. Specifically, these virtual vector edges are fictitious vector line segments connecting two vertices and used to compute the relative Euclidean distances between each vertex in multidimensional space. Based on the established virtual vector triangles, which are placed in multidimensional space, and the novel spatial coordinate transformation method, the cooperative regulation problem of the time-varying nonlinear multi-agent system is transformed into a mathematical problem of compressing the virtual vector triangles with exponential magnitude. The created distributed compression control protocol asymptotically shrinks the magnitude of the virtual vector triangles by exponential oscillatory decay towards the same dynamic point aligned with the motion trajectory of the leader or the leader, where the states of the time-varying nonlinear multi-agent systems achieve asymptotic convergence consensus. The reliable stability of the asymptotic compression convergence process of the virtual vector triangles was verified by establishing a Lyapunov function and relying on the Lyapunov stability theory. Finally, the example of time-varying nonlinear multi-agent systems are presented for simulation experiments to further validate the effectiveness and feasibility of the proposed control protocol in addressing the cooperative regulation issue.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"157 ","pages":"Pages 258-268"},"PeriodicalIF":6.3,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142901454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.isatra.2024.12.013
Lingli Cui , Wenjie Li , Xin Wang , Dongdong Liu
Dual-impulse behaviors of rolling bearings have been widely researched for quantitative diagnosis. However, it is challenging to accurately extract entry and exit moments of the fault from noise-contaminated raw signals. To address this issue, a novel quantitative diagnosis method based on digital twin model is proposed to assess the fault severity from the original signal waveform. Specifically, the quantitative diagnostic criterion for bearing faults is derived to reveal the instantaneous response characteristics of dual-impulse behaviors, and then a digital twin model is constructed to characterize the fault characteristics of the measured signal with noise-free twin signals. Subsequently, a recursive parameter optimization strategy based on cosine similarity (RPOS-CS) is proposed to optimize the twin model in real time, and fault parameters of the optimal signal will be applied to evaluate the fault size of the bearing. Finally, kernel density estimation is employed to perform uncertainty analysis on multiple diagnosis results, thereby realizing interval estimation and significantly enhancing the reliability of diagnosis results. Both simulated and experimental signals are utilized to validate the efficacy of the proposed method, and the further comparative analysis shows that it exhibits high diagnostic accuracy and outstanding reliability.
{"title":"A novel quantitative diagnosis method for rolling bearing faults based on digital twin model","authors":"Lingli Cui , Wenjie Li , Xin Wang , Dongdong Liu","doi":"10.1016/j.isatra.2024.12.013","DOIUrl":"10.1016/j.isatra.2024.12.013","url":null,"abstract":"<div><div>Dual-impulse behaviors of rolling bearings have been widely researched for quantitative diagnosis. However, it is challenging to accurately extract entry and exit moments of the fault from noise-contaminated raw signals. To address this issue, a novel quantitative diagnosis method based on digital twin model is proposed to assess the fault severity from the original signal waveform. Specifically, the quantitative diagnostic criterion for bearing faults is derived to reveal the instantaneous response characteristics of dual-impulse behaviors, and then a digital twin model is constructed to characterize the fault characteristics of the measured signal with noise-free twin signals. Subsequently, a recursive parameter optimization strategy based on cosine similarity (RPOS-CS) is proposed to optimize the twin model in real time, and fault parameters of the optimal signal will be applied to evaluate the fault size of the bearing. Finally, kernel density estimation is employed to perform uncertainty analysis on multiple diagnosis results, thereby realizing interval estimation and significantly enhancing the reliability of diagnosis results. Both simulated and experimental signals are utilized to validate the efficacy of the proposed method, and the further comparative analysis shows that it exhibits high diagnostic accuracy and outstanding reliability.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"157 ","pages":"Pages 381-391"},"PeriodicalIF":6.3,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142911172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.isatra.2024.12.024
Caoyuan Gu , Qi Wu , Baokang Zhang , Yaowei Wang , Wen-An Zhang , Hongjie Ni
This paper proposes an improved remaining useful life (RUL) prediction method for stochastic degradation devices monitored by multi-source sensors under data-model interactive framework. Firstly, the interrelationships among sensors are established using k-nearest neighbor (KNN), and the composite health index (CHI) is constructed by aggregating the multi-source sensor information through the graph convolutional network (GCN). Secondly, a stochastic degradation model with triple uncertainty at any initial degradation level is established to improve the matching degree between the stochastic degradation model and the actual degradation process. Then, a data-model interactive mechanism is proposed to form a closed-loop optimization between the CHI construction and the stochastic degradation model to enhance the RUL prediction accuracy of the device. Finally, experiments on aero-engine and tool datasets indicate that the proposed method can improve the comprehensive performance by at least 20% compared with the original method of the data-model interactive framework, which verifies its effectiveness and superiority.
{"title":"Data-model interactive Rul prediction of stochastic degradation devices with multiple uncertainty quantification and multi-sensor information fusion","authors":"Caoyuan Gu , Qi Wu , Baokang Zhang , Yaowei Wang , Wen-An Zhang , Hongjie Ni","doi":"10.1016/j.isatra.2024.12.024","DOIUrl":"10.1016/j.isatra.2024.12.024","url":null,"abstract":"<div><div>This paper proposes an improved remaining useful life (RUL) prediction method for stochastic degradation devices monitored by multi-source sensors under data-model interactive framework. Firstly, the interrelationships among sensors are established using k-nearest neighbor (KNN), and the composite health index (CHI) is constructed by aggregating the multi-source sensor information through the graph convolutional network (GCN). Secondly, a stochastic degradation model with triple uncertainty at any initial degradation level is established to improve the matching degree between the stochastic degradation model and the actual degradation process. Then, a data-model interactive mechanism is proposed to form a closed-loop optimization between the CHI construction and the stochastic degradation model to enhance the RUL prediction accuracy of the device. Finally, experiments on aero-engine and tool datasets indicate that the proposed method can improve the comprehensive performance by at least 20% compared with the original method of the data-model interactive framework, which verifies its effectiveness and superiority.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"157 ","pages":"Pages 293-305"},"PeriodicalIF":6.3,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142960720","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.isatra.2024.12.032
John Cortés-Romero , Brian Gómez-León , Hebertt Sira-Ramírez
In control system design, managing measurement noise is a critical challenge, requiring a balance between responsiveness and noise suppression. Traditional methods often involve trade-offs, compromising either aspect. This paper proposes a novel solution by integrating a Finite Impulse Response (FIR) filter within the discrete controller transfer function, coupled with disturbance rejection through the Internal Model Principle (IMP). This approach decouples tracking reference dynamics from noise attenuation, ensuring precise disturbance rejection and reference tracking. Numerical analysis highlights significant immunity to high-frequency noise compared to established methods. Practical feasibility is demonstrated through two experimental validations: implementation on a Programmable Logic Controller (PLC) and robustness testing with a DC–DC boost converter in a high-noise environment, underscoring its applicability in real-world industrial scenarios.
{"title":"Discrete FIR filter-based Control","authors":"John Cortés-Romero , Brian Gómez-León , Hebertt Sira-Ramírez","doi":"10.1016/j.isatra.2024.12.032","DOIUrl":"10.1016/j.isatra.2024.12.032","url":null,"abstract":"<div><div>In control system design, managing measurement noise is a critical challenge, requiring a balance between responsiveness and noise suppression. Traditional methods often involve trade-offs, compromising either aspect. This paper proposes a novel solution by integrating a Finite Impulse Response (FIR) filter within the discrete controller transfer function, coupled with disturbance rejection through the Internal Model Principle (IMP). This approach decouples tracking reference dynamics from noise attenuation, ensuring precise disturbance rejection and reference tracking. Numerical analysis highlights significant immunity to high-frequency noise compared to established methods. Practical feasibility is demonstrated through two experimental validations: implementation on a Programmable Logic Controller (PLC) and robustness testing with a DC–DC boost converter in a high-noise environment, underscoring its applicability in real-world industrial scenarios.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"157 ","pages":"Pages 591-602"},"PeriodicalIF":6.3,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143349663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a robust hybrid method using an analytical model and morphological analysis for estimating defect sizes on bearings, addressing inaccuracies in traditional peak-based approaches. First, a spatial contact analytical model is developed to analyze roller movement through defects. Numerical results indicate that radial defect sizes significantly impact dual-impulse intervals for accurate estimation, while axial size has minimal effect. An analysis of contact deformation, force, and vibration response reveals that the defect entry aligns before the first low-frequency peak, and the exit aligns after the first high-frequency peak, preceding the second low-frequency peak. Finally, testing of the derived model on four samples, incorporating vibration characteristics and geometry, achieves a minimum estimation error of 0.03%, demonstrating high accuracy and practical value.
{"title":"A robust hybrid estimation method for local bearing defect size based on analytical model and morphological analysis","authors":"Zepeng Ma, Lei Fu, Dapeng Tan, Meiya Ding, Fang Xu, Libin Zhang","doi":"10.1016/j.isatra.2024.11.056","DOIUrl":"10.1016/j.isatra.2024.11.056","url":null,"abstract":"<div><div>This paper presents a robust hybrid method using an analytical model and morphological analysis for estimating defect sizes on bearings, addressing inaccuracies in traditional peak-based approaches. First, a spatial contact analytical model is developed to analyze roller movement through defects. Numerical results indicate that radial defect sizes significantly impact dual-impulse intervals for accurate estimation, while axial size has minimal effect. An analysis of contact deformation, force, and vibration response reveals that the defect entry aligns before the first low-frequency peak, and the exit aligns after the first high-frequency peak, preceding the second low-frequency peak. Finally, testing of the derived model on four samples, incorporating vibration characteristics and geometry, achieves a minimum estimation error of 0.03%, demonstrating high accuracy and practical value.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"157 ","pages":"Pages 392-407"},"PeriodicalIF":6.3,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142796541","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.isatra.2024.12.007
Jingxin Huang , Xiangze Lin , Xueling Li , Shihua Li
In this paper, non-smooth fixed-time state feedback stabilization of output-constrained uncertain switched systems has been investigated. First, to deal with the output constraints implicitly, a tangent-type barrier Lyapunov function (-BLF), which degenerates into a quadratic function as the output constraints tend to infinity, is constructed deliberately. Then, the fixed-time state feedback control scheme for the output-constrained uncertain switched systems is established with remoulding the technique called adding a power integrator (AAPI). The upper bound of the settling time does not depend on the initial state and the arbitrarily small settling time can be obtained by just adjusting the specific parameters. The proposed method is operated in a unified framework, because it can tackle the cases for the uncertain switched systems with/without output constraints without requiring changes of switched controllers structure. At last, the effectiveness of the method is verified by a numerical simulation and a two-tank liquid-level system.
{"title":"Non-smooth fixed-time state feedback stabilization of uncertain switched systems with output constraints","authors":"Jingxin Huang , Xiangze Lin , Xueling Li , Shihua Li","doi":"10.1016/j.isatra.2024.12.007","DOIUrl":"10.1016/j.isatra.2024.12.007","url":null,"abstract":"<div><div>In this paper, non-smooth fixed-time state feedback stabilization of output-constrained uncertain switched systems has been investigated. First, to deal with the output constraints implicitly, a tangent-type barrier Lyapunov function (<span><math><msub><mrow><mi>T</mi></mrow><mrow><mi>a</mi><mi>n</mi></mrow></msub></math></span>-BLF), which degenerates into a quadratic function as the output constraints tend to infinity, is constructed deliberately. Then, the fixed-time state feedback control scheme for the output-constrained uncertain switched systems is established with remoulding the technique called adding a power integrator (AAPI). The upper bound of the settling time does not depend on the initial state and the arbitrarily small settling time can be obtained by just adjusting the specific parameters. The proposed method is operated in a unified framework, because it can tackle the cases for the uncertain switched systems with/without output constraints without requiring changes of switched controllers structure. At last, the effectiveness of the method is verified by a numerical simulation and a two-tank liquid-level system.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"157 ","pages":"Pages 269-279"},"PeriodicalIF":6.3,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142866252","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.isatra.2024.12.023
Yawei Sun , Hongfeng Tao , Vladimir Stojanovic
When the fault diagnosis datasets contains noise disturbances, small samples, compound faults, and mixed conditions, the feature extraction capability of the neural network will face significant challenges. This paper proposes an end-to-end multi-scale residual network with parallel attention mechanism to address the above complex problems. Firstly, the adaptive mixing pooling method is employed to facilitate the model’s ability to retain effective feature information present within the timing signal. Then, we propose parallel attention mechanism that can obtain the attention information in both channel and temporal domain of the input features. Moreover, the multi-scale feature parallel fusion can better capture effective information contained in different scale features. The experimental results demonstrate that the proposed model attains , , and accuracy on four datasets comprising small samples. Furthermore, the accuracy of to is sustained when the noise level is increased to 0dB.
{"title":"End-to-end multi-scale residual network with parallel attention mechanism for fault diagnosis under noise and small samples","authors":"Yawei Sun , Hongfeng Tao , Vladimir Stojanovic","doi":"10.1016/j.isatra.2024.12.023","DOIUrl":"10.1016/j.isatra.2024.12.023","url":null,"abstract":"<div><div>When the fault diagnosis datasets contains noise disturbances, small samples, compound faults, and mixed conditions, the feature extraction capability of the neural network will face significant challenges. This paper proposes an end-to-end multi-scale residual network with parallel attention mechanism to address the above complex problems. Firstly, the adaptive mixing pooling method is employed to facilitate the model’s ability to retain effective feature information present within the timing signal. Then, we propose parallel attention mechanism that can obtain the attention information in both channel and temporal domain of the input features. Moreover, the multi-scale feature parallel fusion can better capture effective information contained in different scale features. The experimental results demonstrate that the proposed model attains <span><math><mrow><mn>99</mn><mo>.</mo><mn>67</mn></mrow></math></span> <span><math><mtext>%</mtext></math></span>, <span><math><mrow><mn>99</mn><mo>.</mo><mn>83</mn></mrow></math></span> <span><math><mtext>%</mtext></math></span>, <span><math><mrow><mn>99</mn><mo>.</mo><mn>71</mn></mrow></math></span> <span><math><mtext>%</mtext></math></span> and <span><math><mrow><mn>99</mn><mo>.</mo><mn>70</mn></mrow></math></span> <span><math><mtext>%</mtext></math></span> accuracy on four datasets comprising small samples. Furthermore, the accuracy of <span><math><mrow><mn>60</mn></mrow></math></span> <span><math><mtext>%</mtext></math></span> to <span><math><mrow><mn>80</mn></mrow></math></span> <span><math><mtext>%</mtext></math></span> is sustained when the noise level is increased to 0dB.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"157 ","pages":"Pages 419-433"},"PeriodicalIF":6.3,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142879125","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.isatra.2024.12.025
Yuan-Jay Wang
This article introduces a novel setpoint weighted PI-FOPD (SWPI-FOPD) cascade controller with a prefilter. It further describes a four-stage design strategy that sequentially enhances tracking responses, reduces overshoot, and ensures robustness for unstable time-delayed (UTD) processes. The controller applies to integrating and non-integrating UTD processes of any order and does not necessitate model order reduction or delay approximation. The strategy involves four stages, each targeting a specific aspect of performance improvement. The first stage designs the optimal and robust FOPD (ORFOPD) controller to accelerate tracking responses. In the second stage, the cascade optimal and robust PI (ORPI) controller ensures zero steady-state error. A gain and phase margin (GPM) tester determines the feasible specification-oriented regions (FSORs) on the KP1-KD1 and KP2-KI2 planes, aiding in selecting the ORFOPD and ORPI controllers to minimize the IAE or ISE index under ± 10% loop gain perturbations (LGPs). The third stage schedules setpoint weighting coefficients to reduce overshoot typically associated with UTD processes. The final stage uses a prefilter to refine the reference input signal and mitigate overshoot. A systematic design procedure is summarized to rapidly compute all controller gains in a step-by-step sequence. This SWPI-FOPD controller with a prefilter enhances robustness, provides rapid tracking response with zero steady-state error, and effectively reduces overshoot and IAE or ISE, even with ± 10% LGPs. Comparative simulations across three examples, along with a practical jacketed CSTR application, validate the feasibility and demonstrate the superior performance of the proposed controller.
{"title":"Setpoint weighted PI-FOPD cascade controllers synthesis for unstable time-delayed processes satisfying prespecified safety margins","authors":"Yuan-Jay Wang","doi":"10.1016/j.isatra.2024.12.025","DOIUrl":"10.1016/j.isatra.2024.12.025","url":null,"abstract":"<div><div>This article introduces a novel setpoint weighted PI-FOPD (SWPI-FOPD) cascade controller with a prefilter. It further describes a four-stage design strategy that sequentially enhances tracking responses, reduces overshoot, and ensures robustness for unstable time-delayed (UTD) processes. The controller applies to integrating and non-integrating UTD processes of any order and does not necessitate model order reduction or delay approximation. The strategy involves four stages, each targeting a specific aspect of performance improvement. The first stage designs the optimal and robust FOPD (ORFOPD) controller to accelerate tracking responses. In the second stage, the cascade optimal and robust PI (ORPI) controller ensures zero steady-state error. A gain and phase margin (GPM) tester determines the feasible specification-oriented regions (FSORs) on the <em>K</em><sub><em>P1</em></sub>-<em>K</em><sub><em>D1</em></sub> and <em>K</em><sub><em>P2</em></sub>-<em>K</em><sub><em>I2</em></sub> planes, aiding in selecting the ORFOPD and ORPI controllers to minimize the IAE or ISE index under ± 10% loop gain perturbations (LGPs). The third stage schedules setpoint weighting coefficients to reduce overshoot typically associated with UTD processes. The final stage uses a prefilter to refine the reference input signal and mitigate overshoot. A systematic design procedure is summarized to rapidly compute all controller gains in a step-by-step sequence. This SWPI-FOPD controller with a prefilter enhances robustness, provides rapid tracking response with zero steady-state error, and effectively reduces overshoot and IAE or ISE, even with ± 10% LGPs. Comparative simulations across three examples, along with a practical jacketed CSTR application, validate the feasibility and demonstrate the superior performance of the proposed controller.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"157 ","pages":"Pages 164-185"},"PeriodicalIF":6.3,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142901458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号