Recently, deep learning based fingerprint localization has attracted significant interest due to its simplicity in implementation and effectiveness in complex multipath environments, especially for the Internet of Things (IoT) devices in multiple-input multiple-output (MIMO)-orthogonal frequency-division multiplexing (OFDM) system. However, the huge amount of training data collection has become a challenge, which increases the labor burden of fingerprint localization heavily and hinders its large-scale implementation. In this paper, we propose a novel fingerprint localization system, termed as SiamResNet, which can be trained only on the radio map by contrastive self-supervised learning without the need for any other additional data. To be more specific, we first model the fingerprint localization problem as a dictionary look-up task. Subsequently, a channel fingerprint capturing the multipath angle and delay of wireless propagation is introduced, which exhibits excellent uniqueness, stability, and distinguishability. Meanwhile, we propose the corresponding data augmentation strategy to ensure data diversity when generating the training data from the radio map. Thus, the cost of data collection for training can be significantly reduced. Lastly, the Siamese architecture based SiamResNet is applied for location estimation, which can comprehensively extract the features of fingerprints and accurately compare the similarity of any fingerprint to the radio map in the representation space. The performance of the proposed localization method is validated through extensive simulations with a ray-tracing channel model, which demonstrates promising localization accuracy for our SiamResNet with reduced training costs.
{"title":"Contrastive Self-Supervised Learning-Based Wireless Fingerprint Localization","authors":"Qiao Li;Zhili Zhang;Zhaofa Zhou","doi":"10.23919/CSMS.2024.0032","DOIUrl":"https://doi.org/10.23919/CSMS.2024.0032","url":null,"abstract":"Recently, deep learning based fingerprint localization has attracted significant interest due to its simplicity in implementation and effectiveness in complex multipath environments, especially for the Internet of Things (IoT) devices in multiple-input multiple-output (MIMO)-orthogonal frequency-division multiplexing (OFDM) system. However, the huge amount of training data collection has become a challenge, which increases the labor burden of fingerprint localization heavily and hinders its large-scale implementation. In this paper, we propose a novel fingerprint localization system, termed as SiamResNet, which can be trained only on the radio map by contrastive self-supervised learning without the need for any other additional data. To be more specific, we first model the fingerprint localization problem as a dictionary look-up task. Subsequently, a channel fingerprint capturing the multipath angle and delay of wireless propagation is introduced, which exhibits excellent uniqueness, stability, and distinguishability. Meanwhile, we propose the corresponding data augmentation strategy to ensure data diversity when generating the training data from the radio map. Thus, the cost of data collection for training can be significantly reduced. Lastly, the Siamese architecture based SiamResNet is applied for location estimation, which can comprehensively extract the features of fingerprints and accurately compare the similarity of any fingerprint to the radio map in the representation space. The performance of the proposed localization method is validated through extensive simulations with a ray-tracing channel model, which demonstrates promising localization accuracy for our SiamResNet with reduced training costs.","PeriodicalId":65786,"journal":{"name":"复杂系统建模与仿真(英文)","volume":"5 3","pages":"261-281"},"PeriodicalIF":0.0,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979681","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144896769","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mohammad Fathi;Hossein Bolandi;Bahman Ghobrani Vaghei;Saeed Ebadollahi
The present paper attempts to design an adaptive multi-model predictive control strategy for strongly nonlinear or switched systems with various operating points. The proposed control system guarantees the feasibility and the asymptotic stability of the closed-loop system, considering various challenges such as inherent uncertainties in the local models constituting the model bank, limited prediction/control horizons, and set point changes. To this end, four fundamental challenges in this area, namely guaranteeing feasibility throughout the region assigned to each subspace, ensuring asymptotic stability in each subspace considering the inherent uncertainties of the local models, guaranteeing feasibility and asymptotic stability during changes in the set point and switching between the subspaces, are addressed. By introducing transferring mode concept, this paper presents a novel method for guaranteeing the feasibility and stability of the switched systems without the need for increasing the prediction/control horizons or decreasing the size of the feasibility region. The proposed control structure uses a supervisor algorithm along with a soft-switching technique. The supervisor algorithm is responsible for determining the suitable local model/controller pair, determining the operational mode of the control system, managing the soft switching, and specifying the control objectives in accordance with the defined set point. The efficiency of the proposed control strategy is demonstrated by simulating a Continuous Stirred Tank Reactor (CSTR) as the controlled system. Based on the results, the proposed controller is able to guarantee the feasibility and stability of highly nonlinear and switched systems in a wide operating region under set point changes and uncertainties in the local models.
{"title":"On Feasibility and Asymptotically Stability of Switched Systems Using Adaptive Multi-Model Predictive Control","authors":"Mohammad Fathi;Hossein Bolandi;Bahman Ghobrani Vaghei;Saeed Ebadollahi","doi":"10.23919/CSMS.2024.0030","DOIUrl":"https://doi.org/10.23919/CSMS.2024.0030","url":null,"abstract":"The present paper attempts to design an adaptive multi-model predictive control strategy for strongly nonlinear or switched systems with various operating points. The proposed control system guarantees the feasibility and the asymptotic stability of the closed-loop system, considering various challenges such as inherent uncertainties in the local models constituting the model bank, limited prediction/control horizons, and set point changes. To this end, four fundamental challenges in this area, namely guaranteeing feasibility throughout the region assigned to each subspace, ensuring asymptotic stability in each subspace considering the inherent uncertainties of the local models, guaranteeing feasibility and asymptotic stability during changes in the set point and switching between the subspaces, are addressed. By introducing transferring mode concept, this paper presents a novel method for guaranteeing the feasibility and stability of the switched systems without the need for increasing the prediction/control horizons or decreasing the size of the feasibility region. The proposed control structure uses a supervisor algorithm along with a soft-switching technique. The supervisor algorithm is responsible for determining the suitable local model/controller pair, determining the operational mode of the control system, managing the soft switching, and specifying the control objectives in accordance with the defined set point. The efficiency of the proposed control strategy is demonstrated by simulating a Continuous Stirred Tank Reactor (CSTR) as the controlled system. Based on the results, the proposed controller is able to guarantee the feasibility and stability of highly nonlinear and switched systems in a wide operating region under set point changes and uncertainties in the local models.","PeriodicalId":65786,"journal":{"name":"复杂系统建模与仿真(英文)","volume":"5 2","pages":"155-175"},"PeriodicalIF":0.0,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979683","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143937970","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Hua Xie;Zhe Liu;Zhihong Xu;Tao Zhu;Yihan Hu;Kai Li
Hydrogen production from wind-solar generation is of great importance for consuming renewable energy and it is meeting industrial hydrogen demand. In this paper, the modelling of the off-grid hydrogen production system from wind-solar generation and the simulation of its operating characteristics are investigated. Firstly, the network architecture and hierarchical control architecture of the off-grid hydrogen generation system are designed with the goal of efficiently utilising wind-solar generation output. Then, the components of the off-grid hydrogen generation system are characterised and modelled. Finally, the operating characteristics of the hydrogen production system under three operating conditions, such as hydrogen system startup, wind power fluctuation, and electrolyzer partial failure are simulated and analyzed, revealing the ability of the alkaline electrolytic water hydrogen production system to respond to the fluctuation of wind and solar power.
{"title":"Operational Characteristics Simulation for Off-Grid Hydrogen Production System with Wind-Solar Power Generation","authors":"Hua Xie;Zhe Liu;Zhihong Xu;Tao Zhu;Yihan Hu;Kai Li","doi":"10.23919/CSMS.2024.0034","DOIUrl":"https://doi.org/10.23919/CSMS.2024.0034","url":null,"abstract":"Hydrogen production from wind-solar generation is of great importance for consuming renewable energy and it is meeting industrial hydrogen demand. In this paper, the modelling of the off-grid hydrogen production system from wind-solar generation and the simulation of its operating characteristics are investigated. Firstly, the network architecture and hierarchical control architecture of the off-grid hydrogen generation system are designed with the goal of efficiently utilising wind-solar generation output. Then, the components of the off-grid hydrogen generation system are characterised and modelled. Finally, the operating characteristics of the hydrogen production system under three operating conditions, such as hydrogen system startup, wind power fluctuation, and electrolyzer partial failure are simulated and analyzed, revealing the ability of the alkaline electrolytic water hydrogen production system to respond to the fluctuation of wind and solar power.","PeriodicalId":65786,"journal":{"name":"复杂系统建模与仿真(英文)","volume":"5 3","pages":"282-295"},"PeriodicalIF":0.0,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10979682","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144896765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Affected by the limited interchange spacing, the operational risk of vehicles in expressway small-spacing interchanges (SSls) is more complex compared to other interchanges. In this study, unmanned aerial vehicle (UAV) measurements were integrated with joint simulation data to explore the risk characteristics of SSls with the help of traffic conflict theory. Seven traffic flow parameters, including mainline traffic volume, were selected to evaluate their impact on traffic conflicts. The distribution of four traffic conflict indicators, such as time to collision (TTC), was analyzed, and their severity was categorized using cumulative frequency analysis and minibatch K-means clustering. By varying the spacing, the study scrutinized trends in traffic conflicts, emphasizing the influence of various traffic flow parameters, distinctions in conflict indicators, and the ratio of severe conflicts to total conflicts. Additionally, an analysis of the spatial distribution of severe conflicts was conducted. The results suggested that traffic conflicts in SSls are influenced by multiple factors, with mainline and entry traffic volumes being the most significant. Heavy vehicle proportions and entry ramp speeds had notable effects under certain spacing conditions. Considerable variations were observed in conflict indicators across different spacings, with the maximum conflict speed being the most affected by spacing, while TTC was the least. As spacing increased, the proportion of severe conflicts decreased, with severe TTC dropping from 18% to 10%. High-density conflict zones were identified near merging points in the second and third lanes. With larger spacing, the conflict zone range narrowed while the density of conflict points intensified.
{"title":"Traffic Conflict and Risk Analysis of Small-Spacing Interchange Interweaving Sections on Expressways Using Joint Simulation","authors":"Yanpeng Wang;Fanxing Kong;Zhanji Zheng;Xingliang Liu;Heshan Zhang;Jin Xu","doi":"10.23919/CSMS.2024.0031","DOIUrl":"https://doi.org/10.23919/CSMS.2024.0031","url":null,"abstract":"Affected by the limited interchange spacing, the operational risk of vehicles in expressway small-spacing interchanges (SSls) is more complex compared to other interchanges. In this study, unmanned aerial vehicle (UAV) measurements were integrated with joint simulation data to explore the risk characteristics of SSls with the help of traffic conflict theory. Seven traffic flow parameters, including mainline traffic volume, were selected to evaluate their impact on traffic conflicts. The distribution of four traffic conflict indicators, such as time to collision (TTC), was analyzed, and their severity was categorized using cumulative frequency analysis and minibatch K-means clustering. By varying the spacing, the study scrutinized trends in traffic conflicts, emphasizing the influence of various traffic flow parameters, distinctions in conflict indicators, and the ratio of severe conflicts to total conflicts. Additionally, an analysis of the spatial distribution of severe conflicts was conducted. The results suggested that traffic conflicts in SSls are influenced by multiple factors, with mainline and entry traffic volumes being the most significant. Heavy vehicle proportions and entry ramp speeds had notable effects under certain spacing conditions. Considerable variations were observed in conflict indicators across different spacings, with the maximum conflict speed being the most affected by spacing, while TTC was the least. As spacing increased, the proportion of severe conflicts decreased, with severe TTC dropping from 18% to 10%. High-density conflict zones were identified near merging points in the second and third lanes. With larger spacing, the conflict zone range narrowed while the density of conflict points intensified.","PeriodicalId":65786,"journal":{"name":"复杂系统建模与仿真(英文)","volume":"5 2","pages":"176-189"},"PeriodicalIF":0.0,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10969594","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143938001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Electric motors are pivotal yet vulnerable components in Electric Vehicles (EVs), with stator winding faults, particularly inter-turn faults, being among the most frequent and hazardous. Early detection of these faults is crucial for predictive maintenance and ensuring the reliability and safety of EVs. This study introduces a Time-aware Convolutional Transformer (TaCT) model that integrates transformer architecture with convolutional layers and a novel time-series specific positional encoding to enhance fault diagnosis performance by capturing long-range dependencies in time-series data, which are critical for detecting subtle, developing faults. A novel deep transfer learning approach, integrated within a digital twin framework, which creates a virtual replica of the physical motor, is proposed to improve fault diagnosis efficiency and generalization by treating data from time-varying conditions as a continuous domain shift. Four distinct transfer learning methodologies were employed to update and refine digital twin models for fault diagnosis. The TaCT model demonstrated markedly superior performance, maintaining an accuracy above 0.95 across all update steps, outperforming other deep learning models. Notably, TaCT's architecture proved particularly effective for short-circuit fault detection, as confirmed by a Conover test showing that it achieved the highest average rank. The digital twin transfer learning approach mitigated the issue of catastrophic forgetting, which occurs when a model loses previously acquired knowledge upon learning new information, and significantly improved model performance over multiple update steps. This research highlights the advantages of integrating advanced deep learning models with digital twin frameworks and transfer learning techniques, offering substantial improvements in EV motors' predictive maintenance and fault diagnosis.
{"title":"Leveraging Deep Transfer Learning and Time-Aware Convolutional Transformers for Stator Winding Fault Diagnosis in Electric Motors: A Digital Twin Approach","authors":"Imron Rosyadi;Yul Yunazwin Nazaruddin;Parsaulian Ishaya Siregar","doi":"10.23919/CSMS.2025.0005","DOIUrl":"https://doi.org/10.23919/CSMS.2025.0005","url":null,"abstract":"Electric motors are pivotal yet vulnerable components in Electric Vehicles (EVs), with stator winding faults, particularly inter-turn faults, being among the most frequent and hazardous. Early detection of these faults is crucial for predictive maintenance and ensuring the reliability and safety of EVs. This study introduces a Time-aware Convolutional Transformer (TaCT) model that integrates transformer architecture with convolutional layers and a novel time-series specific positional encoding to enhance fault diagnosis performance by capturing long-range dependencies in time-series data, which are critical for detecting subtle, developing faults. A novel deep transfer learning approach, integrated within a digital twin framework, which creates a virtual replica of the physical motor, is proposed to improve fault diagnosis efficiency and generalization by treating data from time-varying conditions as a continuous domain shift. Four distinct transfer learning methodologies were employed to update and refine digital twin models for fault diagnosis. The TaCT model demonstrated markedly superior performance, maintaining an accuracy above 0.95 across all update steps, outperforming other deep learning models. Notably, TaCT's architecture proved particularly effective for short-circuit fault detection, as confirmed by a Conover test showing that it achieved the highest average rank. The digital twin transfer learning approach mitigated the issue of catastrophic forgetting, which occurs when a model loses previously acquired knowledge upon learning new information, and significantly improved model performance over multiple update steps. This research highlights the advantages of integrating advanced deep learning models with digital twin frameworks and transfer learning techniques, offering substantial improvements in EV motors' predictive maintenance and fault diagnosis.","PeriodicalId":65786,"journal":{"name":"复杂系统建模与仿真(英文)","volume":"5 4","pages":"370-387"},"PeriodicalIF":0.0,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10969589","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145705909","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To improve the disassembly efficiency of a U-shaped disassembly line and reduce the potentially harmful effects on the environment and human health, we study the multi-product U-shaped disassembly line balancing problem with a fixed number of stations (MUDLBPF). Firstly, we formulate a mathematical model aimed at minimizing cycle time, balancing loads, and reducing hazard indicators. Secondly, a multi-objective variable neighborhood search (MOVNS) algorithm is proposed. A multi-segment encoding method is proposed to maintain the independence of different products. Considering the characteristics of multiple products, a two-stage decoding method is presented. The method includes product assignment and task assignment. To optimize decoding efficiency, a minimum deviation method is put forward to generate feasible solutions. A segmented neighborhood structure containing seven operators is developed to improve the search efficiency. Finally, numerical experiments are performed and the results show that the MOVNS can solve the MUDLBPF effectively and efficiently.
{"title":"Multi-Objective Variable Neighborhood Search Algorithm to Optimize the Multi-Product U-Shaped Disassembly Line Balancing Problem with a Fixed Number of Stations","authors":"Xingyu Zhang;Xiuli Wu","doi":"10.23919/CSMS.2024.0033","DOIUrl":"https://doi.org/10.23919/CSMS.2024.0033","url":null,"abstract":"To improve the disassembly efficiency of a U-shaped disassembly line and reduce the potentially harmful effects on the environment and human health, we study the multi-product U-shaped disassembly line balancing problem with a fixed number of stations (MUDLBPF). Firstly, we formulate a mathematical model aimed at minimizing cycle time, balancing loads, and reducing hazard indicators. Secondly, a multi-objective variable neighborhood search (MOVNS) algorithm is proposed. A multi-segment encoding method is proposed to maintain the independence of different products. Considering the characteristics of multiple products, a two-stage decoding method is presented. The method includes product assignment and task assignment. To optimize decoding efficiency, a minimum deviation method is put forward to generate feasible solutions. A segmented neighborhood structure containing seven operators is developed to improve the search efficiency. Finally, numerical experiments are performed and the results show that the MOVNS can solve the MUDLBPF effectively and efficiently.","PeriodicalId":65786,"journal":{"name":"复杂系统建模与仿真(英文)","volume":"5 4","pages":"354-369"},"PeriodicalIF":0.0,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10969536","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145705859","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As the global economy develops and people's awareness of environmental protection increases, the efficient scheduling of production lines in workshops has received more and more attention. However, there is very little research focusing on distributed scheduling for heterogeneous factories. This study addresses a multi-objective distributed heterogeneous permutation flow shop scheduling problem with sequence-dependent setup times (DHPFSP-SDST). The objective is to optimize the trade-off between the maximum completion time (Makespan) and total energy consumption. First, to describe the concerned problems, we establish a mathematical model. Second, we use the artificial bee colony (ABC) algorithm to optimize the two objectives, incorporating five local search strategies tailored to the problem characteristics to enhance the algorithm's performance. Third, to improve the convergence speed of the algorithm, a Q-learning based strategy is designed to select the appropriated local search operator during iterations. Finally, based on experiments conducted on 72 instances, statistical analysis and discussions show that the Q-learning based ABC algorithm can effectively solve the problems better than its peers.
{"title":"Ensemble Artificial Bee Colony Algorithm and Q-Learning for Multi-Objective Distributed Heterogeneous Flowshop Scheduling Problems with Sequence-Dependent Setup Time","authors":"Fubin Liu;Kaizhou Gao;Adam Słowik;Ponnuthurai Nagaratnam Suganthan","doi":"10.23919/CSMS.2024.0040","DOIUrl":"https://doi.org/10.23919/CSMS.2024.0040","url":null,"abstract":"As the global economy develops and people's awareness of environmental protection increases, the efficient scheduling of production lines in workshops has received more and more attention. However, there is very little research focusing on distributed scheduling for heterogeneous factories. This study addresses a multi-objective distributed heterogeneous permutation flow shop scheduling problem with sequence-dependent setup times (DHPFSP-SDST). The objective is to optimize the trade-off between the maximum completion time (Makespan) and total energy consumption. First, to describe the concerned problems, we establish a mathematical model. Second, we use the artificial bee colony (ABC) algorithm to optimize the two objectives, incorporating five local search strategies tailored to the problem characteristics to enhance the algorithm's performance. Third, to improve the convergence speed of the algorithm, a Q-learning based strategy is designed to select the appropriated local search operator during iterations. Finally, based on experiments conducted on 72 instances, statistical analysis and discussions show that the Q-learning based ABC algorithm can effectively solve the problems better than its peers.","PeriodicalId":65786,"journal":{"name":"复杂系统建模与仿真(英文)","volume":"5 3","pages":"221-235"},"PeriodicalIF":0.0,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10969533","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144896788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Multiobjective combinatorial optimization (MOCO) problems have a wide range of applications in the real world. Recently, learning-based methods have achieved good results in solving MOCO problems. However, most of these methods use attention mechanisms and their variants, which have room for further improvement in the speed of solving MOCO problems. In this paper, following the idea of decomposition strategy and neural combinatorial optimization, a novel fast-solving model for MOCO based on retention is proposed. A brand new calculation of retention is proposed, causal masking and exponential decay are deprecated in retention, so that our model could better solve MOCO problems. During model training, a parallel computation of retention is applied, allowing for fast parallel training. When using the model to solve MOCO problems, a recurrent computation of retention is applied, enabling quicker problem-solving. In order to make our model more practical and flexible, a preference-based retention decoder is proposed, which allows generating approximate Pareto solutions for any trade-off preferences directly. An industry-standard deep reinforcement learning algorithm is used to train RM-MOCO. Experimental results show that, while ensuring the quality of problem solving, the proposed method significantly outperforms some other methods in terms of the speed of solving MOCO problems.
{"title":"RM-MOCO: A Fast-Solving Model for Neural Multi-Objective Combinatorial Optimization Based on Retention","authors":"Huiqing Wei;Fei Han;Qing Liu;Henry Han","doi":"10.23919/CSMS.2024.0029","DOIUrl":"https://doi.org/10.23919/CSMS.2024.0029","url":null,"abstract":"Multiobjective combinatorial optimization (MOCO) problems have a wide range of applications in the real world. Recently, learning-based methods have achieved good results in solving MOCO problems. However, most of these methods use attention mechanisms and their variants, which have room for further improvement in the speed of solving MOCO problems. In this paper, following the idea of decomposition strategy and neural combinatorial optimization, a novel fast-solving model for MOCO based on retention is proposed. A brand new calculation of retention is proposed, causal masking and exponential decay are deprecated in retention, so that our model could better solve MOCO problems. During model training, a parallel computation of retention is applied, allowing for fast parallel training. When using the model to solve MOCO problems, a recurrent computation of retention is applied, enabling quicker problem-solving. In order to make our model more practical and flexible, a preference-based retention decoder is proposed, which allows generating approximate Pareto solutions for any trade-off preferences directly. An industry-standard deep reinforcement learning algorithm is used to train RM-MOCO. Experimental results show that, while ensuring the quality of problem solving, the proposed method significantly outperforms some other methods in terms of the speed of solving MOCO problems.","PeriodicalId":65786,"journal":{"name":"复杂系统建模与仿真(英文)","volume":"5 2","pages":"125-137"},"PeriodicalIF":0.0,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10969587","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143937969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Process control systems typically comprise multiple variables that impact system parameters, interacting and constraining each other with the objective of maintaining parameters within a specified range to ensure dynamic equilibrium. The reliability and safety of these systems are also of paramount concern. The multi-factor Balanced Feedback Net (BFN) represents one of the significant models for simulating systems with balanced feedback mechanisms. This paper refines and expands the definition of BFN, creating BFN models that incorporate varying quantities of balance factors. Based on this, we analyze the structural properties of BFN and provide relevant proofs. Given the complexity of BFN modeling, this paper introduces an innovative approach to translating real-world process control systems into BFN and develops an algorithm to assist users in automatically constructing BFN. For extreme situations that may arise in process control systems, we present an early recognition algorithm for extreme states in BFN. Theoretical proofs and case analyses complement each other, as demonstrated by the example of the water level control system of a steam boiler. This illustrates the effectiveness of methods and algorithms in complex system control and optimization.
{"title":"Modeling and Analyzing Multi-Factor Balanced Feedback System Based on Petri Net","authors":"Yumeng Cheng;Wangyang Yu;Xianwen Fang;Qi Guo;Beiming Zhang","doi":"10.23919/CSMS.2024.0035","DOIUrl":"https://doi.org/10.23919/CSMS.2024.0035","url":null,"abstract":"Process control systems typically comprise multiple variables that impact system parameters, interacting and constraining each other with the objective of maintaining parameters within a specified range to ensure dynamic equilibrium. The reliability and safety of these systems are also of paramount concern. The multi-factor Balanced Feedback Net (BFN) represents one of the significant models for simulating systems with balanced feedback mechanisms. This paper refines and expands the definition of BFN, creating BFN models that incorporate varying quantities of balance factors. Based on this, we analyze the structural properties of BFN and provide relevant proofs. Given the complexity of BFN modeling, this paper introduces an innovative approach to translating real-world process control systems into BFN and develops an algorithm to assist users in automatically constructing BFN. For extreme situations that may arise in process control systems, we present an early recognition algorithm for extreme states in BFN. Theoretical proofs and case analyses complement each other, as demonstrated by the example of the water level control system of a steam boiler. This illustrates the effectiveness of methods and algorithms in complex system control and optimization.","PeriodicalId":65786,"journal":{"name":"复杂系统建模与仿真(英文)","volume":"5 4","pages":"323-339"},"PeriodicalIF":0.0,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10969578","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145705896","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：