In high-mix low-volume manufacturing and agile manufacturing, flexibility to cope with fluctuations in production conditions is a key for competitive production. To realize the flexible manufacturing, it is required to immediately generate, evaluate, modify and determine a production schedule which is suitable for the new condition. Decentralized methods of scheduling and simulation for the immediate generation and evaluation have been developed based on a concept of highly-distributed manufacturing systems in which each machine is computerized and can communicate with other machines. Although these methods are useful for performing the flexible manufacturing, it is possible that the original schedule is good even for the new production condition and performing rescheduling is in vain. This research aims at supporting decision-making on rescheduling for realizing lean flexible manufacturing in which rescheduling is performed only when it is effective. In this paper, delay in operation due to inaccurate estimation of processing time is considered, and the following method is proposed: When it turns out completion of an operation delays and the amount of the delay is estimated, the original schedule is repaired by the right-shift policy. The repaired schedule is then analyzed using simple criteria for judging whether rescheduling should be executed or not. Several criteria focused on critical path were given, and the effectiveness of this support method was shown by numerical experiments.
{"title":"Method for Supporting Decision-Making in Rescheduling for Lean Flexible Manufacturing","authors":"Eiji Morinaga, Daiki Yasuda, Hidefumi Wakamatsu","doi":"10.5687/iscie.36.271","DOIUrl":"https://doi.org/10.5687/iscie.36.271","url":null,"abstract":"In high-mix low-volume manufacturing and agile manufacturing, flexibility to cope with fluctuations in production conditions is a key for competitive production. To realize the flexible manufacturing, it is required to immediately generate, evaluate, modify and determine a production schedule which is suitable for the new condition. Decentralized methods of scheduling and simulation for the immediate generation and evaluation have been developed based on a concept of highly-distributed manufacturing systems in which each machine is computerized and can communicate with other machines. Although these methods are useful for performing the flexible manufacturing, it is possible that the original schedule is good even for the new production condition and performing rescheduling is in vain. This research aims at supporting decision-making on rescheduling for realizing lean flexible manufacturing in which rescheduling is performed only when it is effective. In this paper, delay in operation due to inaccurate estimation of processing time is considered, and the following method is proposed: When it turns out completion of an operation delays and the amount of the delay is estimated, the original schedule is repaired by the right-shift policy. The repaired schedule is then analyzed using simple criteria for judging whether rescheduling should be executed or not. Several criteria focused on critical path were given, and the effectiveness of this support method was shown by numerical experiments.","PeriodicalId":489536,"journal":{"name":"Shisutemu Seigyo Jōhō Gakkai ronbunshi","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135165161","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As product systems become larger and more complex, their design space increases, making optimization more difficult. In the past, hierarchical optimization methods have been proposed to solve this problem. However, they are ineffective in difficult cases where subsystems are strongly coupled. Therefore, we focused on optimal solutions using reinforcement learning. However, for large-scale optimization problems, the learning space increases, and optimization becomes difficult. Therefore, we consider subsystem optimizers as agents and propose an algorithm that mitigates the disadvantages of reducing the learning space of reinforcement learning through negotiation between agents. Finally, the proposed method successfully reduces the number of evaluations to derive the optimal solution to less than 10% of the previous one, while maintaining the quality of the optimization solution, by increasing the efficiency of learning from 2.1% to 28.1% by reducing the learning space.
{"title":"Hierarchical Optimization of Product System Using Reinforcement Learning","authors":"Yoshiharu Iwata, Haruhi Kajisaki, Kouji Fujishiro, Hidefumi Wakamatsu, Tomoki Takao","doi":"10.5687/iscie.36.251","DOIUrl":"https://doi.org/10.5687/iscie.36.251","url":null,"abstract":"As product systems become larger and more complex, their design space increases, making optimization more difficult. In the past, hierarchical optimization methods have been proposed to solve this problem. However, they are ineffective in difficult cases where subsystems are strongly coupled. Therefore, we focused on optimal solutions using reinforcement learning. However, for large-scale optimization problems, the learning space increases, and optimization becomes difficult. Therefore, we consider subsystem optimizers as agents and propose an algorithm that mitigates the disadvantages of reducing the learning space of reinforcement learning through negotiation between agents. Finally, the proposed method successfully reduces the number of evaluations to derive the optimal solution to less than 10% of the previous one, while maintaining the quality of the optimization solution, by increasing the efficiency of learning from 2.1% to 28.1% by reducing the learning space.","PeriodicalId":489536,"journal":{"name":"Shisutemu Seigyo Jōhō Gakkai ronbunshi","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135165163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Bilateral control via the communication network is one of the effective methods to realize the teleoperation of robot systems. The general communication line such as the Internet system is with random transmission delay and data loss on the communication path. These communication failures cause instability of the control system for the bilateral robot system and make it difficult to accurately deliver the skills of the operator to remote locations. When controlling a flexible structure, the above-mentioned effects are prominent, such as the generation of vibration. In this research, we investigated the effects of transmission delays and data loss in control systems for flexible structures. The purpose of this research is to reduce the undesirable effect due to the transmission delay and data loss of the communication network with respect to the control system by introducing a switchable compensator. This compensator determines the presence of data loss and compensates the data loss using the past data in the system was designed. Furthermore, in the numerical simulations, the effectiveness of the proposed system was demonstrated. However, it was found that when the data loss rate increased, the amount of overshoot occurred, and it was found that improvement was necessary.
{"title":"Bilateral Control for Stochastic Flexible Master-Slave System with Communication Failures","authors":"Toshiyuki Nagata, Yuichi Sawada","doi":"10.5687/iscie.36.260","DOIUrl":"https://doi.org/10.5687/iscie.36.260","url":null,"abstract":"Bilateral control via the communication network is one of the effective methods to realize the teleoperation of robot systems. The general communication line such as the Internet system is with random transmission delay and data loss on the communication path. These communication failures cause instability of the control system for the bilateral robot system and make it difficult to accurately deliver the skills of the operator to remote locations. When controlling a flexible structure, the above-mentioned effects are prominent, such as the generation of vibration. In this research, we investigated the effects of transmission delays and data loss in control systems for flexible structures. The purpose of this research is to reduce the undesirable effect due to the transmission delay and data loss of the communication network with respect to the control system by introducing a switchable compensator. This compensator determines the presence of data loss and compensates the data loss using the past data in the system was designed. Furthermore, in the numerical simulations, the effectiveness of the proposed system was demonstrated. However, it was found that when the data loss rate increased, the amount of overshoot occurred, and it was found that improvement was necessary.","PeriodicalId":489536,"journal":{"name":"Shisutemu Seigyo Jōhō Gakkai ronbunshi","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135165164","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The construction of approximators for simulations, such as the finite element method using machine learning, has the problem of both reducing training data generation time and achieving approximation accuracy. Hybrid neural networks have been proposed to solve this problem as a fast approximator for simulation. Even if built with a simple perceptron with a linear activation function created based on deductive knowledge using conventional approximation techniques such as multiple regression analysis, the number of phenomena that can be modeled by deductive knowledge is limited in simulations of complex structures. As a result, there will be errors in the predictions of the approximators. In contrast, hybrid neural networks allow neural networks to learn errors in predictions to create correction approximators, allowing approximators to account for effects that cannot be expressed by multiple regression analysis. This paper proposes a neural network with a structure that integrates these approximators. The first proposed Hybrid Neural Network (HNN) approximator trains a linear approximator, and then a nonlinear approximator learns the error part. In contrast, the Integration Neural Network (INN) simultaneously learns the linear and nonlinear approximators to optimize the learning ratio by training. This method allows INNs to improve the accuracy of approximators and reduce the conflict between the number of training data and accuracy.
{"title":"Simulation Approximators Using Linear and Nonlinear Integration Neural Networks","authors":"Yoshiharu Iwata, Kouji Fujishiro, Hidefumi Wakamatsu","doi":"10.5687/iscie.36.243","DOIUrl":"https://doi.org/10.5687/iscie.36.243","url":null,"abstract":"The construction of approximators for simulations, such as the finite element method using machine learning, has the problem of both reducing training data generation time and achieving approximation accuracy. Hybrid neural networks have been proposed to solve this problem as a fast approximator for simulation. Even if built with a simple perceptron with a linear activation function created based on deductive knowledge using conventional approximation techniques such as multiple regression analysis, the number of phenomena that can be modeled by deductive knowledge is limited in simulations of complex structures. As a result, there will be errors in the predictions of the approximators. In contrast, hybrid neural networks allow neural networks to learn errors in predictions to create correction approximators, allowing approximators to account for effects that cannot be expressed by multiple regression analysis. This paper proposes a neural network with a structure that integrates these approximators. The first proposed Hybrid Neural Network (HNN) approximator trains a linear approximator, and then a nonlinear approximator learns the error part. In contrast, the Integration Neural Network (INN) simultaneously learns the linear and nonlinear approximators to optimize the learning ratio by training. This method allows INNs to improve the accuracy of approximators and reduce the conflict between the number of training data and accuracy.","PeriodicalId":489536,"journal":{"name":"Shisutemu Seigyo Jōhō Gakkai ronbunshi","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135165160","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents some stability remarks on the discretized diffusion process models. The motivation arises from our observation of stability preserving property in terms of a model reduction procedure of the 1D model. Before the stability analysis of the non-reduced order model, the state-space model is extended to multi-dimensional cases in a systematic manner. This formulation and the corresponding stability analysis are the first non-trivial contributions here. Then we clarify the fact behind the stability preserving property. As a consequence, one can employ arbitrary size of reduced order model based on the techniques such as the principal component analysis without any concern for the stability. The power of this reduction method is demonstrated via numerical examples for the 1D and 2D cases.
{"title":"Stability Remarks on Discretized Multi-dimensional Diffusion Process Models and its Application to Model Reduction","authors":"Weiqi Zhang, Kentaro Hirata, Yukinori Nakamura, Kunihisa Okano","doi":"10.5687/iscie.36.279","DOIUrl":"https://doi.org/10.5687/iscie.36.279","url":null,"abstract":"This paper presents some stability remarks on the discretized diffusion process models. The motivation arises from our observation of stability preserving property in terms of a model reduction procedure of the 1D model. Before the stability analysis of the non-reduced order model, the state-space model is extended to multi-dimensional cases in a systematic manner. This formulation and the corresponding stability analysis are the first non-trivial contributions here. Then we clarify the fact behind the stability preserving property. As a consequence, one can employ arbitrary size of reduced order model based on the techniques such as the principal component analysis without any concern for the stability. The power of this reduction method is demonstrated via numerical examples for the 1D and 2D cases.","PeriodicalId":489536,"journal":{"name":"Shisutemu Seigyo Jōhō Gakkai ronbunshi","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135165159","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A five-axis machining center is known for its synchronous control capability, which allows complicated three-dimensional surfaces, such as propellers and hypoid gears, to be created rapidly. In this study, we aim to maintain the feed speed vector at the end milling point by controlling two linear axes and a rotary axis with a five-axis machining center to improve the machined surface quality. Previously, we suggested reducing the shape error of machined workpieces by implementing a parameter (referred to as the precedent control coefficient herein) to reduce the differences in the servo characteristics of three axes in the machining method. Moreover, to maintain the feed speed vector at the end milling point when machining complex shapes, a rapid velocity change in each axis is required, which results in inaccuracy due to torque saturation. In this study, to reduce the shape error while avoiding torque saturation during high angular velocity movements, we develop a theoretical method to obtain the suitable precedent control coefficient of each axis based on a block diagram that accounts for torque saturation. The developed method allows shape error reduction and torque saturation avoidance to be realized.
{"title":"Improved Method for Synchronous Accuracy of Linear and Rotary Axes under a Constant Feed Speed Vector at the End Milling Point while Avoiding Torque Saturation","authors":"Takamaru Suzuki, Toshiki Hirogaki, Eiichi Aoyama","doi":"10.5687/iscie.36.233","DOIUrl":"https://doi.org/10.5687/iscie.36.233","url":null,"abstract":"A five-axis machining center is known for its synchronous control capability, which allows complicated three-dimensional surfaces, such as propellers and hypoid gears, to be created rapidly. In this study, we aim to maintain the feed speed vector at the end milling point by controlling two linear axes and a rotary axis with a five-axis machining center to improve the machined surface quality. Previously, we suggested reducing the shape error of machined workpieces by implementing a parameter (referred to as the precedent control coefficient herein) to reduce the differences in the servo characteristics of three axes in the machining method. Moreover, to maintain the feed speed vector at the end milling point when machining complex shapes, a rapid velocity change in each axis is required, which results in inaccuracy due to torque saturation. In this study, to reduce the shape error while avoiding torque saturation during high angular velocity movements, we develop a theoretical method to obtain the suitable precedent control coefficient of each axis based on a block diagram that accounts for torque saturation. The developed method allows shape error reduction and torque saturation avoidance to be realized.","PeriodicalId":489536,"journal":{"name":"Shisutemu Seigyo Jōhō Gakkai ronbunshi","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135165162","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this study, the first step we investigate the bifurcation phenomenon in a current-controlled step-up DC–DC converter with photovoltaic (PV) module using a one-dimensional circuit model. Specifically, we analyze the occurrence of the bifurcation phenomenon in this circuit model by defining the output voltage of the converter as constant and utilizing a PV module nonlinear output model with a relatively small number of parameters. Our results reveal the existence of a unique bifurcation structure in DC–DC converters with PV module, which has not been previously reported. We further elucidate the mechanism of occurrence through a stability analysis based on the brute-force method. The findings of this study contribute significantly to the development of circuit and bifurcation theory for current-controlled step-up DC–DC converters with PV module.
{"title":"A Unique Bifurcation Structure in One-Dimensional Model of Current-Controlled Step-Up DC–DC Converter with Photovoltaic Module","authors":"Shota Uchino, Satoshi Aoki, Takuji Kousaka, Hiroyuki Asahara","doi":"10.5687/iscie.36.155","DOIUrl":"https://doi.org/10.5687/iscie.36.155","url":null,"abstract":"In this study, the first step we investigate the bifurcation phenomenon in a current-controlled step-up DC–DC converter with photovoltaic (PV) module using a one-dimensional circuit model. Specifically, we analyze the occurrence of the bifurcation phenomenon in this circuit model by defining the output voltage of the converter as constant and utilizing a PV module nonlinear output model with a relatively small number of parameters. Our results reveal the existence of a unique bifurcation structure in DC–DC converters with PV module, which has not been previously reported. We further elucidate the mechanism of occurrence through a stability analysis based on the brute-force method. The findings of this study contribute significantly to the development of circuit and bifurcation theory for current-controlled step-up DC–DC converters with PV module.","PeriodicalId":489536,"journal":{"name":"Shisutemu Seigyo Jōhō Gakkai ronbunshi","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135859845","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we investigate the effect of norm selection in the scenario approach to robust feasibility problems. A robust feasibility problem can be converted into a robust optimization problem by minimizing a fictitious cost function, and the scenario approach is a method to find a stochastic approximation of the optimal solution in an affordable computational time. When considering a norm of the decision variable as the fictitious cost function, the choice of the norm may affect the robustness of the solution obtained by the scenario approach. To demonstrate the effect of norm selection, we compare the 2-, ∞-, and 1-norms by simulating two illustrative examples. Moreover, the effectiveness of the ∞- and 1-norms compared to the 2-norm is demonstrated in a numerical example of a finite-horizon optimal control of uncertain linear systems.
{"title":"Comparison of Norm Minimizations in Scenario Approach and Its Application to Robust Optimal Control","authors":"Tomoyuki Iori, Yasumasa Fujisaki","doi":"10.5687/iscie.36.163","DOIUrl":"https://doi.org/10.5687/iscie.36.163","url":null,"abstract":"In this paper, we investigate the effect of norm selection in the scenario approach to robust feasibility problems. A robust feasibility problem can be converted into a robust optimization problem by minimizing a fictitious cost function, and the scenario approach is a method to find a stochastic approximation of the optimal solution in an affordable computational time. When considering a norm of the decision variable as the fictitious cost function, the choice of the norm may affect the robustness of the solution obtained by the scenario approach. To demonstrate the effect of norm selection, we compare the 2-, ∞-, and 1-norms by simulating two illustrative examples. Moreover, the effectiveness of the ∞- and 1-norms compared to the 2-norm is demonstrated in a numerical example of a finite-horizon optimal control of uncertain linear systems.","PeriodicalId":489536,"journal":{"name":"Shisutemu Seigyo Jōhō Gakkai ronbunshi","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135859848","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Kazutoshi Sakakibara, Kosuke Nakata, Jin Matsuzaki, Masaki Nakamura
This paper addresses the scheduling of an operator and a machine in the machining process. While both schedules are reduced to the job-shop type, the operator's completion time is relatively uncertain, so rolling scheduling is required. For problems with these characteristics, we apply relax-and-fix heuristics, in which some decision variables in each rolling are relaxed based on a mixed-integer programming model. Although the relax-and-fix heuristics can reduce the solution cost by mathematical programming, it may significantly reduce the optimality of the derived solution, depending on how the set of variables to be relaxed is selected. Therefore, we incorporate metaheuristics into the relax-and-fix heuristics to always choose the appropriate relaxation variables. The effectiveness of the proposed method is confirmed by applying it to real-world machining scheduling.
{"title":"Rolling Scheduling by Hybrid Optimization Mixing Mathematical Programming and Meta-heuristics","authors":"Kazutoshi Sakakibara, Kosuke Nakata, Jin Matsuzaki, Masaki Nakamura","doi":"10.5687/iscie.36.172","DOIUrl":"https://doi.org/10.5687/iscie.36.172","url":null,"abstract":"This paper addresses the scheduling of an operator and a machine in the machining process. While both schedules are reduced to the job-shop type, the operator's completion time is relatively uncertain, so rolling scheduling is required. For problems with these characteristics, we apply relax-and-fix heuristics, in which some decision variables in each rolling are relaxed based on a mixed-integer programming model. Although the relax-and-fix heuristics can reduce the solution cost by mathematical programming, it may significantly reduce the optimality of the derived solution, depending on how the set of variables to be relaxed is selected. Therefore, we incorporate metaheuristics into the relax-and-fix heuristics to always choose the appropriate relaxation variables. The effectiveness of the proposed method is confirmed by applying it to real-world machining scheduling.","PeriodicalId":489536,"journal":{"name":"Shisutemu Seigyo Jōhō Gakkai ronbunshi","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135859846","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","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：