Pub Date : 2001-06-27DOI: 10.1109/SMCIA.2001.936730
M. Dotoli, G. Maione, D. Naso, B. Turchiano
This paper describes the application of genetic algorithms (GA) to identify a class of nonlinear SISO models composed of a memoryless nonlinearity in series with a linear transfer function. In contrast with recent literature on the considered problem, we encode in the chromosomes also the structure of the model (type of nonlinearity, number of zeros and poles), and use the GA to identify both the optimal structure and the associated parameters. New operators for mutation and crossover to deal with chromosomes with variable length are introduced. The effectiveness of the approach is tested on a set of case studies derived from literature.
{"title":"Genetic identification of dynamical systems with static nonlinearities","authors":"M. Dotoli, G. Maione, D. Naso, B. Turchiano","doi":"10.1109/SMCIA.2001.936730","DOIUrl":"https://doi.org/10.1109/SMCIA.2001.936730","url":null,"abstract":"This paper describes the application of genetic algorithms (GA) to identify a class of nonlinear SISO models composed of a memoryless nonlinearity in series with a linear transfer function. In contrast with recent literature on the considered problem, we encode in the chromosomes also the structure of the model (type of nonlinearity, number of zeros and poles), and use the GA to identify both the optimal structure and the associated parameters. New operators for mutation and crossover to deal with chromosomes with variable length are introduced. The effectiveness of the approach is tested on a set of case studies derived from literature.","PeriodicalId":104202,"journal":{"name":"SMCia/01. Proceedings of the 2001 IEEE Mountain Workshop on Soft Computing in Industrial Applications (Cat. No.01EX504)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123828137","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}
Pub Date : 2001-06-25DOI: 10.1109/SMCIA.2001.936732
J.R. Thomas, M. Embrechts, R.M. Stringfield, R. M. Wheat
Thermally-induced Doppler broadening of neutron absorption resonances can be used as a unique signature of the temperature of individual isotopes in a mixture. This principle can be exploited for temperature measurements in situations where conventional measurement techniques are not available, such as measurement of temperatures of individual parts of a system in a severe environment, or of components selectively heated by chemical, electromagnetic, or nuclear processes. Interpretation of the broadened absorption data is normally done by comparison to a nuclear physics model of the absorption process. This paper reports a study of the feasibility of interpreting the data with a trained neural network model.
{"title":"Neural network analysis of Doppler-broadened neutron absorption resonance data","authors":"J.R. Thomas, M. Embrechts, R.M. Stringfield, R. M. Wheat","doi":"10.1109/SMCIA.2001.936732","DOIUrl":"https://doi.org/10.1109/SMCIA.2001.936732","url":null,"abstract":"Thermally-induced Doppler broadening of neutron absorption resonances can be used as a unique signature of the temperature of individual isotopes in a mixture. This principle can be exploited for temperature measurements in situations where conventional measurement techniques are not available, such as measurement of temperatures of individual parts of a system in a severe environment, or of components selectively heated by chemical, electromagnetic, or nuclear processes. Interpretation of the broadened absorption data is normally done by comparison to a nuclear physics model of the absorption process. This paper reports a study of the feasibility of interpreting the data with a trained neural network model.","PeriodicalId":104202,"journal":{"name":"SMCia/01. Proceedings of the 2001 IEEE Mountain Workshop on Soft Computing in Industrial Applications (Cat. No.01EX504)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130581889","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}
Pub Date : 2001-06-25DOI: 10.1109/SMCIA.2001.936733
Hossam Meshref, H. Vanlandingham
The field of robotics is an important application area for artificial immune systems. We can make use of the immune system properties to control and to identify complex or even unknown systems. Much research has been done in studying the dynamics of the autonomous mobile robot with particular interest in obstacle avoidance and/or trajectory following. Our research is concerned with robot arm manipulator trajectory planning which is based on sensor-based reactive control. In order to control a robot, it must be kinematically analyzed and the result of this analysis entered into the controller of the robot. Kinematics analysis starts with calculating the forward kinematics solution (FKS). The FKS is a prerequisite for obtaining the inverse kinematics solution (IKS) that is entered into the robot controller and forms the basis of the remote control of the robot. While calculating the IKS, there are many possible poses that links of that manipulator can take to reach the designated point in the robot space. An important question, when a manipulator is dealing with a real dynamically changing environment, is what is the best pose that will get to the target point in the shortest time. We use one of the features of the immune system to build a network of manipulator joints that will interact together to reach the desired target point within the shortest time. The use of an artificial immune network simulator in this research is to play the role of prediction for robot behavior in the new environment.
{"title":"Immune network simulation of reactive control of a robot arm manipulator","authors":"Hossam Meshref, H. Vanlandingham","doi":"10.1109/SMCIA.2001.936733","DOIUrl":"https://doi.org/10.1109/SMCIA.2001.936733","url":null,"abstract":"The field of robotics is an important application area for artificial immune systems. We can make use of the immune system properties to control and to identify complex or even unknown systems. Much research has been done in studying the dynamics of the autonomous mobile robot with particular interest in obstacle avoidance and/or trajectory following. Our research is concerned with robot arm manipulator trajectory planning which is based on sensor-based reactive control. In order to control a robot, it must be kinematically analyzed and the result of this analysis entered into the controller of the robot. Kinematics analysis starts with calculating the forward kinematics solution (FKS). The FKS is a prerequisite for obtaining the inverse kinematics solution (IKS) that is entered into the robot controller and forms the basis of the remote control of the robot. While calculating the IKS, there are many possible poses that links of that manipulator can take to reach the designated point in the robot space. An important question, when a manipulator is dealing with a real dynamically changing environment, is what is the best pose that will get to the target point in the shortest time. We use one of the features of the immune system to build a network of manipulator joints that will interact together to reach the desired target point within the shortest time. The use of an artificial immune network simulator in this research is to play the role of prediction for robot behavior in the new environment.","PeriodicalId":104202,"journal":{"name":"SMCia/01. Proceedings of the 2001 IEEE Mountain Workshop on Soft Computing in Industrial Applications (Cat. No.01EX504)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121678101","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}
Pub Date : 2001-06-25DOI: 10.1109/SMCIA.2001.936722
J. Martikainen, J. Tanskanen, Xiao-Zhi Gao, S. Ovaska
The Internet provides ever increasing ways to communicate and spread information. The Institute of Intelligent Power Electronics at the Helsinki University of Technology organized an online conference during fall 2000. This new type of conference proved to be a success, but a lot more could be achieved. We sum up the experience gathered from organizing WSC5.
{"title":"Organizing an online soft computing conference: a case study","authors":"J. Martikainen, J. Tanskanen, Xiao-Zhi Gao, S. Ovaska","doi":"10.1109/SMCIA.2001.936722","DOIUrl":"https://doi.org/10.1109/SMCIA.2001.936722","url":null,"abstract":"The Internet provides ever increasing ways to communicate and spread information. The Institute of Intelligent Power Electronics at the Helsinki University of Technology organized an online conference during fall 2000. This new type of conference proved to be a success, but a lot more could be achieved. We sum up the experience gathered from organizing WSC5.","PeriodicalId":104202,"journal":{"name":"SMCia/01. Proceedings of the 2001 IEEE Mountain Workshop on Soft Computing in Industrial Applications (Cat. No.01EX504)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124492189","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}
Pub Date : 2001-06-25DOI: 10.1109/SMCIA.2001.936734
S. Satoh, M. S. Shaikh, Y. Dote
We compare empirically the performance of nonlinear radial basis function neural networks (RBFN) and time delay neural networks (TDNN) in accuracy and speed for fault detection in rotational machine parts. We use the advantageous general parameter (GP) approach for initializing the weights of the RBFN model in the beginning of the offline system identification phase, as well as for fine-tuning the modeling accuracy of RBFN. The GP-RBFN scheme is adaptive but still computationally efficient due to the single adaptive parameter and its simple learning rule. The fault measure is the moving average of a general parameter. In order to verify the performance of the proposed schemes, they are applied to fault detection of automobile transmission gears. As the acoustic time series is slightly nonlinear, the RBFN gives high-speed fault detection, but detection accuracy is not so high. To overcome this problem a TDNN is developed that achieves more accurate fault detection although it needs more computational time. A fault is detected through regression lines. Both methods are empirically compared in speed and accuracy for fault detection of automobile transmission gears.
{"title":"Fast fuzzy neural network for fault diagnosis of rotational machine parts using general parameter learning and adaptation","authors":"S. Satoh, M. S. Shaikh, Y. Dote","doi":"10.1109/SMCIA.2001.936734","DOIUrl":"https://doi.org/10.1109/SMCIA.2001.936734","url":null,"abstract":"We compare empirically the performance of nonlinear radial basis function neural networks (RBFN) and time delay neural networks (TDNN) in accuracy and speed for fault detection in rotational machine parts. We use the advantageous general parameter (GP) approach for initializing the weights of the RBFN model in the beginning of the offline system identification phase, as well as for fine-tuning the modeling accuracy of RBFN. The GP-RBFN scheme is adaptive but still computationally efficient due to the single adaptive parameter and its simple learning rule. The fault measure is the moving average of a general parameter. In order to verify the performance of the proposed schemes, they are applied to fault detection of automobile transmission gears. As the acoustic time series is slightly nonlinear, the RBFN gives high-speed fault detection, but detection accuracy is not so high. To overcome this problem a TDNN is developed that achieves more accurate fault detection although it needs more computational time. A fault is detected through regression lines. Both methods are empirically compared in speed and accuracy for fault detection of automobile transmission gears.","PeriodicalId":104202,"journal":{"name":"SMCia/01. Proceedings of the 2001 IEEE Mountain Workshop on Soft Computing in Industrial Applications (Cat. No.01EX504)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124229915","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}
Pub Date : 2001-06-25DOI: 10.1109/SMCIA.2001.936737
G. Maione, D. Naso
We describe the control system of a manufacturing plant as a network of atomic agents (discrete event controllers) of different kinds. Since agents of the same type do not interact, the control leads to low programming effort and data exchange but, at the same time, it can provide poor performance of the plant. To overcome this drawback, we introduce a discrete event supervisor which modifies the decision laws of the members of the agent population, according to a genetic evolutionary mechanism. Simulation shows that, due to the supervisor action, the network of agents can efficiently face a variety of disturbances.
{"title":"Modeling evolutionary supervisors for multi-agent manufacturing control with discrete event formalism","authors":"G. Maione, D. Naso","doi":"10.1109/SMCIA.2001.936737","DOIUrl":"https://doi.org/10.1109/SMCIA.2001.936737","url":null,"abstract":"We describe the control system of a manufacturing plant as a network of atomic agents (discrete event controllers) of different kinds. Since agents of the same type do not interact, the control leads to low programming effort and data exchange but, at the same time, it can provide poor performance of the plant. To overcome this drawback, we introduce a discrete event supervisor which modifies the decision laws of the members of the agent population, according to a genetic evolutionary mechanism. Simulation shows that, due to the supervisor action, the network of agents can efficiently face a variety of disturbances.","PeriodicalId":104202,"journal":{"name":"SMCia/01. Proceedings of the 2001 IEEE Mountain Workshop on Soft Computing in Industrial Applications (Cat. No.01EX504)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133081555","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}
Pub Date : 2001-06-25DOI: 10.1109/SMCIA.2001.936729
C. F. Jones, M. Christman
This paper presents a set of experiments using a genetic algorithm to develop solutions to a Vigenere alphabetic code. A Vigenere code replaces each character in the plain text with a new character generated by adding the value of a character in the corresponding place in a "keyword". The genetic algorithm uses the number of characters in valid English words as the measure of a solution's fitness; breeding is accomplished by simple crossover and mutation consists of random variations of a single keyword position. To accelerate the convergence of the solution, we introduce a new concept: anti-elitism, or including a small portion of the worst solutions as well as the best. This has the effect of retaining mutations that may not initially benefit the best solution, but will later breed with other individuals in the population to increase the maximum fitness.
{"title":"Genetic algorithm solution of Vigenere alphabetic codes","authors":"C. F. Jones, M. Christman","doi":"10.1109/SMCIA.2001.936729","DOIUrl":"https://doi.org/10.1109/SMCIA.2001.936729","url":null,"abstract":"This paper presents a set of experiments using a genetic algorithm to develop solutions to a Vigenere alphabetic code. A Vigenere code replaces each character in the plain text with a new character generated by adding the value of a character in the corresponding place in a \"keyword\". The genetic algorithm uses the number of characters in valid English words as the measure of a solution's fitness; breeding is accomplished by simple crossover and mutation consists of random variations of a single keyword position. To accelerate the convergence of the solution, we introduce a new concept: anti-elitism, or including a small portion of the worst solutions as well as the best. This has the effect of retaining mutations that may not initially benefit the best solution, but will later breed with other individuals in the population to increase the maximum fitness.","PeriodicalId":104202,"journal":{"name":"SMCia/01. Proceedings of the 2001 IEEE Mountain Workshop on Soft Computing in Industrial Applications (Cat. No.01EX504)","volume":"175 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124318424","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}
Pub Date : 2001-06-25DOI: 10.1109/SMCIA.2001.936725
K. Yasuda, M. Tsunematsu
A robust generation mix is defined explicitly as a mix that assures a fixed performance (economy) despite deviations in these factors. In addition, a new formulation of the robust generation mix problem whose purpose is to determine the most robust combination of plants in a target year is proposed. It is generally difficult to solve the formulated problem with parameters. The simulation results on a typical model show that the generation mix produced by the proposed formulation is more robust than that produced by the conventional formulation.
{"title":"Robust generation mix under uncertain environments in electric power systems","authors":"K. Yasuda, M. Tsunematsu","doi":"10.1109/SMCIA.2001.936725","DOIUrl":"https://doi.org/10.1109/SMCIA.2001.936725","url":null,"abstract":"A robust generation mix is defined explicitly as a mix that assures a fixed performance (economy) despite deviations in these factors. In addition, a new formulation of the robust generation mix problem whose purpose is to determine the most robust combination of plants in a target year is proposed. It is generally difficult to solve the formulated problem with parameters. The simulation results on a typical model show that the generation mix produced by the proposed formulation is more robust than that produced by the conventional formulation.","PeriodicalId":104202,"journal":{"name":"SMCia/01. Proceedings of the 2001 IEEE Mountain Workshop on Soft Computing in Industrial Applications (Cat. No.01EX504)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128831085","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}
Pub Date : 2001-06-25DOI: 10.1109/SMCIA.2001.936738
E. Dummermuth
Fuzzy logic has been used to create nonlinear transfer functions for one or more input variables. For example, an input variable e and its derivative delta e may be used in conjunction with rules and membership functions to create a non-linear PD controller. Integrating the result yields a non-linear fuzzy PI controller. Under the assumption that monotonic behavior is needed for a stable control, the nonlinear PI may be implemented much simpler and more predictable with algebraic equations. This technique allows to adjust the gain of the P and the I portions separately while maintaining conventional stable conditions.
{"title":"A non-linear tunable PI block for improvement of closed-loop control response","authors":"E. Dummermuth","doi":"10.1109/SMCIA.2001.936738","DOIUrl":"https://doi.org/10.1109/SMCIA.2001.936738","url":null,"abstract":"Fuzzy logic has been used to create nonlinear transfer functions for one or more input variables. For example, an input variable e and its derivative delta e may be used in conjunction with rules and membership functions to create a non-linear PD controller. Integrating the result yields a non-linear fuzzy PI controller. Under the assumption that monotonic behavior is needed for a stable control, the nonlinear PI may be implemented much simpler and more predictable with algebraic equations. This technique allows to adjust the gain of the P and the I portions separately while maintaining conventional stable conditions.","PeriodicalId":104202,"journal":{"name":"SMCia/01. Proceedings of the 2001 IEEE Mountain Workshop on Soft Computing in Industrial Applications (Cat. No.01EX504)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127532815","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}
Pub Date : 2001-06-25DOI: 10.1109/SMCIA.2001.936731
S. Laine
This paper presents how process control problems can be studied and solved by combining offline and online information. Offline information is accurate and versatile, it is used to define the problem. Online information describes the state of the process in real time. The online variables containing information of the offline defined problem are selected using a variable selection algorithm. These variables are used to create an online observer of the problem. This observer can be used to solve process control problems. The main algorithms used in this paper are the variable selection technique and the self-organizing map (SOM). The methodology is illustrated using the case of the concentrator of the Outokumpu Hitura mine.
{"title":"Combining off-line and on-line information in process study using the self-organizing map (SOM)","authors":"S. Laine","doi":"10.1109/SMCIA.2001.936731","DOIUrl":"https://doi.org/10.1109/SMCIA.2001.936731","url":null,"abstract":"This paper presents how process control problems can be studied and solved by combining offline and online information. Offline information is accurate and versatile, it is used to define the problem. Online information describes the state of the process in real time. The online variables containing information of the offline defined problem are selected using a variable selection algorithm. These variables are used to create an online observer of the problem. This observer can be used to solve process control problems. The main algorithms used in this paper are the variable selection technique and the self-organizing map (SOM). The methodology is illustrated using the case of the concentrator of the Outokumpu Hitura mine.","PeriodicalId":104202,"journal":{"name":"SMCia/01. Proceedings of the 2001 IEEE Mountain Workshop on Soft Computing in Industrial Applications (Cat. No.01EX504)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128728717","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：