Pub Date : 2003-12-08DOI: 10.1109/CEC.2003.1299403
D. Corne, Joshua D. Knowles
The No-Free-Lunch (NFL) theorems hold for general multiobjective fitness spaces, in the sense that, over a space of problems which is closed under permutation, any two algorithms will produce the same set of multiobjective samples. However, there are salient ways in which NFL does not generally hold in multiobjective optimization. Previously we have shown that a 'free lunch' can arise when comparative metrics (rather than absolute metrics) are used for performance measurement. Here we show that NFL does not generally apply in multiobjective optimization when absolute performance metrics are used. This is because multiobjective optimizers usually combine a generator with an archiver. The generator corresponds to the 'algorithm' in the NFL sense, but the archiver filters the sample generated by the algorithm in a way that undermines the NFL assumptions. Essentially, if two multiobjective approaches have different archivers, their average performance may differ. We prove this, and hence show that we can say, without qualification, that some multiobjective approaches are better than others.
{"title":"Some multiobjective optimizers are better than others","authors":"D. Corne, Joshua D. Knowles","doi":"10.1109/CEC.2003.1299403","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299403","url":null,"abstract":"The No-Free-Lunch (NFL) theorems hold for general multiobjective fitness spaces, in the sense that, over a space of problems which is closed under permutation, any two algorithms will produce the same set of multiobjective samples. However, there are salient ways in which NFL does not generally hold in multiobjective optimization. Previously we have shown that a 'free lunch' can arise when comparative metrics (rather than absolute metrics) are used for performance measurement. Here we show that NFL does not generally apply in multiobjective optimization when absolute performance metrics are used. This is because multiobjective optimizers usually combine a generator with an archiver. The generator corresponds to the 'algorithm' in the NFL sense, but the archiver filters the sample generated by the algorithm in a way that undermines the NFL assumptions. Essentially, if two multiobjective approaches have different archivers, their average performance may differ. We prove this, and hence show that we can say, without qualification, that some multiobjective approaches are better than others.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134278295","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299569
Y. Takahashi, Y. Kawano, M. Kitagawa
We investigate a class QNC/sup 0/ (ADD) that is QNC/sup 0/ with gates for addition of two binary numbers, where QNC/sup 0/ is a class consisting of quantum operations computed by constant-depth quantum circuits. We show that QNC/sup 0/(ADD) = QNC/sup 0/(PAR), where QNC/sup 0/(PAR) is QNC/sup 0/ with Toffoli gates of arbitrary fan-in and gates for parity. Moreover, we show that QNC/sup 0/(ADD) = QAC/sup 0/(MUL) = QAC/sup 0/(DIV), where QAC/sup 0/(MUL) and QAC/sup 0/(DIV) are QNC/sup 0/ with Toffoli gates of arbitrary fan-in and gates for multiplication and division respectively. In the classical setting, similar relationships do not hold. These relationships suggest that QNC/sup 0/ /spl subne/ QNC/sup 0/(ADD); that is, the use of gates for addition increases the computational power of constant-depth quantum circuits. To prove QNC/sup 0/ /spl subne/ QNC/sup 0/(ADD), we present a characterization of this relationship by the one-wayness of a permutation that is constructed explicitly. We conjecture that the permutation is one-way, which implies QNC/sup 0/ /spl subne/ QNC/sup 0/(ADD).
{"title":"On the computational power of constant-depth quantum circuits with gates for addition","authors":"Y. Takahashi, Y. Kawano, M. Kitagawa","doi":"10.1109/CEC.2003.1299569","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299569","url":null,"abstract":"We investigate a class QNC/sup 0/ (ADD) that is QNC/sup 0/ with gates for addition of two binary numbers, where QNC/sup 0/ is a class consisting of quantum operations computed by constant-depth quantum circuits. We show that QNC/sup 0/(ADD) = QNC/sup 0/(PAR), where QNC/sup 0/(PAR) is QNC/sup 0/ with Toffoli gates of arbitrary fan-in and gates for parity. Moreover, we show that QNC/sup 0/(ADD) = QAC/sup 0/(MUL) = QAC/sup 0/(DIV), where QAC/sup 0/(MUL) and QAC/sup 0/(DIV) are QNC/sup 0/ with Toffoli gates of arbitrary fan-in and gates for multiplication and division respectively. In the classical setting, similar relationships do not hold. These relationships suggest that QNC/sup 0/ /spl subne/ QNC/sup 0/(ADD); that is, the use of gates for addition increases the computational power of constant-depth quantum circuits. To prove QNC/sup 0/ /spl subne/ QNC/sup 0/(ADD), we present a characterization of this relationship by the one-wayness of a permutation that is constructed explicitly. We conjecture that the permutation is one-way, which implies QNC/sup 0/ /spl subne/ QNC/sup 0/(ADD).","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134150781","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299601
R. Thomson, T. Arslan
This paper describes the operation of an evolutionary algorithm (EA) for the creation of linear digital VLSI circuit designs. The EA can produce hardware designs from a behavioural description of a problem. The designs are based upon a library of high-level components. The EA performs a multi-objective search, using models of the longest-path delay and the silicon area of a design. These models are based upon the properties of real-world components, implementable in a 0.18 micron technology. The accuracy of these models is investigated. Two important aspects of multi-objective evolution are the population diversity, and the variability of the results. Both of these areas are examined. The population diversity is assessed in terms of conflict between the objectives, and the robustness of the EA is experimentally investigated.
{"title":"On the impact of modelling, robustness and diversity to the performance of a multi-objective evolutionary algorithm for digital VLSI system design","authors":"R. Thomson, T. Arslan","doi":"10.1109/CEC.2003.1299601","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299601","url":null,"abstract":"This paper describes the operation of an evolutionary algorithm (EA) for the creation of linear digital VLSI circuit designs. The EA can produce hardware designs from a behavioural description of a problem. The designs are based upon a library of high-level components. The EA performs a multi-objective search, using models of the longest-path delay and the silicon area of a design. These models are based upon the properties of real-world components, implementable in a 0.18 micron technology. The accuracy of these models is investigated. Two important aspects of multi-objective evolution are the population diversity, and the variability of the results. Both of these areas are examined. The population diversity is assessed in terms of conflict between the objectives, and the robustness of the EA is experimentally investigated.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124372115","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299821
Marcelo A. A. Araújo, E. Teixeira, Fábio R. Camargo, João P. V. Almeida
We present a peculiar parallel implementation of artificial neural networks using the backpropagation training algorithm. The message pass interface PVM is used in the Linux operating system environment, implemented in a cluster of IBM-PC machines. An optimized object-oriented framework to train neural networks, developed in C++, is part of the system presented. A shared memory framework was implemented to improve the training phase. One of the advantages of the system is the low cost, considering that its performance can be compared to similar powerful parallel machines.
{"title":"Parallel training for neural networks using PVM with shared memory","authors":"Marcelo A. A. Araújo, E. Teixeira, Fábio R. Camargo, João P. V. Almeida","doi":"10.1109/CEC.2003.1299821","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299821","url":null,"abstract":"We present a peculiar parallel implementation of artificial neural networks using the backpropagation training algorithm. The message pass interface PVM is used in the Linux operating system environment, implemented in a cluster of IBM-PC machines. An optimized object-oriented framework to train neural networks, developed in C++, is part of the system presented. A shared memory framework was implemented to improve the training phase. One of the advantages of the system is the low cost, considering that its performance can be compared to similar powerful parallel machines.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130124390","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299550
E. Mezura-Montes, C. Coello
In this paper, we propose the use of a simple evolution strategy (SES) (i.e., a (1 + /spl lambda/)-ES with self-adaptation that uses three tournament rules based on feasibility) coupled with a diversity mechanism to solve constrained optimization problems. The proposed mechanism is based on multiobjective optimization concepts taken from an approach called the niched-Pareto genetic algorithm (NPGA). The main advantage of the proposed approach is that it does not require the definition of any extra parameters, other than those required by an evolution strategy. The performance of the proposed approach is shown to be highly competitive with respect to other constraint-handling techniques representative of the state-of-the-art in the area when using a set of well-known benchmarks.
{"title":"Adding a diversity mechanism to a simple evolution strategy to solve constrained optimization problems","authors":"E. Mezura-Montes, C. Coello","doi":"10.1109/CEC.2003.1299550","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299550","url":null,"abstract":"In this paper, we propose the use of a simple evolution strategy (SES) (i.e., a (1 + /spl lambda/)-ES with self-adaptation that uses three tournament rules based on feasibility) coupled with a diversity mechanism to solve constrained optimization problems. The proposed mechanism is based on multiobjective optimization concepts taken from an approach called the niched-Pareto genetic algorithm (NPGA). The main advantage of the proposed approach is that it does not require the definition of any extra parameters, other than those required by an evolution strategy. The performance of the proposed approach is shown to be highly competitive with respect to other constraint-handling techniques representative of the state-of-the-art in the area when using a set of well-known benchmarks.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124791448","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299879
P. Pratibha, S. Borra, A. Muthukaruppan, S. Suresh, V. Kamakoti
This paper introduces a novel parallel evolutionary methodology making use of ANN for solving the spatial partitioning problem for multi-FPGA (field programmable gate arrays) architectures. The algorithm takes as input a HDL (hardware description language) model of the application along with user specified constraints and automatically generates a task graph G; partitions G based on the user specified constraints and maps the blocks of the partitions onto the different FPGAs in the given multi-FPGA architecture, all in a single-shot. The proposed algorithm was successfully employed to spatially partition a reasonably big cryptographic application that involved a 1024-bit modular exponentiation and to map the same onto a network of nine ACEX1K based Altera EP1K30QC208-1 FPGAs. The suggested parallel evolutionary algorithm for the partitioning step was implemented on a 6-node SGI Origin-2000 platform using the message passing interface (MPI) standard. The results obtained by executing the same are extremely encouraging, especially for larger task graphs.
{"title":"An enhanced evolutionary approach to spatial partitioning for reconfigurable environments","authors":"P. Pratibha, S. Borra, A. Muthukaruppan, S. Suresh, V. Kamakoti","doi":"10.1109/CEC.2003.1299879","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299879","url":null,"abstract":"This paper introduces a novel parallel evolutionary methodology making use of ANN for solving the spatial partitioning problem for multi-FPGA (field programmable gate arrays) architectures. The algorithm takes as input a HDL (hardware description language) model of the application along with user specified constraints and automatically generates a task graph G; partitions G based on the user specified constraints and maps the blocks of the partitions onto the different FPGAs in the given multi-FPGA architecture, all in a single-shot. The proposed algorithm was successfully employed to spatially partition a reasonably big cryptographic application that involved a 1024-bit modular exponentiation and to map the same onto a network of nine ACEX1K based Altera EP1K30QC208-1 FPGAs. The suggested parallel evolutionary algorithm for the partitioning step was implemented on a 6-node SGI Origin-2000 platform using the message passing interface (MPI) standard. The results obtained by executing the same are extremely encouraging, especially for larger task graphs.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134565514","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299794
P. Lanzi
In XCS classifier fitness is measured as the relative accuracy of classifier prediction. A classifier is fit if its prediction of the expected payoff is more accurate than that provided by the other classifiers that appear in the same environmental niches. We introduce a modification of Wilson's original definition in which classifier fitness is measured as the absolute (raw) accuracy of classifier prediction. A classifier is fit if the error affecting its prediction is smaller than a given threshold. Then we compare Wilson's relative accuracy and raw accuracy on a number of problems both in terms of learning performance and in terms of generalization capabilities.
{"title":"A comparison of relative accuracy and raw accuracy in XCS","authors":"P. Lanzi","doi":"10.1109/CEC.2003.1299794","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299794","url":null,"abstract":"In XCS classifier fitness is measured as the relative accuracy of classifier prediction. A classifier is fit if its prediction of the expected payoff is more accurate than that provided by the other classifiers that appear in the same environmental niches. We introduce a modification of Wilson's original definition in which classifier fitness is measured as the absolute (raw) accuracy of classifier prediction. A classifier is fit if the error affecting its prediction is smaller than a given threshold. Then we compare Wilson's relative accuracy and raw accuracy on a number of problems both in terms of learning performance and in terms of generalization capabilities.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129109900","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299787
Eiichi Takashima, Y. Murata, N. Shibata, Minoru Ito
Exploration efficiency of GAs largely depends on parameter values. But, it is hard to manually adjust these values. To cope with this problem, several adaptive GAs which automatically adjust parameters have been proposed. However, most of the existing adaptive GAs can adapt only a few parameters at the same time. Although several adaptive GAs can adapt multiple parameters simultaneously, these algorithms require extremely large computation costs. In this paper, we propose self adaptive island GA (SAIGA) which adapts four parameter values simultaneously while finding a solution to a problem. SAIGA is a kind of island GA, and it adapts parameter values using a similar mechanism to meta-GA. Throughout our evaluation experiments, we confirmed that our algorithm outperforms a simple GA using De Jong's rational parameters, and has performance close to a simple GA using manually tuned parameter values.
{"title":"Self adaptive island GA","authors":"Eiichi Takashima, Y. Murata, N. Shibata, Minoru Ito","doi":"10.1109/CEC.2003.1299787","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299787","url":null,"abstract":"Exploration efficiency of GAs largely depends on parameter values. But, it is hard to manually adjust these values. To cope with this problem, several adaptive GAs which automatically adjust parameters have been proposed. However, most of the existing adaptive GAs can adapt only a few parameters at the same time. Although several adaptive GAs can adapt multiple parameters simultaneously, these algorithms require extremely large computation costs. In this paper, we propose self adaptive island GA (SAIGA) which adapts four parameter values simultaneously while finding a solution to a problem. SAIGA is a kind of island GA, and it adapts parameter values using a similar mechanism to meta-GA. Throughout our evaluation experiments, we confirmed that our algorithm outperforms a simple GA using De Jong's rational parameters, and has performance close to a simple GA using manually tuned parameter values.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"158 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132526522","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299913
J. J. Valdés, L. Molina, N. Peris
Many data mining and machine learning algorithms require databases in which objects are described by discrete attributes. However, it is very common that the attributes are in the ratio or interval scales. In order to apply these algorithms, the original attributes must be transformed into the nominal or ordinal scale via discretization. An appropriate transformation is crucial because of the large influence on the results obtained from data mining procedures. This paper presents a hybrid technique for the simultaneous supervised discretization of continuous attributes, based on evolutionary algorithms, in particular, evolution strategies (ES), which is combined with rough set theory and information theory. The purpose is to construct a discretization scheme for all continuous attributes simultaneously (i.e. global) in such a way that class predictability is maximized w.r.t the discrete classes generated for the predictor variables. The ES approach is applied to 17 public data sets and the results are compared with classical discretization methods. ES-based discretization not only outperforms these methods, but leads to much simpler data models and is able to discover irrelevant attributes. These features are not present in classical discretization techniques.
{"title":"An evolution strategies approach to the simultaneous discretization of numeric attributes in data mining","authors":"J. J. Valdés, L. Molina, N. Peris","doi":"10.1109/CEC.2003.1299913","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299913","url":null,"abstract":"Many data mining and machine learning algorithms require databases in which objects are described by discrete attributes. However, it is very common that the attributes are in the ratio or interval scales. In order to apply these algorithms, the original attributes must be transformed into the nominal or ordinal scale via discretization. An appropriate transformation is crucial because of the large influence on the results obtained from data mining procedures. This paper presents a hybrid technique for the simultaneous supervised discretization of continuous attributes, based on evolutionary algorithms, in particular, evolution strategies (ES), which is combined with rough set theory and information theory. The purpose is to construct a discretization scheme for all continuous attributes simultaneously (i.e. global) in such a way that class predictability is maximized w.r.t the discrete classes generated for the predictor variables. The ES approach is applied to 17 public data sets and the results are compared with classical discretization methods. ES-based discretization not only outperforms these methods, but leads to much simpler data models and is able to discover irrelevant attributes. These features are not present in classical discretization techniques.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134197820","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 : 2003-12-08DOI: 10.1109/CEC.2003.1299630
C. S. Chang, L. R. Lu, F. Wang
Mass rapid transit (MRT) operation results in voltage and current harmonic distortions in AC supply systems. Power-supply standards limit the acceptable harmonic-distortion levels. In order to observe these limits, it is necessary to identify the worst-case harmonic distortion in MRT system. Train operating modes and system configurations affect the level of harmonic distortions. Harmonic worst-case identification can be treated as an optimization problem in terms of train separations and traffic conditions. The approach uses approximate train movement and consumption models, and AC/DC harmonic loadflow for evaluating the harmonic distortions. This paper uses a new method called differential evolution (DE) for solving the problem. Parallel studies using genetic algorithm (GA) are also carried out. Comparative results demonstrate the favourable features of DE for large-scale optimization with real variables, as the method is efficient and fast converging.
{"title":"Application of differential evolution for harmonic worst-case identification of mass rapid transit power supply system","authors":"C. S. Chang, L. R. Lu, F. Wang","doi":"10.1109/CEC.2003.1299630","DOIUrl":"https://doi.org/10.1109/CEC.2003.1299630","url":null,"abstract":"Mass rapid transit (MRT) operation results in voltage and current harmonic distortions in AC supply systems. Power-supply standards limit the acceptable harmonic-distortion levels. In order to observe these limits, it is necessary to identify the worst-case harmonic distortion in MRT system. Train operating modes and system configurations affect the level of harmonic distortions. Harmonic worst-case identification can be treated as an optimization problem in terms of train separations and traffic conditions. The approach uses approximate train movement and consumption models, and AC/DC harmonic loadflow for evaluating the harmonic distortions. This paper uses a new method called differential evolution (DE) for solving the problem. Parallel studies using genetic algorithm (GA) are also carried out. Comparative results demonstrate the favourable features of DE for large-scale optimization with real variables, as the method is efficient and fast converging.","PeriodicalId":416243,"journal":{"name":"The 2003 Congress on Evolutionary Computation, 2003. CEC '03.","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132356686","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：