As a crucial data preprocessing method in data mining,feature selection(FS)can be regarded as a bi-objective optimization problem that aims to maximize classification accuracy and minimize the number of selected featu...As a crucial data preprocessing method in data mining,feature selection(FS)can be regarded as a bi-objective optimization problem that aims to maximize classification accuracy and minimize the number of selected features.Evolutionary computing(EC)is promising for FS owing to its powerful search capability.However,in traditional EC-based methods,feature subsets are represented via a length-fixed individual encoding.It is ineffective for high-dimensional data,because it results in a huge search space and prohibitive training time.This work proposes a length-adaptive non-dominated sorting genetic algorithm(LA-NSGA)with a length-variable individual encoding and a length-adaptive evolution mechanism for bi-objective highdimensional FS.In LA-NSGA,an initialization method based on correlation and redundancy is devised to initialize individuals of diverse lengths,and a Pareto dominance-based length change operator is introduced to guide individuals to explore in promising search space adaptively.Moreover,a dominance-based local search method is employed for further improvement.The experimental results based on 12 high-dimensional gene datasets show that the Pareto front of feature subsets produced by LA-NSGA is superior to those of existing algorithms.展开更多
A multi-objective optimization problem has two or more objectives to be minimized or maximized simultaneously. It is usually difficult to arrive at a solution that optimizes every objective. Therefore, the best way of...A multi-objective optimization problem has two or more objectives to be minimized or maximized simultaneously. It is usually difficult to arrive at a solution that optimizes every objective. Therefore, the best way of dealing with the problem is to obtain a set of good solutions for the decision maker to select the one that best serves his/her interest. In this paper, a ratio min-max strategy is incorporated (after Pareto optimal solutions are obtained) under a weighted sum scalarization of the objectives to aid the process of identifying a best compromise solution. The bi-objective discrete optimization problem which has distance and social cost (in rail construction, say) as the criteria was solved by an improved Ant Colony System algorithm developed by the authors. The model and methodology were applied to hypothetical networks of fourteen nodes and twenty edges, and another with twenty nodes and ninety-seven edges as test cases. Pareto optimal solutions and their maximum margins of error were obtained for the problems to assist in decision making. The proposed model and method is user-friendly and provides the decision maker with information on the quality of each of the Pareto optimal solutions obtained, thus facilitating decision making.展开更多
The purpose of this work is to present a methodology to provide a solution to a Bi-objective Green Vehicle Routing Problem (BGVRP). The methodology, illustrated using a case study (newspaper distribution problem) and ...The purpose of this work is to present a methodology to provide a solution to a Bi-objective Green Vehicle Routing Problem (BGVRP). The methodology, illustrated using a case study (newspaper distribution problem) and literature Instances, was divided into three stages: Stage 1, data treatment;Stage 2, “metaheuristic approaches” (hybrid or non-hybrid), used comparatively, more specifically: NSGA-II (Non-dominated Sorting Genetic Algorithm II), MOPSO (Multi-Objective Particle Swarm Optimization), which were compared with the new approaches proposed by the authors, CWNSGA-II (Clarke and Wright’s Savings with the Non-dominated Sorting Genetic Algorithm II) and CWTSNSGA-II (Clarke and Wright’s Savings, Tabu Search and Non-dominated Sorting Genetic Algorithm II);Stage 3, analysis of the results, with a comparison of the algorithms. An optimization of 19.9% was achieved for Objective Function 1 (OF<sub>1</sub>;minimization of CO<sub>2</sub> emissions) and consequently the same percentage for the minimization of total distance, and 87.5% for Objective Function 2 (OF<sub>2</sub>;minimization of the difference in demand). Metaheuristic approaches hybrid achieved superior results for case study and instances. In this way, the procedure presented here can bring benefits to society as it considers environmental issues and also balancing work between the routes, ensuring savings and satisfaction for the users.展开更多
This paper illustrates a new bi-objective optimization model of real-time automatic generation control dispatch in a performance-based frequency regulation market.It attempts to simultaneously minimize the total power...This paper illustrates a new bi-objective optimization model of real-time automatic generation control dispatch in a performance-based frequency regulation market.It attempts to simultaneously minimize the total power deviation and the regulation mileage payment via optimally distributing the realtime total generation command to different regulation units.To handle this problem,an efficient non-dominated sorting genetic algorithm II is adopted to rapidly obtain a high-quality Pareto front for real-time AGC dispatch.A dynamic ideal point based decision making technique is designed to select the best compromise solution from the obtained Pareto front according to the minimization of the total regulation variation,which can effectively avoid an excessive regulation variation for each unit.Finally,two testing systems are used to verify the performance of the proposed technique.展开更多
This study introduces a comprehensive and automated framework that leverages data-driven method-ologies to address various challenges in shale gas development and production.Specifically,it harnesses the power of Auto...This study introduces a comprehensive and automated framework that leverages data-driven method-ologies to address various challenges in shale gas development and production.Specifically,it harnesses the power of Automated Machine Learning(AutoML)to construct an ensemble model to predict the estimated ultimate recovery(EUR)of shale gas wells.To demystify the“black-box”nature of the ensemble model,KernelSHAP,a kernel-based approach to compute Shapley values,is utilized for elucidating the influential factors that affect shale gas production at both global and local scales.Furthermore,a bi-objective optimization algorithm named NSGA-Ⅱ is seamlessly incorporated to opti-mize hydraulic fracturing designs for production boost and cost control.This innovative framework addresses critical limitations often encountered in applying machine learning(ML)to shale gas pro-duction:the challenge of achieving sufficient model accuracy with limited samples,the multidisciplinary expertise required for developing robust ML models,and the need for interpretability in“black-box”models.Validation with field data from the Fuling shale gas field in the Sichuan Basin substantiates the framework's efficacy in enhancing the precision and applicability of data-driven techniques.The test accuracy of the ensemble ML model reached 83%compared to a maximum of 72%of single ML models.The contribution of each geological and engineering factor to the overall production was quantitatively evaluated.Fracturing design optimization raised EUR by 7%-34%under different production and cost tradeoff scenarios.The results empower domain experts to conduct more precise and objective data-driven analyses and optimizations for shale gas production with minimal expertise in data science.展开更多
This paper studies learning effect as a resource utilization technique that can model improvement in worker＇s ability as a result of repeating similar tasks. By considering learning of workers while performing setup ...This paper studies learning effect as a resource utilization technique that can model improvement in worker＇s ability as a result of repeating similar tasks. By considering learning of workers while performing setup times, a schedule can be determined to place jobs that share similar tools and fixtures next to each other. The purpose of this paper is to schedule a set of jobs in a hybrid flow shop （HFS） environment with learning effect while minimizing two objectives that are in conflict： namely maximum completion time （makespan） and total tardiness. Minimizing makespan is desirable from an internal efficiency viewpoint, but may result in individual jobs being scheduled past their due date, causing customer dissatisfaction and penalty costs. A bi-objective mixed integer programming model is developed, and the complexity of the developed bi-objective model is compared against the bi-criteria one through numerical examples. The effect of worker learning on the structure of assigned jobs to machines and their sequences is analyzed. Two solution methods based on the hybrid water flow like algorithm and non-dominated sorting and ranking concepts are proposed to solve the problem. The quality of the approximated sets of Pareto solutions is evaluated using several performance criteria. The results show that the proposed algorithms with learning effect perform well in reducing setup times and eliminate the need for setups itself through proper scheduling.展开更多
Partitioning a complex power network into a number of sub-zones can help realize a divide-and-conquer’management structure for the whole system,such as voltage and reactive power control,coherency identification,powe...Partitioning a complex power network into a number of sub-zones can help realize a divide-and-conquer’management structure for the whole system,such as voltage and reactive power control,coherency identification,power system restoration,etc.Extensive partitioning methods have been proposed by defining various distances,applying different clustering methods,or formulating varying optimization models for one specific objective.However,a power network partition may serve two or more objectives,where a trade-off among these objectives is required.This paper proposes a novel weighted consensus clustering-based approach for bi-objective power network partition.By varying the weights of different partitions for different objectives,Pareto improvement can be explored based on the node-based and subset-based consensus clustering methods.Case studies on the IEEE 300-bus test system are conducted to verify the effectiveness and superiority of our proposed method.展开更多
Traffic congestion in road transportation networks is a persistent problem in major metropolitan cities around the world.In this context,this paper deals with exploiting underutilized road capacities in a network to l...Traffic congestion in road transportation networks is a persistent problem in major metropolitan cities around the world.In this context,this paper deals with exploiting underutilized road capacities in a network to lower the congestion on overutilized links while simultaneously satisfying the system optimal flow assignment for sustainable transportation.Four congestion mitigation strategies are identified based on deviation and relative deviation of link volume from the corresponding capacity.Consequently,four biobjective mathematical programming optimal flow distribution(OFD)models are proposed.The case study results demonstrate that all the proposed models improve system performance and reduce congestion on high volume links by shifting flows to low volumeto-capacity links compared to UE and SO models.Among the models,the system optimality with minimal sum and maximum absolute relative-deviation models(SO-SAR and SO-MAR)showed superior results for different performance measures.The SO-SAR model yielded 50%and 30%fewer links at higher link utilization factors than UE and SO models,respectively.Also,it showed more than 25%improvement in path travel times compared to UE travel time for about 100 paths and resulted in the least network congestion index of1.04 compared to the other OFD and UE models.Conversely,the SO-MAR model yielded the least total distance and total system travel time,resulting in lower fuel consumption and emissions,thus contributing to sustainability.The proposed models contribute towards efficient transportation infrastructure management and will be of interest to transportation planners and traffic managers.展开更多
基金supported in part by the National Natural Science Foundation of China(62172065,62072060)。
文摘As a crucial data preprocessing method in data mining,feature selection(FS)can be regarded as a bi-objective optimization problem that aims to maximize classification accuracy and minimize the number of selected features.Evolutionary computing(EC)is promising for FS owing to its powerful search capability.However,in traditional EC-based methods,feature subsets are represented via a length-fixed individual encoding.It is ineffective for high-dimensional data,because it results in a huge search space and prohibitive training time.This work proposes a length-adaptive non-dominated sorting genetic algorithm(LA-NSGA)with a length-variable individual encoding and a length-adaptive evolution mechanism for bi-objective highdimensional FS.In LA-NSGA,an initialization method based on correlation and redundancy is devised to initialize individuals of diverse lengths,and a Pareto dominance-based length change operator is introduced to guide individuals to explore in promising search space adaptively.Moreover,a dominance-based local search method is employed for further improvement.The experimental results based on 12 high-dimensional gene datasets show that the Pareto front of feature subsets produced by LA-NSGA is superior to those of existing algorithms.
文摘A multi-objective optimization problem has two or more objectives to be minimized or maximized simultaneously. It is usually difficult to arrive at a solution that optimizes every objective. Therefore, the best way of dealing with the problem is to obtain a set of good solutions for the decision maker to select the one that best serves his/her interest. In this paper, a ratio min-max strategy is incorporated (after Pareto optimal solutions are obtained) under a weighted sum scalarization of the objectives to aid the process of identifying a best compromise solution. The bi-objective discrete optimization problem which has distance and social cost (in rail construction, say) as the criteria was solved by an improved Ant Colony System algorithm developed by the authors. The model and methodology were applied to hypothetical networks of fourteen nodes and twenty edges, and another with twenty nodes and ninety-seven edges as test cases. Pareto optimal solutions and their maximum margins of error were obtained for the problems to assist in decision making. The proposed model and method is user-friendly and provides the decision maker with information on the quality of each of the Pareto optimal solutions obtained, thus facilitating decision making.
文摘The purpose of this work is to present a methodology to provide a solution to a Bi-objective Green Vehicle Routing Problem (BGVRP). The methodology, illustrated using a case study (newspaper distribution problem) and literature Instances, was divided into three stages: Stage 1, data treatment;Stage 2, “metaheuristic approaches” (hybrid or non-hybrid), used comparatively, more specifically: NSGA-II (Non-dominated Sorting Genetic Algorithm II), MOPSO (Multi-Objective Particle Swarm Optimization), which were compared with the new approaches proposed by the authors, CWNSGA-II (Clarke and Wright’s Savings with the Non-dominated Sorting Genetic Algorithm II) and CWTSNSGA-II (Clarke and Wright’s Savings, Tabu Search and Non-dominated Sorting Genetic Algorithm II);Stage 3, analysis of the results, with a comparison of the algorithms. An optimization of 19.9% was achieved for Objective Function 1 (OF<sub>1</sub>;minimization of CO<sub>2</sub> emissions) and consequently the same percentage for the minimization of total distance, and 87.5% for Objective Function 2 (OF<sub>2</sub>;minimization of the difference in demand). Metaheuristic approaches hybrid achieved superior results for case study and instances. In this way, the procedure presented here can bring benefits to society as it considers environmental issues and also balancing work between the routes, ensuring savings and satisfaction for the users.
基金supported by National Natural Science Foundation of China(51907112,61963020)Natural Science Foundation of Guangdong Province of China(2019A1515011671).
文摘This paper illustrates a new bi-objective optimization model of real-time automatic generation control dispatch in a performance-based frequency regulation market.It attempts to simultaneously minimize the total power deviation and the regulation mileage payment via optimally distributing the realtime total generation command to different regulation units.To handle this problem,an efficient non-dominated sorting genetic algorithm II is adopted to rapidly obtain a high-quality Pareto front for real-time AGC dispatch.A dynamic ideal point based decision making technique is designed to select the best compromise solution from the obtained Pareto front according to the minimization of the total regulation variation,which can effectively avoid an excessive regulation variation for each unit.Finally,two testing systems are used to verify the performance of the proposed technique.
基金funded by the National Natural Science Foundation of China(42050104).
文摘This study introduces a comprehensive and automated framework that leverages data-driven method-ologies to address various challenges in shale gas development and production.Specifically,it harnesses the power of Automated Machine Learning(AutoML)to construct an ensemble model to predict the estimated ultimate recovery(EUR)of shale gas wells.To demystify the“black-box”nature of the ensemble model,KernelSHAP,a kernel-based approach to compute Shapley values,is utilized for elucidating the influential factors that affect shale gas production at both global and local scales.Furthermore,a bi-objective optimization algorithm named NSGA-Ⅱ is seamlessly incorporated to opti-mize hydraulic fracturing designs for production boost and cost control.This innovative framework addresses critical limitations often encountered in applying machine learning(ML)to shale gas pro-duction:the challenge of achieving sufficient model accuracy with limited samples,the multidisciplinary expertise required for developing robust ML models,and the need for interpretability in“black-box”models.Validation with field data from the Fuling shale gas field in the Sichuan Basin substantiates the framework's efficacy in enhancing the precision and applicability of data-driven techniques.The test accuracy of the ensemble ML model reached 83%compared to a maximum of 72%of single ML models.The contribution of each geological and engineering factor to the overall production was quantitatively evaluated.Fracturing design optimization raised EUR by 7%-34%under different production and cost tradeoff scenarios.The results empower domain experts to conduct more precise and objective data-driven analyses and optimizations for shale gas production with minimal expertise in data science.
文摘This paper studies learning effect as a resource utilization technique that can model improvement in worker＇s ability as a result of repeating similar tasks. By considering learning of workers while performing setup times, a schedule can be determined to place jobs that share similar tools and fixtures next to each other. The purpose of this paper is to schedule a set of jobs in a hybrid flow shop （HFS） environment with learning effect while minimizing two objectives that are in conflict： namely maximum completion time （makespan） and total tardiness. Minimizing makespan is desirable from an internal efficiency viewpoint, but may result in individual jobs being scheduled past their due date, causing customer dissatisfaction and penalty costs. A bi-objective mixed integer programming model is developed, and the complexity of the developed bi-objective model is compared against the bi-criteria one through numerical examples. The effect of worker learning on the structure of assigned jobs to machines and their sequences is analyzed. Two solution methods based on the hybrid water flow like algorithm and non-dominated sorting and ranking concepts are proposed to solve the problem. The quality of the approximated sets of Pareto solutions is evaluated using several performance criteria. The results show that the proposed algorithms with learning effect perform well in reducing setup times and eliminate the need for setups itself through proper scheduling.
基金supported in part by the National Key R&D Program of China(No.2016YFB0900100)the Major Smart Grid Joint Project of National Natural Science Foundation of China and State Grid(No.U1766212).
文摘Partitioning a complex power network into a number of sub-zones can help realize a divide-and-conquer’management structure for the whole system,such as voltage and reactive power control,coherency identification,power system restoration,etc.Extensive partitioning methods have been proposed by defining various distances,applying different clustering methods,or formulating varying optimization models for one specific objective.However,a power network partition may serve two or more objectives,where a trade-off among these objectives is required.This paper proposes a novel weighted consensus clustering-based approach for bi-objective power network partition.By varying the weights of different partitions for different objectives,Pareto improvement can be explored based on the node-based and subset-based consensus clustering methods.Case studies on the IEEE 300-bus test system are conducted to verify the effectiveness and superiority of our proposed method.
文摘Traffic congestion in road transportation networks is a persistent problem in major metropolitan cities around the world.In this context,this paper deals with exploiting underutilized road capacities in a network to lower the congestion on overutilized links while simultaneously satisfying the system optimal flow assignment for sustainable transportation.Four congestion mitigation strategies are identified based on deviation and relative deviation of link volume from the corresponding capacity.Consequently,four biobjective mathematical programming optimal flow distribution(OFD)models are proposed.The case study results demonstrate that all the proposed models improve system performance and reduce congestion on high volume links by shifting flows to low volumeto-capacity links compared to UE and SO models.Among the models,the system optimality with minimal sum and maximum absolute relative-deviation models(SO-SAR and SO-MAR)showed superior results for different performance measures.The SO-SAR model yielded 50%and 30%fewer links at higher link utilization factors than UE and SO models,respectively.Also,it showed more than 25%improvement in path travel times compared to UE travel time for about 100 paths and resulted in the least network congestion index of1.04 compared to the other OFD and UE models.Conversely,the SO-MAR model yielded the least total distance and total system travel time,resulting in lower fuel consumption and emissions,thus contributing to sustainability.The proposed models contribute towards efficient transportation infrastructure management and will be of interest to transportation planners and traffic managers.
