Due to the NP-hardness of the two-sided assembly line balancing （TALB） problem, multiple constraints existing in real applications are less studied, especially when one task is involved with several constraints. In ...Due to the NP-hardness of the two-sided assembly line balancing （TALB） problem, multiple constraints existing in real applications are less studied, especially when one task is involved with several constraints. In this paper, an effective hybrid algorithm is proposed to address the TALB problem with multiple constraints （TALB-MC）. Considering the discrete attribute of TALB-MC and the continuous attribute of the standard teaching-learning-based optimization （TLBO） algorithm, the random-keys method is hired in task permutation representation, for the purpose of bridging the gap between them. Subsequently, a special mechanism for handling multiple constraints is developed. In the mechanism, the directions constraint of each task is ensured by the direction check and adjustment. The zoning constraints and the synchronism constraints are satisfied by teasing out the hidden correlations among constraints. The positional constraint is allowed to be violated to some extent in decoding and punished in cost fimction. Finally, with the TLBO seeking for the global optimum, the variable neighborhood search （VNS） is further hybridized to extend the local search space. The experimental results show that the proposed hybrid algorithm outperforms the late acceptance hill-climbing algorithm （LAHC） for TALB-MC in most cases, especially for large-size problems with multiple constraints, and demonstrates well balance between the exploration and the exploitation. This research proposes an effective and efficient algorithm for solving TALB-MC problem by hybridizing the TLBO and VNS.展开更多
Solving constrained multi-objective optimization problems with evolutionary algorithms has attracted considerable attention.Various constrained multi-objective optimization evolutionary algorithms(CMOEAs)have been dev...Solving constrained multi-objective optimization problems with evolutionary algorithms has attracted considerable attention.Various constrained multi-objective optimization evolutionary algorithms(CMOEAs)have been developed with the use of different algorithmic strategies,evolutionary operators,and constraint-handling techniques.The performance of CMOEAs may be heavily dependent on the operators used,however,it is usually difficult to select suitable operators for the problem at hand.Hence,improving operator selection is promising and necessary for CMOEAs.This work proposes an online operator selection framework assisted by Deep Reinforcement Learning.The dynamics of the population,including convergence,diversity,and feasibility,are regarded as the state;the candidate operators are considered as actions;and the improvement of the population state is treated as the reward.By using a Q-network to learn a policy to estimate the Q-values of all actions,the proposed approach can adaptively select an operator that maximizes the improvement of the population according to the current state and thereby improve the algorithmic performance.The framework is embedded into four popular CMOEAs and assessed on 42 benchmark problems.The experimental results reveal that the proposed Deep Reinforcement Learning-assisted operator selection significantly improves the performance of these CMOEAs and the resulting algorithm obtains better versatility compared to nine state-of-the-art CMOEAs.展开更多
The operation variables,including feed rate of ore slurry,caustic solution and live steams in the double-stream alumina digestion process,determine the product quality,process costs and the environment pollution.Previ...The operation variables,including feed rate of ore slurry,caustic solution and live steams in the double-stream alumina digestion process,determine the product quality,process costs and the environment pollution.Previously,they were set by the technical workers according to the offline analysis results and an empirical formula,which leads to unstable process indices and high consumption frequently.So,a multi-objective optimization model is built to maintain the balance between resource consumptions and process indices by taking technical indices and energy efficiency as objectives,where the key technical indices are predicted based on the digestion kinetics of diaspore.A multi-objective state transition algorithm(MOSTA)is improved to solve the problem,in which a self-adaptive strategy is applied to dynamically adjust the operator factors of the MOSTA and dynamic infeasible threshold is used to handle constraints to enhance searching efficiency and ability of the algorithm.Then a rule based strategy is designed to make the final decision from the Pareto frontiers.The method is integrated into an optimal control system for the industrial digestion process and tested in the actual production.Results show that the proposed method can achieve the technical target while reducing the energy consumption.展开更多
Accurate determination of rock mass parameters is essential for ensuring the accuracy of numericalsimulations. Displacement back-analysis is the most widely used method;however, the reliability of thecurrent approache...Accurate determination of rock mass parameters is essential for ensuring the accuracy of numericalsimulations. Displacement back-analysis is the most widely used method;however, the reliability of thecurrent approaches remains unsatisfactory. Therefore, in this paper, a multistage rock mass parameterback-analysis method, that considers the construction process and displacement losses is proposed andimplemented through the coupling of numerical simulation, auto-machine learning (AutoML), andmulti-objective optimization algorithms (MOOAs). First, a parametric modeling platform for mechanizedtwin tunnels is developed, generating a dataset through extensive numerical simulations. Next, theAutoML method is utilized to establish a surrogate model linking rock parameters and displacements.The tunnel construction process is divided into multiple stages, transforming the rock mass parameterback-analysis into a multi-objective optimization problem, for which multi-objective optimization algorithmsare introduced to obtain the rock mass parameters. The newly proposed rock mass parameterback-analysis method is validated in a mechanized twin tunnel project, and its accuracy and effectivenessare demonstrated. Compared with traditional single-stage back-analysis methods, the proposedmodel decreases the average absolute percentage error from 12.73% to 4.34%, significantly improving theaccuracy of the back-analysis. Moreover, although the accuracy of back analysis significantly increaseswith the number of construction stages considered, the back analysis time is acceptable. This studyprovides a new method for displacement back analysis that is efficient and accurate, thereby paving theway for precise parameter determination in numerical simulations.展开更多
Workload balancing in cloud computing is not yet resolved,particularly considering Infrastructure as a Service(IaaS)in the cloud network.The problem of being underloaded or overloaded should not occur at the time of t...Workload balancing in cloud computing is not yet resolved,particularly considering Infrastructure as a Service(IaaS)in the cloud network.The problem of being underloaded or overloaded should not occur at the time of the server or host accessing the cloud which may lead to create system crash problem.Thus,to resolve these existing problems,an efficient task scheduling algorithm is required for distributing the tasks over the entire feasible resources,which is termed load balancing.The load balancing approach assures that the entire Virtual Machines(VMs)are utilized appropriately.So,it is highly essential to develop a load-balancing model in a cloud environment based on machine learning and optimization strategies.Here,the computing and networking data is utilized for the analysis to observe the traffic as well as performance patterns.The acquired data is offered to the machine learning decision to select the right server by predicting the performance effectively by employing an Optimal Kernel-based Extreme Learning Machine(OK-ELM)and their parameter is tuned by the developed hybrid approach Population Size-based Mud Ring Tunicate Swarm Algorithm(PS-MRTSA).Further,effective scheduling is performed to resolve the load balancing issues by employing the developed model MR-TSA.Here,the developed approach effectively resolves the multi-objective constraints such as Response time,Resource cost,and energy consumption.Thus,the recommended load balancing model securesan enhanced performance rate than the traditional approaches over several experimental analyses.展开更多
Evolutionary algorithms have been shown to be very successful in solving multi-objective optimization problems(MOPs).However,their performance often deteriorates when solving MOPs with irregular Pareto fronts.To remed...Evolutionary algorithms have been shown to be very successful in solving multi-objective optimization problems(MOPs).However,their performance often deteriorates when solving MOPs with irregular Pareto fronts.To remedy this issue,a large body of research has been performed in recent years and many new algorithms have been proposed.This paper provides a comprehensive survey of the research on MOPs with irregular Pareto fronts.We start with a brief introduction to the basic concepts,followed by a summary of the benchmark test problems with irregular problems,an analysis of the causes of the irregularity,and real-world optimization problems with irregular Pareto fronts.Then,a taxonomy of the existing methodologies for handling irregular problems is given and representative algorithms are reviewed with a discussion of their strengths and weaknesses.Finally,open challenges are pointed out and a few promising future directions are suggested.展开更多
In order to solve the flexible job shop scheduling problem with variable batches,we propose an improved multiobjective optimization algorithm,which combines the idea of inverse scheduling.First,a flexible job shop pro...In order to solve the flexible job shop scheduling problem with variable batches,we propose an improved multiobjective optimization algorithm,which combines the idea of inverse scheduling.First,a flexible job shop problem with the variable batches scheduling model is formulated.Second,we propose a batch optimization algorithm with inverse scheduling in which the batch size is adjusted by the dynamic feedback batch adjusting method.Moreover,in order to increase the diversity of the population,two methods are developed.One is the threshold to control the neighborhood updating,and the other is the dynamic clustering algorithm to update the population.Finally,a group of experiments are carried out.The results show that the improved multi-objective optimization algorithm can ensure the diversity of Pareto solutions effectively,and has effective performance in solving the flexible job shop scheduling problem with variable batches.展开更多
For training the present Neural Network(NN)models,the standard technique is to utilize decaying Learning Rates(LR).While the majority of these techniques commence with a large LR,they will decay multiple times over ti...For training the present Neural Network(NN)models,the standard technique is to utilize decaying Learning Rates(LR).While the majority of these techniques commence with a large LR,they will decay multiple times over time.Decaying has been proved to enhance generalization as well as optimization.Other parameters,such as the network’s size,the number of hidden layers,drop-outs to avoid overfitting,batch size,and so on,are solely based on heuristics.This work has proposed Adaptive Teaching Learning Based(ATLB)Heuristic to identify the optimal hyperparameters for diverse networks.Here we consider three architec-tures Recurrent Neural Networks(RNN),Long Short Term Memory(LSTM),Bidirectional Long Short Term Memory(BiLSTM)of Deep Neural Networks for classification.The evaluation of the proposed ATLB is done through the various learning rate schedulers Cyclical Learning Rate(CLR),Hyperbolic Tangent Decay(HTD),and Toggle between Hyperbolic Tangent Decay and Triangular mode with Restarts(T-HTR)techniques.Experimental results have shown the performance improvement on the 20Newsgroup,Reuters Newswire and IMDB dataset.展开更多
This research paper presents a comprehensive investigation into the effectiveness of the DeepSurNet-NSGA II(Deep Surrogate Model-Assisted Non-dominated Sorting Genetic Algorithm II)for solving complex multiobjective o...This research paper presents a comprehensive investigation into the effectiveness of the DeepSurNet-NSGA II(Deep Surrogate Model-Assisted Non-dominated Sorting Genetic Algorithm II)for solving complex multiobjective optimization problems,with a particular focus on robotic leg-linkage design.The study introduces an innovative approach that integrates deep learning-based surrogate models with the robust Non-dominated Sorting Genetic Algorithm II,aiming to enhance the efficiency and precision of the optimization process.Through a series of empirical experiments and algorithmic analyses,the paper demonstrates a high degree of correlation between solutions generated by the DeepSurNet-NSGA II and those obtained from direct experimental methods,underscoring the algorithm’s capability to accurately approximate the Pareto-optimal frontier while significantly reducing computational demands.The methodology encompasses a detailed exploration of the algorithm’s configuration,the experimental setup,and the criteria for performance evaluation,ensuring the reproducibility of results and facilitating future advancements in the field.The findings of this study not only confirm the practical applicability and theoretical soundness of the DeepSurNet-NSGA II in navigating the intricacies of multi-objective optimization but also highlight its potential as a transformative tool in engineering and design optimization.By bridging the gap between complex optimization challenges and achievable solutions,this research contributes valuable insights into the optimization domain,offering a promising direction for future inquiries and technological innovations.展开更多
Recently,mega Low Earth Orbit(LEO)Satellite Network(LSN)systems have gained more and more attention due to low latency,broadband communications and global coverage for ground users.One of the primary challenges for LS...Recently,mega Low Earth Orbit(LEO)Satellite Network(LSN)systems have gained more and more attention due to low latency,broadband communications and global coverage for ground users.One of the primary challenges for LSN systems with inter-satellite links is the routing strategy calculation and maintenance,due to LSN constellation scale and dynamic network topology feature.In order to seek an efficient routing strategy,a Q-learning-based dynamic distributed Routing scheme for LSNs(QRLSN)is proposed in this paper.To achieve low end-toend delay and low network traffic overhead load in LSNs,QRLSN adopts a multi-objective optimization method to find the optimal next hop for forwarding data packets.Experimental results demonstrate that the proposed scheme can effectively discover the initial routing strategy and provide long-term Quality of Service(QoS)optimization during the routing maintenance process.In addition,comparison results demonstrate that QRLSN is superior to the virtual-topology-based shortest path routing algorithm.展开更多
A multi-objective optimization model for draft scheduling of hot strip mill was presented, rolling power minimizing, rolling force ratio distribution and good strip shape as the objective functions. A multi-objective ...A multi-objective optimization model for draft scheduling of hot strip mill was presented, rolling power minimizing, rolling force ratio distribution and good strip shape as the objective functions. A multi-objective differential evolution algorithm based on decomposition （MODE/D）. The two-objective and three-objective optimization experiments were performed respectively to demonstrate the optimal solutions of trade-off. The simulation results show that MODE/D can obtain a good Pareto-optimal front, which suggests a series of alternative solutions to draft scheduling. The extreme Pareto solutions are found feasible and the centres of the Pareto fronts give a good compromise. The conflict exists between each two ones of three objectives. The final optimal solution is selected from the Pareto-optimal front by the importance of objectives, and it can achieve a better performance in all objective dimensions than the empirical solutions. Finally, the practical application cases confirm the feasibility of the multi-objective approach, and the optimal solutions can gain a better rolling stability than the empirical solutions, and strip flatness decreases from （0± 63） IU to （0±45） IU in industrial production.展开更多
The Salp Swarm Algorithm(SSA)may have trouble in dropping into stagnation as a kind of swarm intelligence method.This paper developed an adaptive barebones salp swarm algorithm with quasi-oppositional-based learning t...The Salp Swarm Algorithm(SSA)may have trouble in dropping into stagnation as a kind of swarm intelligence method.This paper developed an adaptive barebones salp swarm algorithm with quasi-oppositional-based learning to compensate for the above weakness called QBSSA.In the proposed QBSSA,an adaptive barebones strategy can help to reach both accurate convergence speed and high solution quality;quasi-oppositional-based learning can make the population away from traping into local optimal and expand the search space.To estimate the performance of the presented method,a series of tests are performed.Firstly,CEC 2017 benchmark test suit is used to test the ability to solve the high dimensional and multimodal problems;then,based on QBSSA,an improved Kernel Extreme Learning Machine(KELM)model,named QBSSA–KELM,is built to handle medical disease diagnosis problems.All the test results and discussions state clearly that the QBSSA is superior to and very competitive to all the compared algorithms on both convergence speed and solutions accuracy.展开更多
This research provides academic and practical contributions. From a theoretical standpoint, a hybrid harmony search(HS)algorithm, namely the oppositional global-based HS(OGHS), is proposed for solving the multi-object...This research provides academic and practical contributions. From a theoretical standpoint, a hybrid harmony search(HS)algorithm, namely the oppositional global-based HS(OGHS), is proposed for solving the multi-objective flexible job-shop scheduling problems(MOFJSPs) to minimize makespan, total machine workload and critical machine workload. An initialization program embedded in opposition-based learning(OBL) is developed for enabling the individuals to scatter in a well-distributed manner in the initial harmony memory(HM). In addition, the recursive halving technique based on opposite number is employed for shrinking the neighbourhood space in the searching phase of the OGHS. From a practice-related standpoint, a type of dual vector code technique is introduced for allowing the OGHS algorithm to adapt the discrete nature of the MOFJSP. Two practical techniques, namely Pareto optimality and technique for order preference by similarity to an ideal solution(TOPSIS), are implemented for solving the MOFJSP.Furthermore, the algorithm performance is tested by using different strategies, including OBL and recursive halving, and the OGHS is compared with existing algorithms in the latest studies.Experimental results on representative examples validate the performance of the proposed algorithm for solving the MOFJSP.展开更多
SS304 is a commercial grade stainless steel which is used for various engineering applications like shafts, guides, jigs, fixtures, etc. Ceramic coating of the wear areas of such parts is a regular practice which sign...SS304 is a commercial grade stainless steel which is used for various engineering applications like shafts, guides, jigs, fixtures, etc. Ceramic coating of the wear areas of such parts is a regular practice which significantly enhances the Mean Time Between Failure (MTBF). The final coating quality depends mainly on the coating thickness, surface roughness and hardness which ultimately decides the life. This paper presents an experimental study to effectively optimize the Atmospheric Plasma Spray (APS) process input parameters of Al<sub>2</sub>O<sub>3</sub>-40% TiO2 ceramic coatings to get the best quality of coating on commercial SS304 substrate. The experiments are conducted with a three-level L<sub>18</sub> Orthogonal Array (OA) Design of Experiments (DoE). Critical input parameters considered are: spray nozzle distance, substrate rotating speed, current of the arc, carrier gas flow and coating powder flow rate. The surface roughness, coating thickness and hardness are considered as the output parameters. Mathematical models are generated using regression analysis for individual output parameters. The Analytic Hierarchy Process (AHP) method is applied to generate weights for the individual objective functions and a combined objective function is generated. An advanced optimization method, Teaching-Learning-Based Optimization algorithm (TLBO), is applied to the combined objective function to optimize the values of input parameters to get the best output parameters and confirmation tests are conducted based on that. The significant effects of spray parameters on surface roughness, coating thickness and coating hardness are studied in detail.展开更多
Financial market has systemic complexity and uncertainty.For investors,return and risk often coexist.How to rationally allocate funds into different assets and achieve excess returns with effectively controlling risk ...Financial market has systemic complexity and uncertainty.For investors,return and risk often coexist.How to rationally allocate funds into different assets and achieve excess returns with effectively controlling risk are main problems to be solved in the field of portfolio optimization(PO).At present,due to the influence of modeling and algorithm solving,the PO models established by many researchers are still mainly focused on single-stage single-objective models or single-stage multiobjective models.PO is actually considered as a multi-stage multi-objective optimization problem in real investment scenarios.It is more difficult than the previous single-stage PO model for meeting the realistic requirements.In this paper,the authors proposed a mean-improved stable tail adjusted return ratio-maximum drawdown rate(M-ISTARR-MD)PO model which effectively characterizes the real investment scenario.In order to solve the multi-stage multi-objective PO model with complex multi-constraints,the authors designed a multi-stage constrained multi-objective evolutionary algorithm with orthogonal learning(MSCMOEA-OL).Comparing with four well-known intelligence algorithms,the MSCMOEA-OL algorithm has competitive advantages in solving the M-ISTARR-MD model on the proposed constructed carbon neutral stock dataset.This paper provides a new way to construct and solve the complex PO model.展开更多
For increasing the overall performance of modem manufacturing systems, effective integration of process planning and scheduling functions has been an important area of consideration among researchers. Owing to the com...For increasing the overall performance of modem manufacturing systems, effective integration of process planning and scheduling functions has been an important area of consideration among researchers. Owing to the complexity of handling process planning and scheduling simultaneously, most of the research work has been limited to solving the integrated process planning and scheduling （IPPS） problem for a single objective function. As there are many conflicting objectives when dealing with process planning and scheduling, real world problems cannot be fully captured considering only a single objective for optimization. Therefore considering multi-objective IPPS （MOIPPS） problem is inevitable. Unfortunately, only a handful of research papers are available on solving MOIPPS problem. In this paper, an optimization algorithm for solving MOIPPS problem is presented. The proposed algorithm uses a set of dispatch- ing rules coupled with priority assignment to optimize the IPPS problem for various objectives like makespan, total machine load, total tardiness, etc. A fixed sized external archive coupled with a crowding distance mechanism is used to store and maintain the non-dominated solutions. To compare the results with other algorithms, a C-matric based method has been used. Instances from four recent papers have been solved to demonstrate the effectiveness of the proposed algorithm. The experimental results show that the proposed method is an efficient approach for solving the MOIPPS problem.展开更多
Crowd Anomaly Detection has become a challenge in intelligent video surveillance system and security.Intelligent video surveillance systems make extensive use of data mining,machine learning and deep learning methods....Crowd Anomaly Detection has become a challenge in intelligent video surveillance system and security.Intelligent video surveillance systems make extensive use of data mining,machine learning and deep learning methods.In this paper a novel approach is proposed to identify abnormal occurrences in crowded situations using deep learning.In this approach,Adaptive GoogleNet Neural Network Classifier with Multi-Objective Whale Optimization Algorithm are applied to predict the abnormal video frames in the crowded scenes.We use multiple instance learning(MIL)to dynamically develop a deep anomalous ranking framework.This technique predicts higher anomalous values for abnormal video frames by treating regular and irregular video bags and video sections.We use the multi-objective whale optimization algorithm to optimize the entire process and get the best results.The performance parameters such as accuracy,precision,recall,and F-score are considered to evaluate the proposed technique using the Python simulation tool.Our simulation results show that the proposed method performs better than the conventional methods on the public live video dataset.展开更多
Surface coating is a critical procedure in the case of maintenance engineering. Ceramic coating of the wear areas is of the best practice which substantially enhances the Mean Time between Failure (MTBF). EN24 is a co...Surface coating is a critical procedure in the case of maintenance engineering. Ceramic coating of the wear areas is of the best practice which substantially enhances the Mean Time between Failure (MTBF). EN24 is a commercial grade alloy which is used for various industrial applications like sleeves, nuts, bolts, shafts, etc. EN24 is having comparatively low corrosion resistance, and ceramic coating of the wear and corroding areas of such parts is a best followed practice which highly improves the frequent failures. The coating quality mainly depends on the coating thickness, surface roughness and coating hardness which finally decides the operability. This paper describes an experimental investigation to effectively optimize the Atmospheric Plasma Spray process input parameters of Al<sub>2</sub>O<sub>3</sub>-40% TiO<sub>2</sub> coatings to get the best quality of coating on EN24 alloy steel substrate. The experiments are conducted with an Orthogonal Array (OA) design of experiments (DoE). In the current experiment, critical input parameters are considered and some of the vital output parameters are monitored accordingly and separate mathematical models are generated using regression analysis. The Analytic Hierarchy Process (AHP) method is used to generate weights for the individual objective functions and based on that, a combined objective function is made. An advanced optimization method, Teaching-Learning-Based Optimization algorithm (TLBO), is practically utilized to the combined objective function to optimize the values of input parameters to get the best output parameters. Confirmation tests are also conducted and their output results are compared with predicted values obtained through mathematical models. The dominating effects of Al<sub>2</sub>O<sub>3</sub>-40% TiO<sub>2</sub> spray parameters on output parameters: surface roughness, coating thickness and coating hardness are discussed in detail. It is concluded that the input parameters variation directly affects the characteristics of output parameters and any number of input as well as output parameters can be easily optimized using the current approach.展开更多
Urban block form significantly impacts energy and environmental performance.Therefore,optimizing urban block design in the early stages contributes to enhancing urban energy efficiency and environmental sustainability...Urban block form significantly impacts energy and environmental performance.Therefore,optimizing urban block design in the early stages contributes to enhancing urban energy efficiency and environmental sustainability.However,widely used multi-objective optimization methods based on performance simulation face the challenges of high computational loads and low efficiency.This study introduces a framework using machine learning,especially the XGBoost model,to accelerate multi-objective optimization of energy-efficient urban block forms.A residential block in Nanjing serves as the case study.The framework commences with a parametric block form model driven by design variables,focusing on minimizing building energy consumption(EUI),maximizing photovoltaic energy generation(PVE)and outdoor sunlight hours(SH).Data generated through Latin Hypercube Sampling and performance simulations inform the model training.Through training and hyperparameter tuning,XGBoost’s predictive accuracy was validated against artificial neural network(ANN),support vector machine(SVM),and random forest(RF)models.Subsequently,XGBoost replaced traditional performance simulations,conducting multi-objective optimization via the NSGA-II algorithm.Results showcase the framework’s significant acceleration of the optimization process,improving computational efficiency by over 420 times and producing 185 Pareto optimal solutions with improved performance metrics.SHAP analysis highlighted shape factor(SF),building density(BD),and building orientation(BO)as key morphological parameters influencing EUI,PVE,and SH.This study presents an efficient approach to energy-efficient urban block design,contributing valuable insights for sustainable urban development.展开更多
Neural network pruning is a popular approach to reducing the computational complexity of deep neural networks.In recent years,as growing evidence shows that conventional network pruning methods employ inappropriate pr...Neural network pruning is a popular approach to reducing the computational complexity of deep neural networks.In recent years,as growing evidence shows that conventional network pruning methods employ inappropriate proxy metrics,and as new types of hardware become increasingly available,hardware-aware network pruning that incorporates hardware characteristics in the loop of network pruning has gained growing attention,Both network accuracy and hardware efficiency(latency,memory consumption,etc.)are critical objectives to the success of network pruning,but the conflict between the multiple objectives makes it impossible to find a single optimal solution.Previous studies mostly convert the hardware-aware network pruning to optimization problems with a single objective.In this paper,we propose to solve the hardware-aware network pruning problem with Multi-Objective Evolutionary Algorithms(MOEAs).Specifically,we formulate the problem as a multi-objective optimization problem,and propose a novel memetic MOEA,namely HAMP,that combines an efficient portfoliobased selection and a surrogate-assisted local search,to solve it.Empirical studies demonstrate the potential of MOEAs in providing simultaneously a set of alternative solutions and the superiority of HAMP compared to the state-of-the-art hardware-aware network pruning method.展开更多
基金Supported by National Natural Science Foundation of China(Grant Nos.51275366,50875190,51305311)Specialized Research Fund for the Doctoral Program of Higher Education of China(Grant No.20134219110002)
文摘Due to the NP-hardness of the two-sided assembly line balancing （TALB） problem, multiple constraints existing in real applications are less studied, especially when one task is involved with several constraints. In this paper, an effective hybrid algorithm is proposed to address the TALB problem with multiple constraints （TALB-MC）. Considering the discrete attribute of TALB-MC and the continuous attribute of the standard teaching-learning-based optimization （TLBO） algorithm, the random-keys method is hired in task permutation representation, for the purpose of bridging the gap between them. Subsequently, a special mechanism for handling multiple constraints is developed. In the mechanism, the directions constraint of each task is ensured by the direction check and adjustment. The zoning constraints and the synchronism constraints are satisfied by teasing out the hidden correlations among constraints. The positional constraint is allowed to be violated to some extent in decoding and punished in cost fimction. Finally, with the TLBO seeking for the global optimum, the variable neighborhood search （VNS） is further hybridized to extend the local search space. The experimental results show that the proposed hybrid algorithm outperforms the late acceptance hill-climbing algorithm （LAHC） for TALB-MC in most cases, especially for large-size problems with multiple constraints, and demonstrates well balance between the exploration and the exploitation. This research proposes an effective and efficient algorithm for solving TALB-MC problem by hybridizing the TLBO and VNS.
基金the National Natural Science Foundation of China(62076225,62073300)the Natural Science Foundation for Distinguished Young Scholars of Hubei(2019CFA081)。
文摘Solving constrained multi-objective optimization problems with evolutionary algorithms has attracted considerable attention.Various constrained multi-objective optimization evolutionary algorithms(CMOEAs)have been developed with the use of different algorithmic strategies,evolutionary operators,and constraint-handling techniques.The performance of CMOEAs may be heavily dependent on the operators used,however,it is usually difficult to select suitable operators for the problem at hand.Hence,improving operator selection is promising and necessary for CMOEAs.This work proposes an online operator selection framework assisted by Deep Reinforcement Learning.The dynamics of the population,including convergence,diversity,and feasibility,are regarded as the state;the candidate operators are considered as actions;and the improvement of the population state is treated as the reward.By using a Q-network to learn a policy to estimate the Q-values of all actions,the proposed approach can adaptively select an operator that maximizes the improvement of the population according to the current state and thereby improve the algorithmic performance.The framework is embedded into four popular CMOEAs and assessed on 42 benchmark problems.The experimental results reveal that the proposed Deep Reinforcement Learning-assisted operator selection significantly improves the performance of these CMOEAs and the resulting algorithm obtains better versatility compared to nine state-of-the-art CMOEAs.
基金Project(62073342)supported by the National Natural Science Foundation of ChinaProject(2014 AA 041803)supported by the Hi-tech Research and Development Program of China。
文摘The operation variables,including feed rate of ore slurry,caustic solution and live steams in the double-stream alumina digestion process,determine the product quality,process costs and the environment pollution.Previously,they were set by the technical workers according to the offline analysis results and an empirical formula,which leads to unstable process indices and high consumption frequently.So,a multi-objective optimization model is built to maintain the balance between resource consumptions and process indices by taking technical indices and energy efficiency as objectives,where the key technical indices are predicted based on the digestion kinetics of diaspore.A multi-objective state transition algorithm(MOSTA)is improved to solve the problem,in which a self-adaptive strategy is applied to dynamically adjust the operator factors of the MOSTA and dynamic infeasible threshold is used to handle constraints to enhance searching efficiency and ability of the algorithm.Then a rule based strategy is designed to make the final decision from the Pareto frontiers.The method is integrated into an optimal control system for the industrial digestion process and tested in the actual production.Results show that the proposed method can achieve the technical target while reducing the energy consumption.
基金supported by the National Natural Science Foundation of China(Grant Nos.52090081,52079068)the State Key Laboratory of Hydroscience and Hydraulic Engineering(Grant No.2021-KY-04).
文摘Accurate determination of rock mass parameters is essential for ensuring the accuracy of numericalsimulations. Displacement back-analysis is the most widely used method;however, the reliability of thecurrent approaches remains unsatisfactory. Therefore, in this paper, a multistage rock mass parameterback-analysis method, that considers the construction process and displacement losses is proposed andimplemented through the coupling of numerical simulation, auto-machine learning (AutoML), andmulti-objective optimization algorithms (MOOAs). First, a parametric modeling platform for mechanizedtwin tunnels is developed, generating a dataset through extensive numerical simulations. Next, theAutoML method is utilized to establish a surrogate model linking rock parameters and displacements.The tunnel construction process is divided into multiple stages, transforming the rock mass parameterback-analysis into a multi-objective optimization problem, for which multi-objective optimization algorithmsare introduced to obtain the rock mass parameters. The newly proposed rock mass parameterback-analysis method is validated in a mechanized twin tunnel project, and its accuracy and effectivenessare demonstrated. Compared with traditional single-stage back-analysis methods, the proposedmodel decreases the average absolute percentage error from 12.73% to 4.34%, significantly improving theaccuracy of the back-analysis. Moreover, although the accuracy of back analysis significantly increaseswith the number of construction stages considered, the back analysis time is acceptable. This studyprovides a new method for displacement back analysis that is efficient and accurate, thereby paving theway for precise parameter determination in numerical simulations.
文摘Workload balancing in cloud computing is not yet resolved,particularly considering Infrastructure as a Service(IaaS)in the cloud network.The problem of being underloaded or overloaded should not occur at the time of the server or host accessing the cloud which may lead to create system crash problem.Thus,to resolve these existing problems,an efficient task scheduling algorithm is required for distributing the tasks over the entire feasible resources,which is termed load balancing.The load balancing approach assures that the entire Virtual Machines(VMs)are utilized appropriately.So,it is highly essential to develop a load-balancing model in a cloud environment based on machine learning and optimization strategies.Here,the computing and networking data is utilized for the analysis to observe the traffic as well as performance patterns.The acquired data is offered to the machine learning decision to select the right server by predicting the performance effectively by employing an Optimal Kernel-based Extreme Learning Machine(OK-ELM)and their parameter is tuned by the developed hybrid approach Population Size-based Mud Ring Tunicate Swarm Algorithm(PS-MRTSA).Further,effective scheduling is performed to resolve the load balancing issues by employing the developed model MR-TSA.Here,the developed approach effectively resolves the multi-objective constraints such as Response time,Resource cost,and energy consumption.Thus,the recommended load balancing model securesan enhanced performance rate than the traditional approaches over several experimental analyses.
基金supported in part by the National Natural Science Foundation of China(61806051,61903078)Natural Science Foundation of Shanghai(20ZR1400400)+2 种基金Agricultural Project of the Shanghai Committee of Science and Technology(16391902800)the Fundamental Research Funds for the Central Universities(2232020D-48)the Project of the Humanities and Social Sciences on Young Fund of the Ministry of Education in China(Research on swarm intelligence collaborative robust optimization scheduling for high-dimensional dynamic decisionmaking system(20YJCZH052))。
文摘Evolutionary algorithms have been shown to be very successful in solving multi-objective optimization problems(MOPs).However,their performance often deteriorates when solving MOPs with irregular Pareto fronts.To remedy this issue,a large body of research has been performed in recent years and many new algorithms have been proposed.This paper provides a comprehensive survey of the research on MOPs with irregular Pareto fronts.We start with a brief introduction to the basic concepts,followed by a summary of the benchmark test problems with irregular problems,an analysis of the causes of the irregularity,and real-world optimization problems with irregular Pareto fronts.Then,a taxonomy of the existing methodologies for handling irregular problems is given and representative algorithms are reviewed with a discussion of their strengths and weaknesses.Finally,open challenges are pointed out and a few promising future directions are suggested.
基金supported by the National Key R&D Plan(2020YFB1712902)the National Natural Science Foundation of China(52075036).
文摘In order to solve the flexible job shop scheduling problem with variable batches,we propose an improved multiobjective optimization algorithm,which combines the idea of inverse scheduling.First,a flexible job shop problem with the variable batches scheduling model is formulated.Second,we propose a batch optimization algorithm with inverse scheduling in which the batch size is adjusted by the dynamic feedback batch adjusting method.Moreover,in order to increase the diversity of the population,two methods are developed.One is the threshold to control the neighborhood updating,and the other is the dynamic clustering algorithm to update the population.Finally,a group of experiments are carried out.The results show that the improved multi-objective optimization algorithm can ensure the diversity of Pareto solutions effectively,and has effective performance in solving the flexible job shop scheduling problem with variable batches.
文摘For training the present Neural Network(NN)models,the standard technique is to utilize decaying Learning Rates(LR).While the majority of these techniques commence with a large LR,they will decay multiple times over time.Decaying has been proved to enhance generalization as well as optimization.Other parameters,such as the network’s size,the number of hidden layers,drop-outs to avoid overfitting,batch size,and so on,are solely based on heuristics.This work has proposed Adaptive Teaching Learning Based(ATLB)Heuristic to identify the optimal hyperparameters for diverse networks.Here we consider three architec-tures Recurrent Neural Networks(RNN),Long Short Term Memory(LSTM),Bidirectional Long Short Term Memory(BiLSTM)of Deep Neural Networks for classification.The evaluation of the proposed ATLB is done through the various learning rate schedulers Cyclical Learning Rate(CLR),Hyperbolic Tangent Decay(HTD),and Toggle between Hyperbolic Tangent Decay and Triangular mode with Restarts(T-HTR)techniques.Experimental results have shown the performance improvement on the 20Newsgroup,Reuters Newswire and IMDB dataset.
文摘This research paper presents a comprehensive investigation into the effectiveness of the DeepSurNet-NSGA II(Deep Surrogate Model-Assisted Non-dominated Sorting Genetic Algorithm II)for solving complex multiobjective optimization problems,with a particular focus on robotic leg-linkage design.The study introduces an innovative approach that integrates deep learning-based surrogate models with the robust Non-dominated Sorting Genetic Algorithm II,aiming to enhance the efficiency and precision of the optimization process.Through a series of empirical experiments and algorithmic analyses,the paper demonstrates a high degree of correlation between solutions generated by the DeepSurNet-NSGA II and those obtained from direct experimental methods,underscoring the algorithm’s capability to accurately approximate the Pareto-optimal frontier while significantly reducing computational demands.The methodology encompasses a detailed exploration of the algorithm’s configuration,the experimental setup,and the criteria for performance evaluation,ensuring the reproducibility of results and facilitating future advancements in the field.The findings of this study not only confirm the practical applicability and theoretical soundness of the DeepSurNet-NSGA II in navigating the intricacies of multi-objective optimization but also highlight its potential as a transformative tool in engineering and design optimization.By bridging the gap between complex optimization challenges and achievable solutions,this research contributes valuable insights into the optimization domain,offering a promising direction for future inquiries and technological innovations.
基金co-supported by the National Natural Science Foundation of China(No.U20B2056)the office of Military and Civilian Integration Devel-opment Committee of Shanghai(No.2020-jmrh1-kj25)the X LAB Joint Innovation Foundation with the Second Academy of CASIC(No.21GFC-JJ02-322)。
文摘Recently,mega Low Earth Orbit(LEO)Satellite Network(LSN)systems have gained more and more attention due to low latency,broadband communications and global coverage for ground users.One of the primary challenges for LSN systems with inter-satellite links is the routing strategy calculation and maintenance,due to LSN constellation scale and dynamic network topology feature.In order to seek an efficient routing strategy,a Q-learning-based dynamic distributed Routing scheme for LSNs(QRLSN)is proposed in this paper.To achieve low end-toend delay and low network traffic overhead load in LSNs,QRLSN adopts a multi-objective optimization method to find the optimal next hop for forwarding data packets.Experimental results demonstrate that the proposed scheme can effectively discover the initial routing strategy and provide long-term Quality of Service(QoS)optimization during the routing maintenance process.In addition,comparison results demonstrate that QRLSN is superior to the virtual-topology-based shortest path routing algorithm.
基金Projects(50974039,50634030)supported by the National Natural Science Foundation of China
文摘A multi-objective optimization model for draft scheduling of hot strip mill was presented, rolling power minimizing, rolling force ratio distribution and good strip shape as the objective functions. A multi-objective differential evolution algorithm based on decomposition （MODE/D）. The two-objective and three-objective optimization experiments were performed respectively to demonstrate the optimal solutions of trade-off. The simulation results show that MODE/D can obtain a good Pareto-optimal front, which suggests a series of alternative solutions to draft scheduling. The extreme Pareto solutions are found feasible and the centres of the Pareto fronts give a good compromise. The conflict exists between each two ones of three objectives. The final optimal solution is selected from the Pareto-optimal front by the importance of objectives, and it can achieve a better performance in all objective dimensions than the empirical solutions. Finally, the practical application cases confirm the feasibility of the multi-objective approach, and the optimal solutions can gain a better rolling stability than the empirical solutions, and strip flatness decreases from （0± 63） IU to （0±45） IU in industrial production.
基金supported by the National Natural Science Foundation of China(62076185,U1809209)supported by Zhejiang Provincial Natural Science Foundation of China(LY21F020030)+1 种基金Wenzhou Major Scientific and Technological Innovation Project(ZY2019019)Wenzhou Science and Technology Bureau(2018ZG016)。
文摘The Salp Swarm Algorithm(SSA)may have trouble in dropping into stagnation as a kind of swarm intelligence method.This paper developed an adaptive barebones salp swarm algorithm with quasi-oppositional-based learning to compensate for the above weakness called QBSSA.In the proposed QBSSA,an adaptive barebones strategy can help to reach both accurate convergence speed and high solution quality;quasi-oppositional-based learning can make the population away from traping into local optimal and expand the search space.To estimate the performance of the presented method,a series of tests are performed.Firstly,CEC 2017 benchmark test suit is used to test the ability to solve the high dimensional and multimodal problems;then,based on QBSSA,an improved Kernel Extreme Learning Machine(KELM)model,named QBSSA–KELM,is built to handle medical disease diagnosis problems.All the test results and discussions state clearly that the QBSSA is superior to and very competitive to all the compared algorithms on both convergence speed and solutions accuracy.
基金supported by the National Key Research and Development Program of China(2016YFD0700605)the Fundamental Research Funds for the Central Universities(JZ2016HGBZ1035)the Anhui University Natural Science Research Project(KJ2017A891)
文摘This research provides academic and practical contributions. From a theoretical standpoint, a hybrid harmony search(HS)algorithm, namely the oppositional global-based HS(OGHS), is proposed for solving the multi-objective flexible job-shop scheduling problems(MOFJSPs) to minimize makespan, total machine workload and critical machine workload. An initialization program embedded in opposition-based learning(OBL) is developed for enabling the individuals to scatter in a well-distributed manner in the initial harmony memory(HM). In addition, the recursive halving technique based on opposite number is employed for shrinking the neighbourhood space in the searching phase of the OGHS. From a practice-related standpoint, a type of dual vector code technique is introduced for allowing the OGHS algorithm to adapt the discrete nature of the MOFJSP. Two practical techniques, namely Pareto optimality and technique for order preference by similarity to an ideal solution(TOPSIS), are implemented for solving the MOFJSP.Furthermore, the algorithm performance is tested by using different strategies, including OBL and recursive halving, and the OGHS is compared with existing algorithms in the latest studies.Experimental results on representative examples validate the performance of the proposed algorithm for solving the MOFJSP.
文摘SS304 is a commercial grade stainless steel which is used for various engineering applications like shafts, guides, jigs, fixtures, etc. Ceramic coating of the wear areas of such parts is a regular practice which significantly enhances the Mean Time Between Failure (MTBF). The final coating quality depends mainly on the coating thickness, surface roughness and hardness which ultimately decides the life. This paper presents an experimental study to effectively optimize the Atmospheric Plasma Spray (APS) process input parameters of Al<sub>2</sub>O<sub>3</sub>-40% TiO2 ceramic coatings to get the best quality of coating on commercial SS304 substrate. The experiments are conducted with a three-level L<sub>18</sub> Orthogonal Array (OA) Design of Experiments (DoE). Critical input parameters considered are: spray nozzle distance, substrate rotating speed, current of the arc, carrier gas flow and coating powder flow rate. The surface roughness, coating thickness and hardness are considered as the output parameters. Mathematical models are generated using regression analysis for individual output parameters. The Analytic Hierarchy Process (AHP) method is applied to generate weights for the individual objective functions and a combined objective function is generated. An advanced optimization method, Teaching-Learning-Based Optimization algorithm (TLBO), is applied to the combined objective function to optimize the values of input parameters to get the best output parameters and confirmation tests are conducted based on that. The significant effects of spray parameters on surface roughness, coating thickness and coating hardness are studied in detail.
基金supported by the National Natural Science Foundation of China under Grant No.61973042Beijing Natural Science Foundation under Grant No.1202020。
文摘Financial market has systemic complexity and uncertainty.For investors,return and risk often coexist.How to rationally allocate funds into different assets and achieve excess returns with effectively controlling risk are main problems to be solved in the field of portfolio optimization(PO).At present,due to the influence of modeling and algorithm solving,the PO models established by many researchers are still mainly focused on single-stage single-objective models or single-stage multiobjective models.PO is actually considered as a multi-stage multi-objective optimization problem in real investment scenarios.It is more difficult than the previous single-stage PO model for meeting the realistic requirements.In this paper,the authors proposed a mean-improved stable tail adjusted return ratio-maximum drawdown rate(M-ISTARR-MD)PO model which effectively characterizes the real investment scenario.In order to solve the multi-stage multi-objective PO model with complex multi-constraints,the authors designed a multi-stage constrained multi-objective evolutionary algorithm with orthogonal learning(MSCMOEA-OL).Comparing with four well-known intelligence algorithms,the MSCMOEA-OL algorithm has competitive advantages in solving the M-ISTARR-MD model on the proposed constructed carbon neutral stock dataset.This paper provides a new way to construct and solve the complex PO model.
文摘For increasing the overall performance of modem manufacturing systems, effective integration of process planning and scheduling functions has been an important area of consideration among researchers. Owing to the complexity of handling process planning and scheduling simultaneously, most of the research work has been limited to solving the integrated process planning and scheduling （IPPS） problem for a single objective function. As there are many conflicting objectives when dealing with process planning and scheduling, real world problems cannot be fully captured considering only a single objective for optimization. Therefore considering multi-objective IPPS （MOIPPS） problem is inevitable. Unfortunately, only a handful of research papers are available on solving MOIPPS problem. In this paper, an optimization algorithm for solving MOIPPS problem is presented. The proposed algorithm uses a set of dispatch- ing rules coupled with priority assignment to optimize the IPPS problem for various objectives like makespan, total machine load, total tardiness, etc. A fixed sized external archive coupled with a crowding distance mechanism is used to store and maintain the non-dominated solutions. To compare the results with other algorithms, a C-matric based method has been used. Instances from four recent papers have been solved to demonstrate the effectiveness of the proposed algorithm. The experimental results show that the proposed method is an efficient approach for solving the MOIPPS problem.
文摘Crowd Anomaly Detection has become a challenge in intelligent video surveillance system and security.Intelligent video surveillance systems make extensive use of data mining,machine learning and deep learning methods.In this paper a novel approach is proposed to identify abnormal occurrences in crowded situations using deep learning.In this approach,Adaptive GoogleNet Neural Network Classifier with Multi-Objective Whale Optimization Algorithm are applied to predict the abnormal video frames in the crowded scenes.We use multiple instance learning(MIL)to dynamically develop a deep anomalous ranking framework.This technique predicts higher anomalous values for abnormal video frames by treating regular and irregular video bags and video sections.We use the multi-objective whale optimization algorithm to optimize the entire process and get the best results.The performance parameters such as accuracy,precision,recall,and F-score are considered to evaluate the proposed technique using the Python simulation tool.Our simulation results show that the proposed method performs better than the conventional methods on the public live video dataset.
文摘Surface coating is a critical procedure in the case of maintenance engineering. Ceramic coating of the wear areas is of the best practice which substantially enhances the Mean Time between Failure (MTBF). EN24 is a commercial grade alloy which is used for various industrial applications like sleeves, nuts, bolts, shafts, etc. EN24 is having comparatively low corrosion resistance, and ceramic coating of the wear and corroding areas of such parts is a best followed practice which highly improves the frequent failures. The coating quality mainly depends on the coating thickness, surface roughness and coating hardness which finally decides the operability. This paper describes an experimental investigation to effectively optimize the Atmospheric Plasma Spray process input parameters of Al<sub>2</sub>O<sub>3</sub>-40% TiO<sub>2</sub> coatings to get the best quality of coating on EN24 alloy steel substrate. The experiments are conducted with an Orthogonal Array (OA) design of experiments (DoE). In the current experiment, critical input parameters are considered and some of the vital output parameters are monitored accordingly and separate mathematical models are generated using regression analysis. The Analytic Hierarchy Process (AHP) method is used to generate weights for the individual objective functions and based on that, a combined objective function is made. An advanced optimization method, Teaching-Learning-Based Optimization algorithm (TLBO), is practically utilized to the combined objective function to optimize the values of input parameters to get the best output parameters. Confirmation tests are also conducted and their output results are compared with predicted values obtained through mathematical models. The dominating effects of Al<sub>2</sub>O<sub>3</sub>-40% TiO<sub>2</sub> spray parameters on output parameters: surface roughness, coating thickness and coating hardness are discussed in detail. It is concluded that the input parameters variation directly affects the characteristics of output parameters and any number of input as well as output parameters can be easily optimized using the current approach.
基金sponsored by the National Natural Science Foundation of China(NSFC No.52478011,No.52378046).
文摘Urban block form significantly impacts energy and environmental performance.Therefore,optimizing urban block design in the early stages contributes to enhancing urban energy efficiency and environmental sustainability.However,widely used multi-objective optimization methods based on performance simulation face the challenges of high computational loads and low efficiency.This study introduces a framework using machine learning,especially the XGBoost model,to accelerate multi-objective optimization of energy-efficient urban block forms.A residential block in Nanjing serves as the case study.The framework commences with a parametric block form model driven by design variables,focusing on minimizing building energy consumption(EUI),maximizing photovoltaic energy generation(PVE)and outdoor sunlight hours(SH).Data generated through Latin Hypercube Sampling and performance simulations inform the model training.Through training and hyperparameter tuning,XGBoost’s predictive accuracy was validated against artificial neural network(ANN),support vector machine(SVM),and random forest(RF)models.Subsequently,XGBoost replaced traditional performance simulations,conducting multi-objective optimization via the NSGA-II algorithm.Results showcase the framework’s significant acceleration of the optimization process,improving computational efficiency by over 420 times and producing 185 Pareto optimal solutions with improved performance metrics.SHAP analysis highlighted shape factor(SF),building density(BD),and building orientation(BO)as key morphological parameters influencing EUI,PVE,and SH.This study presents an efficient approach to energy-efficient urban block design,contributing valuable insights for sustainable urban development.
基金the National Natural Science Foundation of China(62106098)the Stable Support Plan Program of Shenzhen Natural Science Fund(20200925154942002)the M0E University Scientific-Technological Innovation Plan Program.
文摘Neural network pruning is a popular approach to reducing the computational complexity of deep neural networks.In recent years,as growing evidence shows that conventional network pruning methods employ inappropriate proxy metrics,and as new types of hardware become increasingly available,hardware-aware network pruning that incorporates hardware characteristics in the loop of network pruning has gained growing attention,Both network accuracy and hardware efficiency(latency,memory consumption,etc.)are critical objectives to the success of network pruning,but the conflict between the multiple objectives makes it impossible to find a single optimal solution.Previous studies mostly convert the hardware-aware network pruning to optimization problems with a single objective.In this paper,we propose to solve the hardware-aware network pruning problem with Multi-Objective Evolutionary Algorithms(MOEAs).Specifically,we formulate the problem as a multi-objective optimization problem,and propose a novel memetic MOEA,namely HAMP,that combines an efficient portfoliobased selection and a surrogate-assisted local search,to solve it.Empirical studies demonstrate the potential of MOEAs in providing simultaneously a set of alternative solutions and the superiority of HAMP compared to the state-of-the-art hardware-aware network pruning method.
