Traditional large-scale multi-objective optimization algorithms(LSMOEAs)encounter difficulties when dealing with sparse large-scale multi-objective optimization problems(SLM-OPs)where most decision variables are zero....Traditional large-scale multi-objective optimization algorithms(LSMOEAs)encounter difficulties when dealing with sparse large-scale multi-objective optimization problems(SLM-OPs)where most decision variables are zero.As a result,many algorithms use a two-layer encoding approach to optimize binary variable Mask and real variable Dec separately.Nevertheless,existing optimizers often focus on locating non-zero variable posi-tions to optimize the binary variables Mask.However,approxi-mating the sparse distribution of real Pareto optimal solutions does not necessarily mean that the objective function is optimized.In data mining,it is common to mine frequent itemsets appear-ing together in a dataset to reveal the correlation between data.Inspired by this,we propose a novel two-layer encoding learning swarm optimizer based on frequent itemsets(TELSO)to address these SLMOPs.TELSO mined the frequent terms of multiple particles with better target values to find mask combinations that can obtain better objective values for fast convergence.Experi-mental results on five real-world problems and eight benchmark sets demonstrate that TELSO outperforms existing state-of-the-art sparse large-scale multi-objective evolutionary algorithms(SLMOEAs)in terms of performance and convergence speed.展开更多
Sparse large-scale multi-objective optimization problems(SLMOPs)are common in science and engineering.However,the large-scale problem represents the high dimensionality of the decision space,requiring algorithms to tr...Sparse large-scale multi-objective optimization problems(SLMOPs)are common in science and engineering.However,the large-scale problem represents the high dimensionality of the decision space,requiring algorithms to traverse vast expanse with limited computational resources.Furthermore,in the context of sparse,most variables in Pareto optimal solutions are zero,making it difficult for algorithms to identify non-zero variables efficiently.This paper is dedicated to addressing the challenges posed by SLMOPs.To start,we introduce innovative objective functions customized to mine maximum and minimum candidate sets.This substantial enhancement dramatically improves the efficacy of frequent pattern mining.In this way,selecting candidate sets is no longer based on the quantity of nonzero variables they contain but on a higher proportion of nonzero variables within specific dimensions.Additionally,we unveil a novel approach to association rule mining,which delves into the intricate relationships between non-zero variables.This novel methodology aids in identifying sparse distributions that can potentially expedite reductions in the objective function value.We extensively tested our algorithm across eight benchmark problems and four real-world SLMOPs.The results demonstrate that our approach achieves competitive solutions across various challenges.展开更多
During the last three decades,evolutionary algorithms(EAs)have shown superiority in solving complex optimization problems,especially those with multiple objectives and non-differentiable landscapes.However,due to the ...During the last three decades,evolutionary algorithms(EAs)have shown superiority in solving complex optimization problems,especially those with multiple objectives and non-differentiable landscapes.However,due to the stochastic search strategies,the performance of most EAs deteriorates drastically when handling a large number of decision variables.To tackle the curse of dimensionality,this work proposes an efficient EA for solving super-large-scale multi-objective optimization problems with sparse optimal solutions.The proposed algorithm estimates the sparse distribution of optimal solutions by optimizing a binary vector for each solution,and provides a fast clustering method to highly reduce the dimensionality of the search space.More importantly,all the operations related to the decision variables only contain several matrix calculations,which can be directly accelerated by GPUs.While existing EAs are capable of handling fewer than 10000 real variables,the proposed algorithm is verified to be effective in handling 1000000 real variables.Furthermore,since the proposed algorithm handles the large number of variables via accelerated matrix calculations,its runtime can be reduced to less than 10%of the runtime of existing EAs.展开更多
Large-scale multi-objective optimization problems(LSMOPs)pose challenges to existing optimizers since a set of well-converged and diverse solutions should be found in huge search spaces.While evolutionary algorithms a...Large-scale multi-objective optimization problems(LSMOPs)pose challenges to existing optimizers since a set of well-converged and diverse solutions should be found in huge search spaces.While evolutionary algorithms are good at solving small-scale multi-objective optimization problems,they are criticized for low efficiency in converging to the optimums of LSMOPs.By contrast,mathematical programming methods offer fast convergence speed on large-scale single-objective optimization problems,but they have difficulties in finding diverse solutions for LSMOPs.Currently,how to integrate evolutionary algorithms with mathematical programming methods to solve LSMOPs remains unexplored.In this paper,a hybrid algorithm is tailored for LSMOPs by coupling differential evolution and a conjugate gradient method.On the one hand,conjugate gradients and differential evolution are used to update different decision variables of a set of solutions,where the former drives the solutions to quickly converge towards the Pareto front and the latter promotes the diversity of the solutions to cover the whole Pareto front.On the other hand,objective decomposition strategy of evolutionary multi-objective optimization is used to differentiate the conjugate gradients of solutions,and the line search strategy of mathematical programming is used to ensure the higher quality of each offspring than its parent.In comparison with state-of-the-art evolutionary algorithms,mathematical programming methods,and hybrid algorithms,the proposed algorithm exhibits better convergence and diversity performance on a variety of benchmark and real-world LSMOPs.展开更多
Large-scale multi-objective optimization problems(MOPs)that involve a large number of decision variables,have emerged from many real-world applications.While evolutionary algorithms(EAs)have been widely acknowledged a...Large-scale multi-objective optimization problems(MOPs)that involve a large number of decision variables,have emerged from many real-world applications.While evolutionary algorithms(EAs)have been widely acknowledged as a mainstream method for MOPs,most research progress and successful applications of EAs have been restricted to MOPs with small-scale decision variables.More recently,it has been reported that traditional multi-objective EAs(MOEAs)suffer severe deterioration with the increase of decision variables.As a result,and motivated by the emergence of real-world large-scale MOPs,investigation of MOEAs in this aspect has attracted much more attention in the past decade.This paper reviews the progress of evolutionary computation for large-scale multi-objective optimization from two angles.From the key difficulties of the large-scale MOPs,the scalability analysis is discussed by focusing on the performance of existing MOEAs and the challenges induced by the increase of the number of decision variables.From the perspective of methodology,the large-scale MOEAs are categorized into three classes and introduced respectively:divide and conquer based,dimensionality reduction based and enhanced search-based approaches.Several future research directions are also discussed.展开更多
The performance of analytical derivative and sparse matrix techniques applied to a traditional dense sequential quadratic programming (SQP) is studied, and the strategy utilizing those techniques is also presented.Com...The performance of analytical derivative and sparse matrix techniques applied to a traditional dense sequential quadratic programming (SQP) is studied, and the strategy utilizing those techniques is also presented.Computational results on two typical chemical optimization problems demonstrate significant enhancement in efficiency, which shows this strategy is promising and suitable for large-scale process optimization problems.展开更多
In this paper we report a sparse truncated Newton algorithm for handling large-scale simple bound nonlinear constrained minimixation problem. The truncated Newton method is used to update the variables with indices ou...In this paper we report a sparse truncated Newton algorithm for handling large-scale simple bound nonlinear constrained minimixation problem. The truncated Newton method is used to update the variables with indices outside of the active set, while the projected gradient method is used to update the active variables. At each iterative level, the search direction consists of three parts, one of which is a subspace truncated Newton direction, the other two are subspace gradient and modified gradient directions. The subspace truncated Newton direction is obtained by solving a sparse system of linear equations. The global convergence and quadratic convergence rate of the algorithm are proved and some numerical tests are given.展开更多
The large-scale multi-objective optimization algorithm(LSMOA),based on the grouping of decision variables,is an advanced method for handling high-dimensional decision variables.However,in practical problems,the intera...The large-scale multi-objective optimization algorithm(LSMOA),based on the grouping of decision variables,is an advanced method for handling high-dimensional decision variables.However,in practical problems,the interaction among decision variables is intricate,leading to large group sizes and suboptimal optimization effects;hence a large-scale multi-objective optimization algorithm based on weighted overlapping grouping of decision variables(MOEAWOD)is proposed in this paper.Initially,the decision variables are perturbed and categorized into convergence and diversity variables;subsequently,the convergence variables are subdivided into groups based on the interactions among different decision variables.If the size of a group surpasses the set threshold,that group undergoes a process of weighting and overlapping grouping.Specifically,the interaction strength is evaluated based on the interaction frequency and number of objectives among various decision variables.The decision variable with the highest interaction in the group is identified and disregarded,and the remaining variables are then reclassified into subgroups.Finally,the decision variable with the strongest interaction is added to each subgroup.MOEAWOD minimizes the interactivity between different groups and maximizes the interactivity of decision variables within groups,which contributed to the optimized direction of convergence and diversity exploration with different groups.MOEAWOD was subjected to testing on 18 benchmark large-scale optimization problems,and the experimental results demonstrate the effectiveness of our methods.Compared with the other algorithms,our method is still at an advantage.展开更多
With the load growth and the power grid expansion,the problem of short-circuit current(SCC)exceeding the secure limit in large-scale power grids has become more serious,which poses great challenge to the optimal secur...With the load growth and the power grid expansion,the problem of short-circuit current(SCC)exceeding the secure limit in large-scale power grids has become more serious,which poses great challenge to the optimal secure operation.Aiming at the SCC limitations,we use multiple back-toback voltage source converter based(B2B VSC)systems to separate a large-scale AC power grid into two asynchronous power grids.A multi-objective robust optimal secure operation model of large-scale power grid with multiple B2B VSC systems considering the SCC limitation is established based on the AC power flow equations.The decision variables include the on/off states of synchronous generators,power output,terminal voltage,transmission switching,bus sectionalization,and modulation ratios of B2B VSC systems.The influence of inner current sources of renewable energy generators on the system SCC is also considered.To improve the computational efficiency,a mixedinteger convex programming(MICP)framework based on convex relaxation methods including the inscribed N-sided approximation for the nonlinear SCC limitation constraints is proposed.Moreover,combined with the column-and-constraint generation(C&CG)algorithm,a method to directly solve the compromise optimal solution(COS)of the multi-objective robust optimal secure operation model is proposed.Finally,the effectiveness and computational efficiency of the proposed solution method is demonstrated by an actual 4407-bus provincial power grid and the modified IEEE 39-bus power grid,which can reduce the consumed CPU time of solving the COS by more than 90%and obtain a better COS.展开更多
A mathematical approach was proposed to investigate the impact of high penetration of large-scale photovoltaic park(LPP) on small-signal stability of a power network and design of hybrid controller for these units.A s...A mathematical approach was proposed to investigate the impact of high penetration of large-scale photovoltaic park(LPP) on small-signal stability of a power network and design of hybrid controller for these units.A systematic procedure was performed to obtain the complete model of a multi-machine power network including LPP.For damping of oscillations focusing on inter-area oscillatory modes,a hybrid controller for LPP was proposed.The performance of the suggested controller was tested using a 16-machine 5-area network.The results indicate that the proposed hybrid controller for LPP provides sufficient damping to the low-frequency modes of power system for a wide range of operating conditions.The method presented in this work effectively indentifies the impact of increased PV penetration and its controller on dynamic performance of multi-machine power network containing LPP.Simulation results demonstrate that the model presented can be used in designing of essential controllers for LPP.展开更多
Power system optimal dispatch with transient security constraints is commonly represented as transient securityconstrained optimal power flow(TSC-OPF).Deep reinforcement learning(DRL)-based TSC-OPF trains efficient de...Power system optimal dispatch with transient security constraints is commonly represented as transient securityconstrained optimal power flow(TSC-OPF).Deep reinforcement learning(DRL)-based TSC-OPF trains efficient decisionmaking agents that are adaptable to various scenarios and provide solution results quickly.However,due to the high dimensionality of the state space and action spaces,as well as the nonsmoothness of dynamic constraints,existing DRL-based TSCOPF solution methods face a significant challenge of the sparse reward problem.To address this issue,a fast-converging DRL method for optimal dispatch of large-scale power systems under transient security constraints is proposed in this paper.The Markov decision process(MDP)modeling of TSC-OPF is improved by reducing the observation space and smoothing the reward design,thus facilitating agent training.An improved deep deterministic policy gradient algorithm with curriculum learning,parallel exploration,and ensemble decision-making(DDPGCL-PE-ED)is introduced to drastically enhance the efficiency of agent training and the accuracy of decision-making.The effectiveness,efficiency,and accuracy of the proposed method are demonstrated through experiments in the IEEE 39-bus system and a practical 710-bus regional power grid.The source code of the proposed method is made public on GitHub.展开更多
基金supported by the Scientific Research Project of Xiang Jiang Lab(22XJ02003)the University Fundamental Research Fund(23-ZZCX-JDZ-28)+5 种基金the National Science Fund for Outstanding Young Scholars(62122093)the National Natural Science Foundation of China(72071205)the Hunan Graduate Research Innovation Project(ZC23112101-10)the Hunan Natural Science Foundation Regional Joint Project(2023JJ50490)the Science and Technology Project for Young and Middle-aged Talents of Hunan(2023TJ-Z03)the Science and Technology Innovation Program of Humnan Province(2023RC1002)。
文摘Traditional large-scale multi-objective optimization algorithms(LSMOEAs)encounter difficulties when dealing with sparse large-scale multi-objective optimization problems(SLM-OPs)where most decision variables are zero.As a result,many algorithms use a two-layer encoding approach to optimize binary variable Mask and real variable Dec separately.Nevertheless,existing optimizers often focus on locating non-zero variable posi-tions to optimize the binary variables Mask.However,approxi-mating the sparse distribution of real Pareto optimal solutions does not necessarily mean that the objective function is optimized.In data mining,it is common to mine frequent itemsets appear-ing together in a dataset to reveal the correlation between data.Inspired by this,we propose a novel two-layer encoding learning swarm optimizer based on frequent itemsets(TELSO)to address these SLMOPs.TELSO mined the frequent terms of multiple particles with better target values to find mask combinations that can obtain better objective values for fast convergence.Experi-mental results on five real-world problems and eight benchmark sets demonstrate that TELSO outperforms existing state-of-the-art sparse large-scale multi-objective evolutionary algorithms(SLMOEAs)in terms of performance and convergence speed.
基金support by the Open Project of Xiangjiang Laboratory(22XJ02003)the University Fundamental Research Fund(23-ZZCX-JDZ-28,ZK21-07)+5 种基金the National Science Fund for Outstanding Young Scholars(62122093)the National Natural Science Foundation of China(72071205)the Hunan Graduate Research Innovation Project(CX20230074)the Hunan Natural Science Foundation Regional Joint Project(2023JJ50490)the Science and Technology Project for Young and Middle-aged Talents of Hunan(2023TJZ03)the Science and Technology Innovation Program of Humnan Province(2023RC1002).
文摘Sparse large-scale multi-objective optimization problems(SLMOPs)are common in science and engineering.However,the large-scale problem represents the high dimensionality of the decision space,requiring algorithms to traverse vast expanse with limited computational resources.Furthermore,in the context of sparse,most variables in Pareto optimal solutions are zero,making it difficult for algorithms to identify non-zero variables efficiently.This paper is dedicated to addressing the challenges posed by SLMOPs.To start,we introduce innovative objective functions customized to mine maximum and minimum candidate sets.This substantial enhancement dramatically improves the efficacy of frequent pattern mining.In this way,selecting candidate sets is no longer based on the quantity of nonzero variables they contain but on a higher proportion of nonzero variables within specific dimensions.Additionally,we unveil a novel approach to association rule mining,which delves into the intricate relationships between non-zero variables.This novel methodology aids in identifying sparse distributions that can potentially expedite reductions in the objective function value.We extensively tested our algorithm across eight benchmark problems and four real-world SLMOPs.The results demonstrate that our approach achieves competitive solutions across various challenges.
基金This work was supported in part by the National Key Research and Development Program of China(2018AAA0100100)the National Natural Science Foundation of China(61822301,61876123,61906001)+2 种基金the Collaborative Innovation Program of Universities in Anhui Province(GXXT-2020-051)the Hong Kong Scholars Program(XJ2019035)Anhui Provincial Natural Science Foundation(1908085QF271).
文摘During the last three decades,evolutionary algorithms(EAs)have shown superiority in solving complex optimization problems,especially those with multiple objectives and non-differentiable landscapes.However,due to the stochastic search strategies,the performance of most EAs deteriorates drastically when handling a large number of decision variables.To tackle the curse of dimensionality,this work proposes an efficient EA for solving super-large-scale multi-objective optimization problems with sparse optimal solutions.The proposed algorithm estimates the sparse distribution of optimal solutions by optimizing a binary vector for each solution,and provides a fast clustering method to highly reduce the dimensionality of the search space.More importantly,all the operations related to the decision variables only contain several matrix calculations,which can be directly accelerated by GPUs.While existing EAs are capable of handling fewer than 10000 real variables,the proposed algorithm is verified to be effective in handling 1000000 real variables.Furthermore,since the proposed algorithm handles the large number of variables via accelerated matrix calculations,its runtime can be reduced to less than 10%of the runtime of existing EAs.
基金supported in part by the National Key Research and Development Program of China(2018AAA0100100)the National Natural Science Foundation of China(61906001,62136008,U21A20512)+1 种基金the Key Program of Natural Science Project of Educational Commission of Anhui Province(KJ2020A0036)Alexander von Humboldt Professorship for Artificial Intelligence Funded by the Federal Ministry of Education and Research,Germany。
文摘Large-scale multi-objective optimization problems(LSMOPs)pose challenges to existing optimizers since a set of well-converged and diverse solutions should be found in huge search spaces.While evolutionary algorithms are good at solving small-scale multi-objective optimization problems,they are criticized for low efficiency in converging to the optimums of LSMOPs.By contrast,mathematical programming methods offer fast convergence speed on large-scale single-objective optimization problems,but they have difficulties in finding diverse solutions for LSMOPs.Currently,how to integrate evolutionary algorithms with mathematical programming methods to solve LSMOPs remains unexplored.In this paper,a hybrid algorithm is tailored for LSMOPs by coupling differential evolution and a conjugate gradient method.On the one hand,conjugate gradients and differential evolution are used to update different decision variables of a set of solutions,where the former drives the solutions to quickly converge towards the Pareto front and the latter promotes the diversity of the solutions to cover the whole Pareto front.On the other hand,objective decomposition strategy of evolutionary multi-objective optimization is used to differentiate the conjugate gradients of solutions,and the line search strategy of mathematical programming is used to ensure the higher quality of each offspring than its parent.In comparison with state-of-the-art evolutionary algorithms,mathematical programming methods,and hybrid algorithms,the proposed algorithm exhibits better convergence and diversity performance on a variety of benchmark and real-world LSMOPs.
基金This work was supported by the Natural Science Foundation of China(Nos.61672478 and 61806090)the National Key Research and Development Program of China(No.2017YFB1003102)+4 种基金the Guangdong Provincial Key Laboratory(No.2020B121201001)the Shenzhen Peacock Plan(No.KQTD2016112514355531)the Guangdong-Hong Kong-Macao Greater Bay Area Center for Brain Science and Brain-inspired Intelligence Fund(No.2019028)the Fellowship of China Postdoctoral Science Foundation(No.2020M671900)the National Leading Youth Talent Support Program of China.
文摘Large-scale multi-objective optimization problems(MOPs)that involve a large number of decision variables,have emerged from many real-world applications.While evolutionary algorithms(EAs)have been widely acknowledged as a mainstream method for MOPs,most research progress and successful applications of EAs have been restricted to MOPs with small-scale decision variables.More recently,it has been reported that traditional multi-objective EAs(MOEAs)suffer severe deterioration with the increase of decision variables.As a result,and motivated by the emergence of real-world large-scale MOPs,investigation of MOEAs in this aspect has attracted much more attention in the past decade.This paper reviews the progress of evolutionary computation for large-scale multi-objective optimization from two angles.From the key difficulties of the large-scale MOPs,the scalability analysis is discussed by focusing on the performance of existing MOEAs and the challenges induced by the increase of the number of decision variables.From the perspective of methodology,the large-scale MOEAs are categorized into three classes and introduced respectively:divide and conquer based,dimensionality reduction based and enhanced search-based approaches.Several future research directions are also discussed.
基金Supported by the National Natural Science Foundation of China(No.29906010).
文摘The performance of analytical derivative and sparse matrix techniques applied to a traditional dense sequential quadratic programming (SQP) is studied, and the strategy utilizing those techniques is also presented.Computational results on two typical chemical optimization problems demonstrate significant enhancement in efficiency, which shows this strategy is promising and suitable for large-scale process optimization problems.
基金The research was supported by the State Education Grant for Retumed Scholars
文摘In this paper we report a sparse truncated Newton algorithm for handling large-scale simple bound nonlinear constrained minimixation problem. The truncated Newton method is used to update the variables with indices outside of the active set, while the projected gradient method is used to update the active variables. At each iterative level, the search direction consists of three parts, one of which is a subspace truncated Newton direction, the other two are subspace gradient and modified gradient directions. The subspace truncated Newton direction is obtained by solving a sparse system of linear equations. The global convergence and quadratic convergence rate of the algorithm are proved and some numerical tests are given.
基金supported in part by the Central Government Guides Local Science and TechnologyDevelopment Funds(Grant No.YDZJSX2021A038)in part by theNational Natural Science Foundation of China under(Grant No.61806138)in part by the China University Industry-University-Research Collaborative Innovation Fund(Future Network Innovation Research and Application Project)(Grant 2021FNA04014).
文摘The large-scale multi-objective optimization algorithm(LSMOA),based on the grouping of decision variables,is an advanced method for handling high-dimensional decision variables.However,in practical problems,the interaction among decision variables is intricate,leading to large group sizes and suboptimal optimization effects;hence a large-scale multi-objective optimization algorithm based on weighted overlapping grouping of decision variables(MOEAWOD)is proposed in this paper.Initially,the decision variables are perturbed and categorized into convergence and diversity variables;subsequently,the convergence variables are subdivided into groups based on the interactions among different decision variables.If the size of a group surpasses the set threshold,that group undergoes a process of weighting and overlapping grouping.Specifically,the interaction strength is evaluated based on the interaction frequency and number of objectives among various decision variables.The decision variable with the highest interaction in the group is identified and disregarded,and the remaining variables are then reclassified into subgroups.Finally,the decision variable with the strongest interaction is added to each subgroup.MOEAWOD minimizes the interactivity between different groups and maximizes the interactivity of decision variables within groups,which contributed to the optimized direction of convergence and diversity exploration with different groups.MOEAWOD was subjected to testing on 18 benchmark large-scale optimization problems,and the experimental results demonstrate the effectiveness of our methods.Compared with the other algorithms,our method is still at an advantage.
基金supported by the National Natural Science Foundation of China(No.51977080).
文摘With the load growth and the power grid expansion,the problem of short-circuit current(SCC)exceeding the secure limit in large-scale power grids has become more serious,which poses great challenge to the optimal secure operation.Aiming at the SCC limitations,we use multiple back-toback voltage source converter based(B2B VSC)systems to separate a large-scale AC power grid into two asynchronous power grids.A multi-objective robust optimal secure operation model of large-scale power grid with multiple B2B VSC systems considering the SCC limitation is established based on the AC power flow equations.The decision variables include the on/off states of synchronous generators,power output,terminal voltage,transmission switching,bus sectionalization,and modulation ratios of B2B VSC systems.The influence of inner current sources of renewable energy generators on the system SCC is also considered.To improve the computational efficiency,a mixedinteger convex programming(MICP)framework based on convex relaxation methods including the inscribed N-sided approximation for the nonlinear SCC limitation constraints is proposed.Moreover,combined with the column-and-constraint generation(C&CG)algorithm,a method to directly solve the compromise optimal solution(COS)of the multi-objective robust optimal secure operation model is proposed.Finally,the effectiveness and computational efficiency of the proposed solution method is demonstrated by an actual 4407-bus provincial power grid and the modified IEEE 39-bus power grid,which can reduce the consumed CPU time of solving the COS by more than 90%and obtain a better COS.
文摘A mathematical approach was proposed to investigate the impact of high penetration of large-scale photovoltaic park(LPP) on small-signal stability of a power network and design of hybrid controller for these units.A systematic procedure was performed to obtain the complete model of a multi-machine power network including LPP.For damping of oscillations focusing on inter-area oscillatory modes,a hybrid controller for LPP was proposed.The performance of the suggested controller was tested using a 16-machine 5-area network.The results indicate that the proposed hybrid controller for LPP provides sufficient damping to the low-frequency modes of power system for a wide range of operating conditions.The method presented in this work effectively indentifies the impact of increased PV penetration and its controller on dynamic performance of multi-machine power network containing LPP.Simulation results demonstrate that the model presented can be used in designing of essential controllers for LPP.
基金supported in part by the National Natural Science Foundation of China(No.52107104)。
文摘Power system optimal dispatch with transient security constraints is commonly represented as transient securityconstrained optimal power flow(TSC-OPF).Deep reinforcement learning(DRL)-based TSC-OPF trains efficient decisionmaking agents that are adaptable to various scenarios and provide solution results quickly.However,due to the high dimensionality of the state space and action spaces,as well as the nonsmoothness of dynamic constraints,existing DRL-based TSCOPF solution methods face a significant challenge of the sparse reward problem.To address this issue,a fast-converging DRL method for optimal dispatch of large-scale power systems under transient security constraints is proposed in this paper.The Markov decision process(MDP)modeling of TSC-OPF is improved by reducing the observation space and smoothing the reward design,thus facilitating agent training.An improved deep deterministic policy gradient algorithm with curriculum learning,parallel exploration,and ensemble decision-making(DDPGCL-PE-ED)is introduced to drastically enhance the efficiency of agent training and the accuracy of decision-making.The effectiveness,efficiency,and accuracy of the proposed method are demonstrated through experiments in the IEEE 39-bus system and a practical 710-bus regional power grid.The source code of the proposed method is made public on GitHub.
