This paper presents a parallel composite local search algorithm based on multiple search neighborhoods to solve a special kind of timetable problem. The new algorithm can also effectively solve those problems that can...This paper presents a parallel composite local search algorithm based on multiple search neighborhoods to solve a special kind of timetable problem. The new algorithm can also effectively solve those problems that can be solved by general local search algorithms. Experimental results show that the new algorithm can generate better solutions than general local search algorithms.展开更多
We apply the transitionless driving on the local adiabatic quantum search algorithm to speed up the adiabatic process. By studying quantum dynamics of the adiabatic search algorithm with the equivalent two-level syste...We apply the transitionless driving on the local adiabatic quantum search algorithm to speed up the adiabatic process. By studying quantum dynamics of the adiabatic search algorithm with the equivalent two-level system, we derive the transi- tionless driving Hamiltonian for the local adiabatic quantum search algorithm. We found that when adding a transitionless quantum driving term Ht~ （t） on the local adiabatic quantum search algorithm, the success rate is 1 exactly with arbitrary evolution time by solving the time-dependent Schr6dinger equation in eigen-picture. Moreover, we show the reason for the drastic decrease of the evolution time is that the driving Hamiltonian increases the lowest eigenvalues to a maximum of展开更多
Safety patrol inspection in chemical industrial parks is a complex multi-objective task with multiple degrees of freedom.Traditional pointer instruments with advantages like high reliability and strong adaptability to...Safety patrol inspection in chemical industrial parks is a complex multi-objective task with multiple degrees of freedom.Traditional pointer instruments with advantages like high reliability and strong adaptability to harsh environment,are widely applied in such parks.However,they rely on manual readings which have problems like heavy patrol workload,high labor cost,high false positives/negatives and poor timeliness.To address the above problems,this study proposes a path planning method for robot patrol in chemical industrial parks,where a path optimization model based on improved iterated local search and random variable neighborhood descent(ILS-RVND)algorithm is established by integrating the actual requirements of patrol tasks in chemical industrial parks.Further,the effectiveness of the model and algorithm is verified by taking real park data as an example.The results show that compared with GA and ILS-RVND,the improved algorithm reduces quantification cost by about 24%and saves patrol time by about 36%.Apart from shortening the patrol time of robots,optimizing their patrol path and reducing their maintenance loss,the proposed algorithm also avoids the untimely patrol of robots and enhances the safety factor of equipment.展开更多
An optimized volt-ampere reactive(VAR)control framework is proposed for transmission-level power systems to simultaneously mitigate voltage deviations and active-power losses through coordinated control of large-scale...An optimized volt-ampere reactive(VAR)control framework is proposed for transmission-level power systems to simultaneously mitigate voltage deviations and active-power losses through coordinated control of large-scale wind/solar farms with shunt static var generators(SVGs).The model explicitly represents reactive-power regulation characteristics of doubly-fed wind turbines and PV inverters under real-time meteorological conditions,and quantifies SVG high-speed compensation capability,enabling seamless transition from localized VAR management to a globally coordinated strategy.An enhanced adaptive gain-sharing knowledge optimizer(AGSK-SD)integrates simulated annealing and diversity maintenance to autonomously tune voltage-control actions,renewable source reactive-power set-points,and SVG output.The algorithm adaptively modulates knowledge factors and ratios across search phases,performs SA-based fine-grained local exploitation,and periodically re-injects population diversity to prevent premature convergence.Comprehensive tests on IEEE 9-bus and 39-bus systems demonstrate AGSK-SD’s superiority over NSGA-II and MOPSO in hypervolume(HV),inverse generative distance(IGD),and spread metrics while maintaining acceptable computational burden.The method reduces network losses from 2.7191 to 2.15 MW(20.79%reduction)and from 15.1891 to 11.22 MW(26.16%reduction)in the 9-bus and 39-bus systems respectively.Simultaneously,the cumulative voltage-deviation index decreases from 0.0277 to 3.42×10^(−4) p.u.(98.77%reduction)in the 9-bus system,and from 0.0556 to 0.0107 p.u.(80.76%reduction)in the 39-bus system.These improvements demonstrate significant suppression of line losses and voltage fluctuations.Comparative analysis with traditional heuristic optimization algorithms confirms the superior performance of the proposed approach.展开更多
Recently,learned heuristics have been widely applied to solve combinatorial optimization problems(e.g.,traveling salesman problem(TSP)).However,the scalability of these learning-based methods hinders the applications ...Recently,learned heuristics have been widely applied to solve combinatorial optimization problems(e.g.,traveling salesman problem(TSP)).However,the scalability of these learning-based methods hinders the applications in practical scenarios.Specifically,models pre-trained on the small-scale data generalize poorly to large-scale problems.Moreover,learning the heuristics directly for large-scale problems costs tremendous time and space.To extend the scalability of learned heuristics on TSP,we propose a Hierarchical neural framework for solving large-scale traveling salesman problems(HiTSPs)based on a divide-and-conquer strategy.In particular,the HiTSP framework first divides the large-scale problem into small-scale subproblems by node clustering.Each subproblem is conquered by a modified pointer network learned from reinforcement learning.The tour of the original TSP is constructed by linking solutions of subproblems and optimized by a novel segmented local search algorithm.Notably,the segmented local search algorithm leverages the node clustering information to prune many unnecessary operations and significantly reduces the complexity in theory.Extensive experiments show that HiTSP outperforms state-of-the-art learning-based methods and Google OR-Tools in large-scale cases.Moreover,compared to the best heuristic algorithms,HiTSP has a significant advantage in efficiency for large-scale TSP problems.展开更多
文摘This paper presents a parallel composite local search algorithm based on multiple search neighborhoods to solve a special kind of timetable problem. The new algorithm can also effectively solve those problems that can be solved by general local search algorithms. Experimental results show that the new algorithm can generate better solutions than general local search algorithms.
基金Project supported by the National Basic Research Program of China(Grant No.2013CB338002)the National Natural Science Foundation of China(Grant Nos.11504430 and 61502526)
文摘We apply the transitionless driving on the local adiabatic quantum search algorithm to speed up the adiabatic process. By studying quantum dynamics of the adiabatic search algorithm with the equivalent two-level system, we derive the transi- tionless driving Hamiltonian for the local adiabatic quantum search algorithm. We found that when adding a transitionless quantum driving term Ht~ （t） on the local adiabatic quantum search algorithm, the success rate is 1 exactly with arbitrary evolution time by solving the time-dependent Schr6dinger equation in eigen-picture. Moreover, we show the reason for the drastic decrease of the evolution time is that the driving Hamiltonian increases the lowest eigenvalues to a maximum of
基金the National Key R&D Plan of China(No.2021YFE0105000)the National Natural Science Foundation of China(No.52074213)+1 种基金the Shaanxi Key R&D Plan Project(No.2021SF-472)the Yulin Science and Technology Plan Project(No.CXY-2020-036).
文摘Safety patrol inspection in chemical industrial parks is a complex multi-objective task with multiple degrees of freedom.Traditional pointer instruments with advantages like high reliability and strong adaptability to harsh environment,are widely applied in such parks.However,they rely on manual readings which have problems like heavy patrol workload,high labor cost,high false positives/negatives and poor timeliness.To address the above problems,this study proposes a path planning method for robot patrol in chemical industrial parks,where a path optimization model based on improved iterated local search and random variable neighborhood descent(ILS-RVND)algorithm is established by integrating the actual requirements of patrol tasks in chemical industrial parks.Further,the effectiveness of the model and algorithm is verified by taking real park data as an example.The results show that compared with GA and ILS-RVND,the improved algorithm reduces quantification cost by about 24%and saves patrol time by about 36%.Apart from shortening the patrol time of robots,optimizing their patrol path and reducing their maintenance loss,the proposed algorithm also avoids the untimely patrol of robots and enhances the safety factor of equipment.
基金supported by Yunnan Power Grid Co.,Ltd.Science and Technology Project:Research and application of key technologies for graphical-based power grid accident reconstruction and simulation(YNKJXM20240333).
文摘An optimized volt-ampere reactive(VAR)control framework is proposed for transmission-level power systems to simultaneously mitigate voltage deviations and active-power losses through coordinated control of large-scale wind/solar farms with shunt static var generators(SVGs).The model explicitly represents reactive-power regulation characteristics of doubly-fed wind turbines and PV inverters under real-time meteorological conditions,and quantifies SVG high-speed compensation capability,enabling seamless transition from localized VAR management to a globally coordinated strategy.An enhanced adaptive gain-sharing knowledge optimizer(AGSK-SD)integrates simulated annealing and diversity maintenance to autonomously tune voltage-control actions,renewable source reactive-power set-points,and SVG output.The algorithm adaptively modulates knowledge factors and ratios across search phases,performs SA-based fine-grained local exploitation,and periodically re-injects population diversity to prevent premature convergence.Comprehensive tests on IEEE 9-bus and 39-bus systems demonstrate AGSK-SD’s superiority over NSGA-II and MOPSO in hypervolume(HV),inverse generative distance(IGD),and spread metrics while maintaining acceptable computational burden.The method reduces network losses from 2.7191 to 2.15 MW(20.79%reduction)and from 15.1891 to 11.22 MW(26.16%reduction)in the 9-bus and 39-bus systems respectively.Simultaneously,the cumulative voltage-deviation index decreases from 0.0277 to 3.42×10^(−4) p.u.(98.77%reduction)in the 9-bus system,and from 0.0556 to 0.0107 p.u.(80.76%reduction)in the 39-bus system.These improvements demonstrate significant suppression of line losses and voltage fluctuations.Comparative analysis with traditional heuristic optimization algorithms confirms the superior performance of the proposed approach.
基金supported by the National Natural Science Foundation of China(No.12071077).
文摘Recently,learned heuristics have been widely applied to solve combinatorial optimization problems(e.g.,traveling salesman problem(TSP)).However,the scalability of these learning-based methods hinders the applications in practical scenarios.Specifically,models pre-trained on the small-scale data generalize poorly to large-scale problems.Moreover,learning the heuristics directly for large-scale problems costs tremendous time and space.To extend the scalability of learned heuristics on TSP,we propose a Hierarchical neural framework for solving large-scale traveling salesman problems(HiTSPs)based on a divide-and-conquer strategy.In particular,the HiTSP framework first divides the large-scale problem into small-scale subproblems by node clustering.Each subproblem is conquered by a modified pointer network learned from reinforcement learning.The tour of the original TSP is constructed by linking solutions of subproblems and optimized by a novel segmented local search algorithm.Notably,the segmented local search algorithm leverages the node clustering information to prune many unnecessary operations and significantly reduces the complexity in theory.Extensive experiments show that HiTSP outperforms state-of-the-art learning-based methods and Google OR-Tools in large-scale cases.Moreover,compared to the best heuristic algorithms,HiTSP has a significant advantage in efficiency for large-scale TSP problems.
