The aim of this paper is to present graphically the behaviour of a simulation model to the varying parameters and to establish the suitability of this representation as a valid tool for the analysis of the same parame...The aim of this paper is to present graphically the behaviour of a simulation model to the varying parameters and to establish the suitability of this representation as a valid tool for the analysis of the same parameters. In this paper, we define parameter combinatorial diagram as the joint graphical representation of all box plots related to the adjustment between real and simulated data, by setting and/or changing the parameters of the simulation model. To do this, we start with a box plot representing the values of an objective adjustment function, achieving these results when varying all the parameters of the simulation model, Then we draw the box plot when setting all the parameters of the model, for example, using the median or average. Later, we get all the box plots when carrying out simulations combining fixed or variable values of the model parameters. Finally, all box plots obtained are represented neatly in a single graph. It is intended that the new parameter combinatorial diagram is used to examine and analyze simulation models useful in practice. This paper presents combinatorial diagrams of different examples of application as in the case of hydrologic models of one, two, three, and five parameters.展开更多
With the development of technology,diffusion model-based solvers have shown significant promise in solving Combinatorial Optimization(CO)problems,particularly in tackling Non-deterministic Polynomial-time hard(NP-hard...With the development of technology,diffusion model-based solvers have shown significant promise in solving Combinatorial Optimization(CO)problems,particularly in tackling Non-deterministic Polynomial-time hard(NP-hard)problems such as the Traveling Salesman Problem(TSP).However,existing diffusion model-based solvers typically employ a fixed,uniform noise schedule(e.g.,linear or cosine annealing)across all training instances,failing to fully account for the unique characteristics of each problem instance.To address this challenge,we present GraphGuided Diffusion Solvers(GGDS),an enhanced method for improving graph-based diffusion models.GGDS leverages Graph Neural Networks(GNNs)to capture graph structural information embedded in node coordinates and adjacency matrices,dynamically adjusting the noise levels in the diffusion model.This study investigates the TSP by examining two distinct time-step noise generation strategies:cosine annealing and a Neural Network(NN)-based approach.We evaluate their performance across different problem scales,particularly after integrating graph structural information.Experimental results indicate that GGDS outperforms previous methods with average performance improvements of 18.7%,6.3%,and 88.7%on TSP-500,TSP-100,and TSP-50,respectively.Specifically,GGDS demonstrates superior performance on TSP-500 and TSP-50,while its performance on TSP-100 is either comparable to or slightly better than that of previous methods,depending on the chosen noise schedule and decoding strategy.展开更多
Multiobjective combinatorial optimization(MOCO)problems have a wide range of applications in the real world.Recently,learning-based methods have achieved good results in solving MOCO problems.However,most of these met...Multiobjective combinatorial optimization(MOCO)problems have a wide range of applications in the real world.Recently,learning-based methods have achieved good results in solving MOCO problems.However,most of these methods use attention mechanisms and their variants,which have room for further improvement in the speed of solving MOCO problems.In this paper,following the idea of decomposition strategy and neural combinatorial optimization,a novel fast-solving model for MOCO based on retention is proposed.A brand new calculation of retention is proposed,causal masking and exponential decay are deprecated in retention,so that our model could better solve MOCO problems.During model training,a parallel computation of retention is applied,allowing for fast parallel training.When using the model to solve MOCO problems,a recurrent computation of retention is applied,enabling quicker problem-solving.In order to make our model more practical and flexible,a preference-based retention decoder is proposed,which allows generating approximate Pareto solutions for any trade-off preferences directly.An industry-standard deep reinforcement learning algorithm is used to train RM-MOCO.Experimental results show that,while ensuring the quality of problem solving,the proposed method significantly outperforms some other methods in terms of the speed of solving MOCO problems.展开更多
电动汽车(electric vehicle,EV)充电行为存在强随机性与高波动性,使其充电站短期充电负荷预测精度较低,作为移动电力存储和负载资源参与车到网(vehicle to grid,V2G)服务中,其调度中心需要在短时间内预测EV的充电负荷来改善其对电网负...电动汽车(electric vehicle,EV)充电行为存在强随机性与高波动性,使其充电站短期充电负荷预测精度较低,作为移动电力存储和负载资源参与车到网(vehicle to grid,V2G)服务中,其调度中心需要在短时间内预测EV的充电负荷来改善其对电网负荷的影响。为了提高EV充电站短期充电负荷预测精度,提出一种冠豪猪优化器变分模态分解双向长短期记忆神经网络(crested porcupine optimizer variational mode decomposition bidirectional long short term memory,CPO VMD BiLSTM)组合模型进行EV充电站短期充电负荷预测的方法。首先,考虑影响EV充电负荷的多种因素和历史充电站充电负荷共同构成输入特征矩阵。然后利用CPO算法对VMD其核心参数进行优化搜索,实现参数自适应优化设置。之后采用CPO VMD对历史充电负荷数据进行分解,弱化负荷的非平稳性,捕捉其局部特征。最后在BiLSTM模型中输入分解后的特征矩阵来实现充电站短期充电负荷的预测目标。以美国ANN DATA公开数据集中位于加州理工大学校园内EV充电站的历史充电负荷数据作为实际算例,与独立模型、未优化组合模型、优化组合模型进行对比,均方根误差(root mean squared error,RMSE)和平均绝对误差(mean absolute error,MAE)平均降低了41.23%和59.04%。因此,验证了提出方法在充电站充电负荷短期预测中精度的提高和实用性。展开更多
文摘The aim of this paper is to present graphically the behaviour of a simulation model to the varying parameters and to establish the suitability of this representation as a valid tool for the analysis of the same parameters. In this paper, we define parameter combinatorial diagram as the joint graphical representation of all box plots related to the adjustment between real and simulated data, by setting and/or changing the parameters of the simulation model. To do this, we start with a box plot representing the values of an objective adjustment function, achieving these results when varying all the parameters of the simulation model, Then we draw the box plot when setting all the parameters of the model, for example, using the median or average. Later, we get all the box plots when carrying out simulations combining fixed or variable values of the model parameters. Finally, all box plots obtained are represented neatly in a single graph. It is intended that the new parameter combinatorial diagram is used to examine and analyze simulation models useful in practice. This paper presents combinatorial diagrams of different examples of application as in the case of hydrologic models of one, two, three, and five parameters.
基金supported by the National Science and Technology Council,Taiwan,under grant no.NSTC 114-2221-E-197-005-MY3.
文摘With the development of technology,diffusion model-based solvers have shown significant promise in solving Combinatorial Optimization(CO)problems,particularly in tackling Non-deterministic Polynomial-time hard(NP-hard)problems such as the Traveling Salesman Problem(TSP).However,existing diffusion model-based solvers typically employ a fixed,uniform noise schedule(e.g.,linear or cosine annealing)across all training instances,failing to fully account for the unique characteristics of each problem instance.To address this challenge,we present GraphGuided Diffusion Solvers(GGDS),an enhanced method for improving graph-based diffusion models.GGDS leverages Graph Neural Networks(GNNs)to capture graph structural information embedded in node coordinates and adjacency matrices,dynamically adjusting the noise levels in the diffusion model.This study investigates the TSP by examining two distinct time-step noise generation strategies:cosine annealing and a Neural Network(NN)-based approach.We evaluate their performance across different problem scales,particularly after integrating graph structural information.Experimental results indicate that GGDS outperforms previous methods with average performance improvements of 18.7%,6.3%,and 88.7%on TSP-500,TSP-100,and TSP-50,respectively.Specifically,GGDS demonstrates superior performance on TSP-500 and TSP-50,while its performance on TSP-100 is either comparable to or slightly better than that of previous methods,depending on the chosen noise schedule and decoding strategy.
基金supported by the National Natural Science Foundation of China(No.62102002).
文摘Multiobjective combinatorial optimization(MOCO)problems have a wide range of applications in the real world.Recently,learning-based methods have achieved good results in solving MOCO problems.However,most of these methods use attention mechanisms and their variants,which have room for further improvement in the speed of solving MOCO problems.In this paper,following the idea of decomposition strategy and neural combinatorial optimization,a novel fast-solving model for MOCO based on retention is proposed.A brand new calculation of retention is proposed,causal masking and exponential decay are deprecated in retention,so that our model could better solve MOCO problems.During model training,a parallel computation of retention is applied,allowing for fast parallel training.When using the model to solve MOCO problems,a recurrent computation of retention is applied,enabling quicker problem-solving.In order to make our model more practical and flexible,a preference-based retention decoder is proposed,which allows generating approximate Pareto solutions for any trade-off preferences directly.An industry-standard deep reinforcement learning algorithm is used to train RM-MOCO.Experimental results show that,while ensuring the quality of problem solving,the proposed method significantly outperforms some other methods in terms of the speed of solving MOCO problems.
文摘电动汽车(electric vehicle,EV)充电行为存在强随机性与高波动性,使其充电站短期充电负荷预测精度较低,作为移动电力存储和负载资源参与车到网(vehicle to grid,V2G)服务中,其调度中心需要在短时间内预测EV的充电负荷来改善其对电网负荷的影响。为了提高EV充电站短期充电负荷预测精度,提出一种冠豪猪优化器变分模态分解双向长短期记忆神经网络(crested porcupine optimizer variational mode decomposition bidirectional long short term memory,CPO VMD BiLSTM)组合模型进行EV充电站短期充电负荷预测的方法。首先,考虑影响EV充电负荷的多种因素和历史充电站充电负荷共同构成输入特征矩阵。然后利用CPO算法对VMD其核心参数进行优化搜索,实现参数自适应优化设置。之后采用CPO VMD对历史充电负荷数据进行分解,弱化负荷的非平稳性,捕捉其局部特征。最后在BiLSTM模型中输入分解后的特征矩阵来实现充电站短期充电负荷的预测目标。以美国ANN DATA公开数据集中位于加州理工大学校园内EV充电站的历史充电负荷数据作为实际算例,与独立模型、未优化组合模型、优化组合模型进行对比,均方根误差(root mean squared error,RMSE)和平均绝对误差(mean absolute error,MAE)平均降低了41.23%和59.04%。因此,验证了提出方法在充电站充电负荷短期预测中精度的提高和实用性。
