In this paper, a mathematical model of real-time simulation is given, and the problem of convergence on real-time Runge-Kutta algorithms is analysed. At last a theorem on the relation between the order of compensation...In this paper, a mathematical model of real-time simulation is given, and the problem of convergence on real-time Runge-Kutta algorithms is analysed. At last a theorem on the relation between the order of compensation and the convergent order of real-time algorithm is proved.展开更多
A class of parallel Runge-Kutta Methods for differential-algebraic equations of index 2are constructed for multiprocessor system. This paper gives the order conditions and investigatesthe convergence theory for such m...A class of parallel Runge-Kutta Methods for differential-algebraic equations of index 2are constructed for multiprocessor system. This paper gives the order conditions and investigatesthe convergence theory for such methods.展开更多
A class of parallel implicit Runge-Kutta formulas is constructed for multiprocessor system. A family of parallel implicit two-stage fourth order Runge-Kutta formulas is given. For these formulas, the convergence is pr...A class of parallel implicit Runge-Kutta formulas is constructed for multiprocessor system. A family of parallel implicit two-stage fourth order Runge-Kutta formulas is given. For these formulas, the convergence is proved and the stability analysis is given. The numerical examples demonstrate that these formulas can solve an extensive class of initial value problems for the ordinary differential equations.展开更多
Error analysis methods in frequency domain are developed in this paper for determining the characteristic root and transfer function errors when the linear multipass algorithms are used to solve linear differential eq...Error analysis methods in frequency domain are developed in this paper for determining the characteristic root and transfer function errors when the linear multipass algorithms are used to solve linear differential equations. The relation between the local truncation error in time domain and the error in frequency domain is established, which is the basis for developing the error estimation methods. The error estimation methods for the digital simulation model constructed by using the Runge-Kutta algorithms and the linear multistep predictor-corrector algorithms are also given.展开更多
实车动力电池的健康状态(state of health,SOH)评估存在数据质量差、工况不统一、数据利用率低等问题,本文面向阶梯倍率充电工况构建多源特征提取及SOH估计模型。首先,通过数据清洗、切割、填充,获取独立的充电片段;其次,基于不同电流...实车动力电池的健康状态(state of health,SOH)评估存在数据质量差、工况不统一、数据利用率低等问题,本文面向阶梯倍率充电工况构建多源特征提取及SOH估计模型。首先,通过数据清洗、切割、填充,获取独立的充电片段;其次,基于不同电流阶段计算容量,实现原始数据利用率达96.9%,并与单独限定SOC范围计算容量的方法相比,误差降低48.1%以上;然后,从当前工况、历史累积两个维度提取多个健康因子,对于当前工况特征值,通过灰色关联度及干扰性随机森林重要度分析双重筛选。对于历史累积特征值,利用Spearson相关性分析和核主成分分析方法(kernel principal component analysis,KPCA)降低信息冗余;最后,对门控循环单元网络模型(gated recurrent unit,GRU)引入注意力机制和龙格库塔优化算法(Runge Kutta optimizer,RUN),建立RUN-GRU-attention模型,基于实车运行数据集与现有5种模型进行对比,实验结果表明,无论是包含单阶段还是多阶段电流的测试样本,优化模型的估计精度更佳,误差不高于0.0086,并且随着充电循环次数增加表现出良好的误差收敛性,可有效预测SOH波动趋势。展开更多
文摘In this paper, a mathematical model of real-time simulation is given, and the problem of convergence on real-time Runge-Kutta algorithms is analysed. At last a theorem on the relation between the order of compensation and the convergent order of real-time algorithm is proved.
文摘A class of parallel Runge-Kutta Methods for differential-algebraic equations of index 2are constructed for multiprocessor system. This paper gives the order conditions and investigatesthe convergence theory for such methods.
基金Project supported by the National Natural Science Foundation of China
文摘A class of parallel implicit Runge-Kutta formulas is constructed for multiprocessor system. A family of parallel implicit two-stage fourth order Runge-Kutta formulas is given. For these formulas, the convergence is proved and the stability analysis is given. The numerical examples demonstrate that these formulas can solve an extensive class of initial value problems for the ordinary differential equations.
基金This project was supported by the National Natural Science Foundation of China (No. 19871080).
文摘Error analysis methods in frequency domain are developed in this paper for determining the characteristic root and transfer function errors when the linear multipass algorithms are used to solve linear differential equations. The relation between the local truncation error in time domain and the error in frequency domain is established, which is the basis for developing the error estimation methods. The error estimation methods for the digital simulation model constructed by using the Runge-Kutta algorithms and the linear multistep predictor-corrector algorithms are also given.
文摘实车动力电池的健康状态(state of health,SOH)评估存在数据质量差、工况不统一、数据利用率低等问题,本文面向阶梯倍率充电工况构建多源特征提取及SOH估计模型。首先,通过数据清洗、切割、填充,获取独立的充电片段;其次,基于不同电流阶段计算容量,实现原始数据利用率达96.9%,并与单独限定SOC范围计算容量的方法相比,误差降低48.1%以上;然后,从当前工况、历史累积两个维度提取多个健康因子,对于当前工况特征值,通过灰色关联度及干扰性随机森林重要度分析双重筛选。对于历史累积特征值,利用Spearson相关性分析和核主成分分析方法(kernel principal component analysis,KPCA)降低信息冗余;最后,对门控循环单元网络模型(gated recurrent unit,GRU)引入注意力机制和龙格库塔优化算法(Runge Kutta optimizer,RUN),建立RUN-GRU-attention模型,基于实车运行数据集与现有5种模型进行对比,实验结果表明,无论是包含单阶段还是多阶段电流的测试样本,优化模型的估计精度更佳,误差不高于0.0086,并且随着充电循环次数增加表现出良好的误差收敛性,可有效预测SOH波动趋势。
