基于肌电信号和极限学习机的下肢关节运动预测

为判断下肢障碍患者的运动意图,通过外骨骼进行康复训练,分析表面肌电信号与下肢关节运动的关系。 提取表面肌电信号的均方根(root mean square,RMS)、绝对值均值(mean absolute value,MAV)、波形长度(waveform length,WL)和方差(variance,VAR)作为特征输入信号,采用极限学习机(extreme learning machine,ELM)建立表面 肌电信号与下肢关节角度之间的映射关系;对输出结果进行优化滤波以降低模型的误差,实现对下肢膝关节角度连 续变化的预测。与传统的反向传播(back propagation,BP)神经网络、径向基神经网络预测结果进行对比,结果证明： 极限学习机在通过表面肌电信号预测下肢关节角度变化中有更高的精度。     

Equilibrium on line method (ELM) for imposition of Neumann boundary conditions in the finite point method (FPM)

Equilibrium on line method (ELM) for imposition of Neumann boundary conditions in the finite point method (FPM) is presented. In contrary to weak‐form‐based methods, strong‐form‐based methods such as the FPM are often unstable and less accurate, especially for problems governed by partial differential equations with Neumann (derivative) boundary conditions. In this paper, a truly meshless approach for imposition of Neumann boundary conditions in the FPM is proposed and adopted for 2D elasticity analyses. In the proposed method, equilibrium on lines on the Neumann boundary conditions is satisfied as Neumann boundary condition equations. Numerical studies show that this method for imposition of Neumann boundary is simple to implement and computationally efficient and also leads to more stable and accurate results. Copyright © 2006 John Wiley & Sons, Ltd.     

计及模型误差的分布式光伏配电网优化调控方法

当前配电网存在信息采集不全、在线获取电网精确模型困难的问题,导致对分布式光伏的调控存在误差,难以满足配电网安全运行的要求,因此提出了一种计及模型误差的分布式光伏配电网调控方法。基于近似灵敏度建立了光伏调控量粗略计算模型;采用极限学习机(ELM)方法建立人工智能辅助决策模型,作为光伏调控量粗略计算模型的修正;进一步地,基于上述2个模型,设计了计及模型误差的分布式光伏优化调控策略;最后进行仿真分析,结果表明提出的调控方法弥补了仅依赖电网模型进行优化带来的误差,提高了优化调控的精度。     

基于改进ELM的风/光/柴/储独立微网分布式电源多目标容量优化配置

针对风/光/柴/储独立混合微电网中分布式电源容量配置问题,以系统供电可靠性作为约束条件,构建考虑微电网经济性、环保性和能源利用效率的多目标优化模型。引入熵理论确定多目标函数各项指标的权重系数,有效地摆脱权重取值的主观性。采用自适应迭代步长、自适应变异策略和搜索空间缩减等措施来改善仿电磁学算法(Electromagnetism-like mechanism,ELM)的收敛性。基于改进的循环充放(ICC)控制策略,用改进的仿电磁学算法求解配置模型。以广西涠洲岛为背景进行算例分析,验证模型和算法的有效性。结果表明改进的ELM算法能有效地解决微电网规划这类多约束、计算量大的组合优化问题,并且在全局最优搜索能力和寻优速度方面具有明显优势。     

Critical Clearing Time prediction within various loads for transient stability assessment by means of the Extreme Learning Machine method

The Critical Clearing Time (CCT) is a key issue for Transient Stability Assessment (TSA) in electrical power system operation, security, and maintenance. However, there are some difficulties in obtaining the CCT, which include the accuracy, fast computation, and robustness for TSA online. Therefore, obtaining the CCT is still an interesting topic for investigation. This paper proposes a new technique for obtaining CCT based on numerical calculations and artificial intelligence techniques. First, the CCT is calculated by the critical trajectory method based on critical generation. Second, the CCT is learned by Extreme Learning Machine (ELM). This proposed method has the ability to obtain the CCT with load changes, different fault occurrences, accuracy, and fast computation, and considering the controller. This proposed method is tested by the IEEE 3-machine 9-bus system and Java-Bali 500 kV 54-machine 25-bus system. The proposed method can provide accurate CCTs with an average error of 0.33% for the Neural Network (NN) method and an average error of 0.06% for the ELM method. The simulation result also shows that this method is a robust algorithm that can address several load changes and different locations of faults occurring. There are 29 load changes used to obtain the CCT, with 20 load changes included for the training process and 9 load changes not included.     

基于OS-ELM和Bootstrap方法的超短期风电功率预测

提出了一种基于在线序贯极限学习机(OS-ELM)的超短期风电功率预测方法。利用OSELM学习速度快、泛化能力强的优点,将批处理和逐次迭代相结合,不断更新训练数据和网络结构,实现了对数值天气预报风速的快速实时修正和风电机组输出功率的快速预测。随后,采用计算机自助(Bootstrap)法构造伪样本,给出了预测功率的置信区间评估。实例和研究结果表明,该预测方法与反向传播(BP)网络、支持向量机(SVM)方法相比,在计算时间上更能满足在线应用需求,而且预测精度相当,有较好的应用前景。     

基于极端学习机的SLFN谐波检测法

为了解决传统电网谐波检测法的实时性和精确度问题,提出一种基于极端学习机（ELM）单隐层前馈神经网络（SLFN）的谐波检测法。该方法把ELM用于SLFN,将傅里叶变换系数选作ELM模型的检测参数,用广义逆矩阵计算输出层权值的最小二乘解,以获得该模型的最优输出向量,从而改善各次谐波电流的检测性能。运用MATLAB仿真软件进行了仿真实验。结果表明：与基于BP算法的SLFN谐波检测法相比,所提出ELM算法的SLFN谐波检测法达到稳定检测的时间缩短,且检测误差的精度有所提高,达到了有效提升谐波检测实时性和精确度的目的。     

风暴灾害下电力断线倒塔概率预测

提出了基于极端学习机(Extreme Learning Machine,ELM)和广义极值(Generalized Extreme Value,GEV)分布的风暴灾害下输电线路断线概率预测模型。该模型首先针对极端风速,通过ELM网络的训练学习,预测出实时变化的风速;随后从概率的角度考虑实时变化的电力线风荷载极值,提出电力线风荷载的实时广义极值分布;从而实现风暴灾害下输电线路的实时断线概率预测。结合输电线路的历史数据展开算例分析,验证了该预测方法的有效性和准确性。     

极限学习机延拓的BS-EMD端点效应抑制算法及应用

针对希尔伯特-黄变换过程中经验模态分解出现的端点效应问题,采用极限学习机（Extreme Learning Machine,ELM）算法对原始数据序列分别向左右两端延拓,对扩展后的数据序列用B样条插值函数求其平均曲线,在此基础上进行下一步分解,结束分解后摒弃两端延展的数据,使算法得到优化,起到了抑制端点效应的作用。通过与未经延拓,BP神经网络延拓和支持向量机延拓各项指标的对比分析表明,该算法不仅有效抑制了经验模态分解过程中的端点效应,在预测速度和分解精度上都有一定的优势。将该方法应用于电力系统的谐波分析中,仿真结果表明该方法能有效抑制EMD的端点效应,更好地分解出谐波中含有的不同频率谐波分量。