%0 Journal Article
%T 基于改进粒子群算法优化回声状态网络的时间序列预测<br>Time Series Forecasting Based on Echo State Network Optimized by Improved Particle Swarm Algorithm
%A 张亚楠
%A 赵丽娜
%J Computer Science and Application
%P 2070-2079
%@ 2161-881X
%D 2021
%I Hans Publishing
%R 10.12677/CSA.2021.118212
%X <div style=\"text-align:justify;\">
	<span style=\"font-size:14px;\">由于结构简单、收敛速度快等优点，回声状态网络(Echo State Network, ESN)已被广泛的用于时间序列的预测。针对回声状态网络中随机生成权值矩阵带来的不适用于特定时间序列的问题，本文提出利用改进的粒子群优化算法来优化回声状态网络部分随机权值。与标准粒子群优化算法相比，惯性权重和学习因子自适应调整，以提高算法的寻优性能。为验证本文方法的有效性，对Mackey-Glass时间序列、非线性自回归滑动模型(Nonlinear Auto Regressive Moving Average, NARMA)和Lorenz时间序列进行仿真实验。实验结果表明，本文提出的方法可以进一步提升回声状态网络对时间序列的预测精度。<br />
Echo State Network (ESN) has been widely used in time series prediction due to its simple structure and fast convergence speed. In order to solve the problem that random weight matrix generated in ESN is not applicable to specific time series, an improved particle swarm optimization (PSO) algorithm is proposed to optimize partial random weights in ESN. Compared with the standard particle swarm optimization algorithm, the inertia weight and learning factor are adjusted to improve the optimization performance of the algorithm. To verify the effectiveness of the proposed method, the Mackey-Glass time series, the nonlinear autoregressive sliding model (NARMA) and the Lorenz time series are simulated. Experimental results show that the method proposed in this paper can further improve the prediction accuracy of ESN for time series.</span> 
</div>
%K 时间序列预测，回声状态网络，粒子群优化算法<br>Time Series Prediction
%K ESN
%K Particle Swarm Optimization Algorithm
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=44404