Time Series Prediction with an Improved Echo State Network Using Small World Network
-
摘要: 为了提高时间序列的预测精度, 提出了利用改进的小世界网络优化泄露积分型回声状态网(Leaky-integrator echo state network, Leaky ESN)的时间序列预测方法. 首先提出一个改进型小世界网络, 其加边概率是节点间距离的负指数函数. 然后, 利用加边概率直接表示Leaky ESN储备池两个神经节点的连接权值, 取值范围为[0,1], 表征了节点间的连接程度. 利用这个新型小世界网络改进Leaky ESN的储备池神经节点的连接方式, 有目的地实现了稀疏连接, 减小了Leaky ESN储备池随机稀疏连接的盲目性, 提高了储备池的适应性.最后, 利用改进的Leaky ESN预测典型的非线性时间序列, 并利用Matlab仿真软件验证了本文提出方法的有效性. 与Leaky ESN相比, 本文提出的方法具有更高的预测精度和更短的训练时间.Abstract: In order to improve the prediction accuracy of echo state network for time series, this paper uses a modified small world network to improve leaky-integrator echo state network (Leaky ESN) and then uses the improved Leaky ESN to predict time series. Firstly, we propose a modified small world network whose adding edge probability is a negative exponential function of the distance between the nodes. Then, we use adding edge probability to directly represent the connection weights of two neural nodes in the reservoir of Leaky ESN. The value of the connection weights belongs to [0,1] and represents the degree of connection between the two nodes. The modified small world network is utilized to improve the connection of reservoir neural nodes, which purposefully achieves the sparse connection to reduce the blindness of the original sparse connections and then improves the adaptive capacity of the reservoir. Finally, we use the modified Leaky ESN to predict a typical nonlinear time series. We use Matlab to verify the effectiveness of the proposed method. Compared with Leaky ESN, the method proposed in this paper has a higher prediction precision and a shorter training time.
-
Key words:
- Echo state network (ESN) /
- small world network /
- time series prediction /
- reservoir
-
[1] Zhou Z J, Hu C H. An effective hybrid approach based on Grey and ARMA for forecasting gyro drift. Chaos, Solitons and Fractals, 2008, 35(3): 525-529 [2] Rao Yun-Zhang, Xu Shui-Tai, Xiong Ling-Yan. Time series prediction of heavy metal pollution in mining areas based on ARIMA model. Metal Mine, 2010, (6): 142-146(饶运章, 徐水太, 熊灵燕. 基于ARIMA模型的矿区重金属污染时间序列预测. 金属矿山, 2010, (6): 142-146) [3] Huo Xiao-Yu, Yang Shi-Jiao, Wu Chang-Zhen. The research of prediction methods and application of chaotic time series based on BPNN. Journal of University of South China (Science and Technology), 2012, 26(2): 26-31(霍晓宇, 杨仕教, 吴长振. 基于BP神经网络的混沌时间序列预测方法及应用研究. 南华大学学报(自然科学版), 2012, 26(2): 26-31) [4] Song R Z, Xiao W D, Sun C Y. Optimal tracking control for a class of unknown discrete-time systems with actuator saturation via data-based ADP algorithm. Acta Automatica Sinica, 2013, 39(9): 1413-1420 [5] Qu Ren-Hui, Song Li-Hua, Di Chao-Sheng. Chaotic time series prediction based on recursive networks. Journal of Jilin University (Information Science Edition), 2008, 26(2): 136-140(曲仁慧, 宋丽华, 邸朝生. 基于递归网络的混沌时间序列预测. 吉林大学学报(信息科学版), 2008, 26(2): 136-140) [6] Han Min, Xu Mei-Ling, Ren Wei-Jie. Research on multivariate chaotic time series prediction using mRSM model. Acta Automatica Sinica, 2014, 40(5): 822-829(韩敏, 许美玲, 任伟杰. 多元混沌时间序列的相关状态机预测模型研究. 自动化学报, 2014, 40(5): 822-829) [7] Hu Shou-Song, Zhang Zheng-Dao. Fault prediction for nonlinear time series based on neural network. Acta Automatica Sinica, 2007, 33(7): 744-748(胡寿松, 张正道. 基于神经网络的非线性时间序列故障预报. 自动化学报, 2007, 33(7): 744-748) [8] Peng Yu, Wang Jian-Min, Peng Xi-Yuan. Researches on time series prediction with echo state networks. Acta Electronica Sinica , 2010, 38(2A): 148-154(彭宇, 王建民, 彭喜元. 基于回声状态网络的时间序列预测方法研究. 电子学报, 2010, 38(2A): 148-154) [9] Qiao Jun-Fei, Bo Ying-Chun, Han Guang. Application of ESN-based multi indices dual heuristic dynamic programming on wastewater treatment process. Acta Automatica Sinica, 2013, 39(7): 1146-1151(乔俊飞, 薄迎春, 韩广. 基于ESN的多指标DHP控制策略在污水处理过程中的应用. 自动化学报, 2013, 39(7): 1146-1151) [10] Zhao Lu-Sha. Research on Nonlinear Time Series Prediction based on Echo State Networks [Master dissertation], Harbin Institute of Technology, China, 2012.(赵露莎. 基于回声状态网络的非线性时间序列预测方法研究 [硕士学位论文], 哈尔滨工业大学, 中国, 2012.) [11] Wang Zhuo-Qun, Sun Zhi-Guo. Method for prediction of multi-scale time series with WDESN. Journal of Electronic Measurement and Instrument, 2010, 24(10): 947-951(王卓群, 孙志国. 一种小波分解回声状态网络时间序列预测方法. 电子测量与仪器学报, 2010, 24(10): 947-951) [12] Bohland J W, Minai A A. Efficient associative memory using small-world architecture. Neurocomputing, 2001, 38-40: 489-496 [13] Jaeger H, Lukovsevivcius M, Popovici D, Siewert U. Optimization and applications of echo state networks with leaky-integrator neurons. Neural Networks, 2007, 20(3): 335-352 [14] Li Han. Nonlinear Time Series Prediction Based on Echo State Networks [Master dissertation], Dalian University of Technology, China, 2013.(李菡. 基于回声状态网络的非线性时间序列预测研究 [硕士学位论文], 大连理工大学, 中国, 2013.) [15] Liebald B. Exploration of Effects of Different Network Topologies on the ESN Signal Crosscorrelation Matrix Spectrum [Master dissertation], International University Bremen, Germany, 2004. [16] Wang Juan. Research on Topologies of Echo State Network [Master dissertation], Chongqing University, China, 2013.(王娟. 回声状态网络的拓扑结构研究 [硕士学位论文], 重庆大学, 中国, 2013.) [17] Chen Feng-Lan. The Network Security Situation Predicting Technology Based on the Small-World Echo State Network [Master dissertation], Lanzhou University, China, 2014.(陈凤兰. 基于小世界回声状态网络的网络安全态势预测技术研究 [硕士学位论文], 兰州大学, 中国, 2014.) [18] Lv J H, Chen G R. A time-varying complex dynamical network model and its controlled synchronization criteria. IEEE Transactions on Automatic Control, 2005, 50(6): 841-846 [19] Lv Jin-Hu. Mathematical models and synchronization criterions of complex dynamical networks. Systems Engineering Theory and Practice, 2004, 24(4): 17-22, 62(吕金虎. 复杂动力网络的数学模型与同步准则. 系统工程理论与实践, 2004, 24(4): 17-22, 62) [20] Chen Guan-Rong. Problems and challenges in control theory under complex dynamical network environments. Acta Automatica Sinica, 2013, 39(4): 312-321(陈关荣. 复杂动态网络环境下控制理论遇到的问题与挑战. 自动化学报, 2013, 39(4): 312-321) [21] Chen L, Lv J H, Lu J A. Synchronization of the time-varying discrete biological networks. In: Proceedings of the 2007 IEEE International Symposium on Circuit and Systems (ISCAS 07). New Orleans, LA: IEEE, 2007. 2650-2653 [22] Li Jun, Yue Wen-Qi. Dynamic soft sensor modeling and its application using leaky-integrator ESN. CIESC Journal, 2014, 65(10): 4004-4014(李军, 岳文琦. 基于泄漏积分型回声状态网络的软测量动态建模方法及应用. 化工学报, 2014, 65(10): 4004-4014) [23] Lun S X, Wang S, Guo T T, Du C J. An I-V model based on time warp invariant echo state network for photovoltaic array with shaded solar cells. Solar Energy, 2014, 105: 529-541 [24] Lun S X, Yao X S, Qi H Y, Hu H F. A novel model of leaky integrator echo state network for time-series prediction. Neurocomputing, 2015, 159: 58-66 [25] Zhou J, Lu J A, Lv J H. Adaptive synchronization of an uncertain complex dynamical network. IEEE Transactions on Automatic Control, 2006, 51(4): 652-656 [26] Lv J H, Yu X H, Chen G R, Cheng D Z. Characterizing the synchronizability of small-world dynamical networks. IEEE Transactions on Circuits and Systems I: Regular Papers, 2004, 51(4): 787-796
点击查看大图
计量
- 文章访问数: 1677
- HTML全文浏览量: 69
- PDF下载量: 1918
- 被引次数: 0