GIHS-TSFNN并行学习算法及其应用研究
|
摘要
针对传统T-S模糊神经网络的随机初始网络参数导致网络学习速度慢、易陷入局部解以及运算精度低等缺陷,提出了一种应用佳点集的改进和声搜索算法(GIHS)优化T-S模糊神经网络的并行学习算法.首先应用佳点集择优构造更加高质量的初始和声库,然后搜索过程中进行参数动态调整,并且每次迭代产生多个新解,充分利用和声记忆库的信息,以提高算法的全局搜索能力和收敛速度.其次,将GIHS算法与T-S神经网络相结合构建并行学习算法,实现两种算法的并行交互集成,得到了最优参数配置以提高T-S模糊神经网络的泛化能力.最后将该算法应用到农业干旱等级预测中以解决旱情评估问题.仿真实验表明,GIHS算法性能优于基本HS和IHS算法,且与T-S模糊神经网络、HS算法优化的T-S模糊神经网络和IHS算法优化的T-S模糊神经网络相比,具有更高的预测准确度.
To improve the prediction accuracy of agricultural drought grade, an improved harmony search algorithm based on good point set(GIHS) is proposed to optimize the T-S neural network model for agricultural drought. Firstly, the good-point set is used to construct a more high quality initial harmony library. Secondly, it adjusts the parameters dynamically during the search process and generates several solutions in each iteration so as to make full use of information of harmony memory to improve the global search ability and convergence speed of algorithm. GIHS is used to train the T-S neural network to get the optimal parameter configuration, thus can improve the generalization ability of T-S fuzzy neural network. The efficiency of the proposed prediction algorithm is tested by the simulation of the prediction of agricultural drought grade. The simulation results show that the performance of GIHS algorithm is better than the basic HS and IHS, and it has higher forecasting accuracy compared with the traditional T-S fuzzy neural network and T-S fuzzy neural network optimized by HS algorithm and T-S neural network optimized by IHS algorithm, so it is feasible and effective in prediction of agricultural drought grade.
To improve the prediction accuracy of agricultural drought grade, an improved harmony search algorithm based on good point set(GIHS) is proposed to optimize the T-S neural network model for agricultural drought. Firstly, the good-point set is used to construct a more high quality initial harmony library. Secondly, it adjusts the parameters dynamically during the search process and generates several solutions in each iteration so as to make full use of information of harmony memory to improve the global search ability and convergence speed of algorithm. GIHS is used to train the T-S neural network to get the optimal parameter configuration, thus can improve the generalization ability of T-S fuzzy neural network. The efficiency of the proposed prediction algorithm is tested by the simulation of the prediction of agricultural drought grade. The simulation results show that the performance of GIHS algorithm is better than the basic HS and IHS, and it has higher forecasting accuracy compared with the traditional T-S fuzzy neural network and T-S fuzzy neural network optimized by HS algorithm and T-S neural network optimized by IHS algorithm, so it is feasible and effective in prediction of agricultural drought grade.
引文
[1] Geem Z W,Kim J H, Loganathan G V. A new heuristic optimization algorithm:Harmony search[J].Simulation, 2001, 76(2):60-68.
[2] Geem Z W.Optimal cost design of water distribution networks using harmony search[J]. Eng Optimiz, 2006, 38(3):259-280.
[3] Kang S L, Geem Z W. A new structural optimization method based on the harmony search algorithm[J].Comput Struct, 2004, 82(9/10):781-798.
[4] Kang S L, Geem Z W. A new meta-heuristic algorithm for continuous engineering optimization:harmony search theory and practice[J]. Computer Methods in Applied Mechanics and Engineering,2005, 194:3902-3933.
[5] Geem Z W, Tseng C, Park Y. Harmony search for generalized orienteering problem:best touring in China[J]. Springer Lecture Notes in Computer Science, 2005, 3412:741-750.
[6]周雅兰,黄韬和.声搜索算法改进与应用[J].计算机科学,2014, 41(s1):52-56.
[7] Mahdav M, Fesangharym, Damangir E. An improved harmony search algorithm for solving optimization problems[J]. Applied Mathematics and Computation, 2007, 188(2):1567-1579.
[8] Omran, Mahdavi M. Global-best harmony search[J]. Applied Mathematics and Computation, 2008,198(2):643-656.
[9] Zhao S Z, Suganthan P N, Pan Q K, et al. Dynamic multi-swarm particle swarm optimizer with harmony search[J]. Expert Systems with Applications, 2011, 38(4):37353-742.
[10] Aghakouchak A. A multivariate approach for persistence-based drought prediction:Application to the 2010-2011 East Africa drought[J]. Journal of Hydrology, 2015, 526:127-135.
[11] Gong Z, Forrest J Y L. Special issue on meteorological disaster risk analysis and assessment:on basis of grey systems theory[J]. Natural Hazards, 2014, 71(2):995-1000.
[12] Cheng J, Tao J P, Cheng J, et al. Fuzzy comprehensive evaluation of drought vulnerability based on the analytic hierarchy process:an empirical study from Xiaogan city in Hubei province[J].Hydrometallurgy, 2010, 1(4):126-135.
[13] Sun Z, Zhang J, Zhang Q, et al. Integrated risk zoning of drought and water logging disasters based on fuzzy comprehensive evaluation in Anhui Province, China[J]. Natural hazards, 2014, 71(3):1639-1657.
[14]苏超,方崇,黄伟军等.SOM神经网络在农业旱情评价中的应用[J].人民黄河,2011,33(7):93-95.
[15]陈兴,孟卫东,严太华.基于T-S模型的模糊神经网络在股市预测中的应用[J].系统工程理论与实践,2001, 21(2):66-72.
[16]王培崇,李丽荣,高文超等.应用佳点集的混合反向学习人工鱼群算法[J].计算机应用研究,2015, 32(7):1992-1995.
[17]刘香品,宣士斌,刘峰.引入佳点集和猴群翻过程的人工蜂群算法[J].模式识别与人工智能,2015, 28(1):80-89.
[18] Hu C H, Wang J X, Wang Y X, et al. Review on research of hydrological drought index[J]. Yangtze River, 2013, 44(7):11-15.
[19]李柏贞,周广胜.干旱指标研究进展[J].生态学报2014, 34(5):1043-1052.
[2] Geem Z W.Optimal cost design of water distribution networks using harmony search[J]. Eng Optimiz, 2006, 38(3):259-280.
[3] Kang S L, Geem Z W. A new structural optimization method based on the harmony search algorithm[J].Comput Struct, 2004, 82(9/10):781-798.
[4] Kang S L, Geem Z W. A new meta-heuristic algorithm for continuous engineering optimization:harmony search theory and practice[J]. Computer Methods in Applied Mechanics and Engineering,2005, 194:3902-3933.
[5] Geem Z W, Tseng C, Park Y. Harmony search for generalized orienteering problem:best touring in China[J]. Springer Lecture Notes in Computer Science, 2005, 3412:741-750.
[6]周雅兰,黄韬和.声搜索算法改进与应用[J].计算机科学,2014, 41(s1):52-56.
[7] Mahdav M, Fesangharym, Damangir E. An improved harmony search algorithm for solving optimization problems[J]. Applied Mathematics and Computation, 2007, 188(2):1567-1579.
[8] Omran, Mahdavi M. Global-best harmony search[J]. Applied Mathematics and Computation, 2008,198(2):643-656.
[9] Zhao S Z, Suganthan P N, Pan Q K, et al. Dynamic multi-swarm particle swarm optimizer with harmony search[J]. Expert Systems with Applications, 2011, 38(4):37353-742.
[10] Aghakouchak A. A multivariate approach for persistence-based drought prediction:Application to the 2010-2011 East Africa drought[J]. Journal of Hydrology, 2015, 526:127-135.
[11] Gong Z, Forrest J Y L. Special issue on meteorological disaster risk analysis and assessment:on basis of grey systems theory[J]. Natural Hazards, 2014, 71(2):995-1000.
[12] Cheng J, Tao J P, Cheng J, et al. Fuzzy comprehensive evaluation of drought vulnerability based on the analytic hierarchy process:an empirical study from Xiaogan city in Hubei province[J].Hydrometallurgy, 2010, 1(4):126-135.
[13] Sun Z, Zhang J, Zhang Q, et al. Integrated risk zoning of drought and water logging disasters based on fuzzy comprehensive evaluation in Anhui Province, China[J]. Natural hazards, 2014, 71(3):1639-1657.
[14]苏超,方崇,黄伟军等.SOM神经网络在农业旱情评价中的应用[J].人民黄河,2011,33(7):93-95.
[15]陈兴,孟卫东,严太华.基于T-S模型的模糊神经网络在股市预测中的应用[J].系统工程理论与实践,2001, 21(2):66-72.
[16]王培崇,李丽荣,高文超等.应用佳点集的混合反向学习人工鱼群算法[J].计算机应用研究,2015, 32(7):1992-1995.
[17]刘香品,宣士斌,刘峰.引入佳点集和猴群翻过程的人工蜂群算法[J].模式识别与人工智能,2015, 28(1):80-89.
[18] Hu C H, Wang J X, Wang Y X, et al. Review on research of hydrological drought index[J]. Yangtze River, 2013, 44(7):11-15.
[19]李柏贞,周广胜.干旱指标研究进展[J].生态学报2014, 34(5):1043-1052.