价值工程
價值工程
개치공정
VALUE ENGINEERING
2014年
8期
6-7,8
,共3页
自适应惯性权重粒子群优化算法%小波神经网络%上证指数预测
自適應慣性權重粒子群優化算法%小波神經網絡%上證指數預測
자괄응관성권중입자군우화산법%소파신경망락%상증지수예측
Adaptive Inertia Weight Particle Swarm Optimization%Wavelet Neural Network%Shanghai Stock Index Prediction
针对小波神经网络(Wavelet Neural Network, WNN)的学习算法的不足，采用一种自适应惯性权重粒子群优化算法(Adaptive Inertia Weight Particle Swarm Optimization，AIW-PSO)作为小波神经网络的学习算法，建立AIW-PSO小波神经网络模型对上证指数进行预测，并将预测结果传统小波神经网络模型比较。结果表明，AIW-PSO小波神经网络模型对上证指数具有更好的预测效果。
針對小波神經網絡(Wavelet Neural Network, WNN)的學習算法的不足，採用一種自適應慣性權重粒子群優化算法(Adaptive Inertia Weight Particle Swarm Optimization，AIW-PSO)作為小波神經網絡的學習算法，建立AIW-PSO小波神經網絡模型對上證指數進行預測，併將預測結果傳統小波神經網絡模型比較。結果錶明，AIW-PSO小波神經網絡模型對上證指數具有更好的預測效果。
침대소파신경망락(Wavelet Neural Network, WNN)적학습산법적불족，채용일충자괄응관성권중입자군우화산법(Adaptive Inertia Weight Particle Swarm Optimization，AIW-PSO)작위소파신경망락적학습산법，건립AIW-PSO소파신경망락모형대상증지수진행예측，병장예측결과전통소파신경망락모형비교。결과표명，AIW-PSO소파신경망락모형대상증지수구유경호적예측효과。
In the view of the shortage of the Wavelet Neural Network Algorithm, adapt Adaptive Inertia Weight Particle Swarm Optimization Algorithm (AIW-PSO) as a study algorithm, build the AIW-PSO Wavelet Neural Network Model to predict the Shanghai stock Index., and make a comparison between the results of improved algorithm prediction model with results of traditional Wavelet Neural Network Model. The results show that the AIW-PSO Wavelet Neural Network Prediction Model has better prediction results on the Shanghai Stock Index.
