基于海洋捕食者算法优化的长短期记忆神经网络径流预测

为提高径流预测精度,研究提出海洋捕食者算法(MPA)与长短期记忆(LSTM)神经网络相结合的径流预测方法.通过6个仿真函数对MPA、粒子群优化(PSO)算法进行测试,利用MPA优化LSTM隐藏层神经元数、训练次数等关键参数,基于主成分分析(PCA)降维和不降维处理分别建立PCA-MPA-LSTM、MPA-LSTM径流预...

Long-term and Short-term Memory Neural Network Runoff Prediction Based on Optimization of Marine Predators Algorithm

To improve the accuracy of runoff prediction,the research proposes a combined method of marine predator algorithm(MPA)and long short-term memory(LSTM)neural network,6 standard test functions are selected to simulate and verify MPA and compare with the simulation results of PSO algorithm.MPA is used to optimize key parameters such as the number of neurons in the hidden layer of LSTM and training times,based on principal component analysis(PCA)dimensionality reduction and non-dimensionality reduction processing to establish PCA-MPA-LSTM,MPA-LSTM runoff prediction models,and build PCA-LSTM,LSTM,PCA-MPA-support vector machine(SVM),MPA-SVM,PCA-MPA-BP,MPA-BP for comparison models.The measured data of Yunnan Provincial Observation Station is used to train and predict 8 models including PCA-MPA-LSTM and MPA-LSTM.The results show that:①The MPA simulation effect is better than the PSO algorithm,and it has better optimization accuracy and global search capability.②The average relative errors of PCA-MPA-LSTM and MPA-LSTM models for instance fitting and prediction are 1.18%,2.35%,1.94%,and 1.96%,respectively,and the prediction effect is better than that of the other 6 models,with good prediction precision and generalization ability.③MPA is used to optimize the key parameters of LSTM can effectively improve the generalization ability and prediction accuracy of LSTM;the prediction accuracy of the data dimensionality reduction model is better than that of the corresponding non-dimensionality reduction model,and the data dimensionality reduction processing can effectively improve the prediction effect of the model.