论文部分内容阅读
对标准粒子群算法进行了简化,并基于简化的算法给出了混沌粒子群算法的优化支持向量机算法.基于延河流域甘谷驿水文站1954—1992年的实测年径流和输沙数据、1992—1995实测月径流和输沙数据,利用该算法和常用的几种粒子群支持向量机算法、误差后向传播神经网络算法预测了1993—1997年期间的年径流量和输沙量、1996年的月径流和输沙量.几种算法的预测结果和实测数据进行了比较,通过相对误差、平均相对误差、均方根误差、一致性指标和有效系数等参数,比较了不同算法的预测效果.结果表明,支持向量机算法模拟效果优于神经网络算法,本文提出的基于改进粒子群算法的支持向量机算法的预测效果更好,可用于流域的径流和输沙量的模拟和预报.
The standard Particle Swarm Optimization (PSO) algorithm is simplified and an optimized support vector machine (SVM) algorithm is proposed based on the simplified algorithm. Based on the measured annual runoff and sediment transport data of Ganguyi Hydrological Station in the Yanhe River Basin from 1954 to 1992, 1992 -1995 measured monthly runoff and sediment transport data, the use of the algorithm and several commonly used particle swarm optimization support vector machine algorithm, error back propagation neural network algorithm to predict the annual runoff and sediment load during 1993-1997, 1996 Monthly runoff and sediment discharge.The prediction results of several algorithms are compared with the measured data, and the prediction results of different algorithms are compared by relative parameters, such as relative error, average relative error, root mean square error, consistency index and effective coefficient The results show that the SVM algorithm is better than the neural network algorithm, and the SVM algorithm based on the improved particle swarm optimization is better than the BP neural network algorithm, which can be used to simulate and forecast the runoff and sediment load in the basin.