Abstract:With a data smoothing process method and mean variance analysis,the precipitation data are divided into certain groups,and the precipitation forecasting model based on Markov model is established.Taking the precipitation data of Rugao in November from 1980 to 2015 as an example,the self-correlation coefficients are calculated in order to obtain Markov weight coefficients,and then the eigenvalues are calculated by the moving weighted Markov chain.The precipitation of November in 2016 is predicted.The relative error between the predicted value and the measured value is 6.6%,which is smaller than that of the ordinary weighted Markov model,and the improvement is effective.
慕凯, 张祥, 余士龙, 李俊, 李宇, 戴笑俊. 滑动加权马尔科夫模型在降水量预测中的应用[J]. 气象水文海洋仪器, 2019, 36(4): 34-36.
Mu Kai, Zhang Xiang, Yu Shilong, Li Jun, Li Yu, Dai Xiaojun. Application of a moving weighted Markov model in precipitation prediction. Meteorological Hydrological and Marine Instrument, 2019, 36(4): 34-36.