说明：  降水短期气候预测是一个非常复杂、重要的研究课题。为了提高其预测能力，拟采用1959—2011 年逐月74 项大气环流特征量序列、月平均500 hPa 高度场和月平均海温场，选取预测因子；用主分量分析方法提取样本数据中主要信息为综合因子。用粒子群优化人工神经网络方法，建立宣城市夏季降水短期气候预测模型。对2007—2011 年宣城市夏季降水预报检验结果表明，粒子群优化人工神经网络收敛速度快，迭代次数少；试报平均绝对误差是66.5 mm，绝对值平均相对误差10.5 ，预测精度高，具有很好的应用推广前景。
(Precipitation of short-term climate prediction is a very complex and important research topic. Intends to adopt in order to improve its ability to predict the the 1959-2011 monthly 74 atmospheric circulation feature series, monthly mean 500 hPa height field and monthly average sea surface temperature field, select the predictor extract the sample data using principal component analysis for the Synthesis factor. Artificial neural network using particle swarm optimization method, Xuancheng City in summer rainfall in short-term climate prediction model. 2007-2011 declared the city in summer precipitation forecast verification results show that the particle swarm optimization artificial neural network convergence speed, fewer iterations trial reported an average absolute error is 66.5 mm, the absolute value of the average relative error of 10.5 , high prediction accuracy, good application prospect.)