摘要
降水状态的规律性隐藏在历史数据中,可用状态转移频率替代状态的转移概率。不同时段状态对预测的状态影响权重不相同,可用降水序列的自相关系数定量描述,由此建立基于随机过程的中长期降水预测模型,可预测年降水状态及相应的降水量范围。
The precipitation course is regarded as random process. The regularity of precipitation state conceals in the historical data and the state transfer probability can be replaced by the state transfer frequency. The states in different periods have different influences on forecasting state and the weight can be described quantitatively with autocorrelation coefficient of randomse- quenee. Accordingly, the middle and long-term precipitation model is constructed, with which annual precipitation state and range can be forecasted.
出处
《唐山学院学报》
2007年第2期7-9,共3页
Journal of Tangshan University
关键词
随机变量
随机过程
状态转移概率
权重
预测模型
random variable
random process
state transfer probability
weight
forecasting model
