非线性运算法在日平均温度计算上的应用

The Application of Non-linear Algorithm in Calculating Daily Mean Air Temperatures

采用的非线性方法是基于24 h的平均温度,通过下列4种方法计算得到的逐时平均、四次平均、三次观测数据加上加权订正所得数据的平均、最高最低平均法得到的日平均温度与逐时平均比较,同时求出各种结果和24 h计算所得平均温度之间的离散度,发现在日平均温度的计算上,除了所述的24 h平均温度计算方法外,4次平均方法求出的日平均温度具有较好代表性,其他两种方法计算偏差相对较大。

All the methods were compared with 24 hourly mean temperature values. Daily mean air temperature was calculated by four methods： mean hourly （Hourly）, four equally spaced hourly mean observations, three equally spaced hourly mean observations weighted with the last observation （Weighted） and the mean of the maximum and minimum daily temperatures （Max/Min）. It was assumed that the Hourly method best represented the daily mean air temperature and the other methods were compared to it. With the analysis of Mean bias errors （MBE） and root mean square errors （RMSE）. There have been found that the method of four equally spaced hourly mean observations best approach to the 24 hourly mean temperature values.