以水司日常记录的日最高气温、日最低气温、天气阴晴量和工作日性质作为主影响因素,采用模式识别方法,识别出与预测日最相似的用水量模式;按照加权平均法计算出预测日的基准用水量;利用实时反馈的用水量,从累计误差趋势和当前误差趋势对预测日基准用水量进行实时动态修正。%In order to predict the short-term water demand, the daily maximum temperature, daily minimum temperature, the weather and the work state of the day, which are recorded in the daily record of the plant, were considered as the main inlfuencing factors. The dates which had the most similar main inlfuencing factors were identiifed through the water consumption pattern recognition process. The baseline water consumption was calculated by sum water consumption of the dates chosed by pattern recognition process with weighted coefifcient. The baseline water consumption was amended with dynamic correction by using of real-time feedback data of the water consumption, from the cumulative and current error trend.
展开▼