基于时间序列ARIMA模型的校园管网漏损分析

Leakage Analysis of Campus Pipeline Network Based on Time Series ARIMA Model

通过分析校园智能水表提供的实时供水系统运行数据,建立了时间序列分析模型ARIMA(p,d,q)。利用该模型对校园管网漏损定位进行分析,可以帮助各大高校即时发现和解决供水系统中存在的问题,提高校园服务和管理水平。首先,利用pandas工具对数据进行清洗操作和数据补零,并结合绘制的校园管网拓展图和0-1整数规划模型计算出每个水表的实际用水量;其次,根据用水量数据绘制时序图,观察ACF和PACF并根据时序图建立ARIMA模型;最后,通过对比实时数据和预测数据可以得知水管是否有漏损情况。

By analyzing the real-time water supply system operation data provided by the campus smart water meter,a time series analysis model ARIMA(p,d,q)is established.Using this model to analyze the location of leakage of the campus pipe network can help universities to find and solve problems in the water supply system in real time,and improve the level of campus service and management.Firstly,use the pandas tool to clean the data and zero-fill the data,and calculate the actual water consumption of each water meter based on the drawn campus pipe network expansion diagram and the 0-1 integer programming model;Secondly,draw the time series diagram according to the water consumption data,observe ACF and PACF and build an ARIMA model according to the time sequence diagram;Finally,by comparing real-time data and predicted data can know whether there is any leakage in the water pipe.