Water fingerprint prediction algorithm

Abstract

The invention relates to the technical field of water supply pipe network management, and discloses a water fingerprint prediction algorithm, which comprises the following steps of: 1) obtaining historical data of partitions needing to be calculated; 2) predicting data in the future 24 hours by using a time sequence method; 3) calculating data in the future 24 hours by using a moving average mode;4) calculating data in the future 24 hours by using a linear regression mode; 5) using a Kalman filtering method to integrate the actual collection data and the above two calculation data; and 6) calculating different confidence intervals of each algorithm by using a Kalman filtering algorithm. According to the algorithm, the credibility of various algorithms is calculated through Kalman filtering; prediction of comprehensiveness, compared with the prior art, the method has the advantages that the method can cope with various different special conditions, the application range is wider, the selected centralized basic algorithms have different characteristics and adapt to different scenes, and the situation that abnormal changes of returned data may be caused by various reasons and normalmonitoring and analysis of the data are affected by water meter flow data collected by remote transmission equipment is avoided.

Description

technical field [0001] The invention relates to the technical field of water supply pipe network management, in particular to a water fingerprint prediction algorithm. Background technique [0002] The water fingerprint prediction algorithm belongs to the water supply network management and can be applied in the fields of zonal metering and leakage control. [0003] At present, for the data collected by flowmeters or water meters in partition metering, due to equipment failure or signal abnormalities, 100% accuracy of the data cannot be guaranteed, and there may be data missing and glitches, and it is easy to receive such bad data. Interference affects the normal analysis of monitoring data, such as false alarms due to glitches. Although the existing technology can process this type of bad data, the processing method is relatively simple and the adaptability is not wide enough. The main problems are as follows: 1. The glitch data is directly eliminated. After the eliminatio...

AI Technical Summary

Problems solved by technology

[0003] At present, for the data collected by flowmeters or water meters in partition metering, due to equipment failure or signal abnormalities, 100% accuracy of the data cannot be guaranteed, and there may be data missing and glitches, and it is easy to receive such bad data. Interference affects the normal analysis of monitoring data, such as false alarms due to glitches. Although the existing technology can process this type of bad data, the processing method is relatively simple and the adaptability is not wide enough. The main problems are as follows: 1. The glitch data is directly eliminated. After the elimination, the data is empty at this moment. Although there is no glitch, it affects the continuity of the data; 2. The compensation for missing data is not smart enough, such as simply calculating the average value, etc., compared with the real data 3. Due to the lack of data and abnormal conditions, it affects the subsequent prediction and alarm algorithms, so a water fingerprint prediction algorithm is proposed to solve the above problems

Images