Fire-fighting pipe network water leakage monitoring method based on intra-class distance constraint auto-encoder

Abstract

The invention discloses a fire-fighting pipe network water leakage monitoring method based on an intra-class distance constraint auto-encoder. The method comprises the following steps of preprocessingacquired data to obtain a training data set with N samples; taking the training data set as input, training the first WICRAE and obtaining a coding result Y (1); training a second WIC-RAE by taking Y(1) as an input, and so on, training K WIC-RAEs in total, and obtaining a final coding result; performing single classification model training based on an RLS-OC loss function, and setting a threshold value by taking a part of training data set as an abnormal fire-fighting pipe network sample; comparing the fitting error of the newly acquired fire-fighting pipe network data with a threshold value, and judging whether the pipeline has abnormal conditions such as water leakage or not; and based on the direction of the water leakage pipeline, troubleshooting a water leakage point in the suspected water leakage pipeline at the foremost end of the direction and maintaining the suspected water leakage pipeline. The method is advantaged in that a problem that in the prior art, a fire-fighting pipe network cannot be effectively monitored is solved.

Description

technical field [0001] The invention relates to a pipeline protection system; the technical field of intelligent monitoring of pipelines, in particular to a water leakage monitoring method of a fire-fighting pipeline network based on an intra-class distance constraint self-encoder. Background technique [0002] Fire protection is of great significance to the protection of people's lives and property, so the daily monitoring and maintenance of fire protection pipelines is of great significance. However, because fire-fighting pipes are either exposed to the air or buried underground, they are in a static state for a long time and are easily affected by corrosion, high pressure and high temperature, resulting in leakage. At present, the leakage rate of my country's fire pipe network is more than 30%, which leads to the waste of water resources, the loss of water pump pressurization power, reduces the use efficiency of water supply facilities, and occupies the effective water ou...

AI Technical Summary

Problems solved by technology

However, in the existing technology, the leakage detection of the fire protection pipe network is mainly done manually, which has a large workload and low work efficiency, and the requirements for the personnel themselves are relatively high. It is required to use special equipment and be able to analyze, and the labor cost is high.

Images