Method for generating pump station unit fault diagnosis system

Abstract

The invention provides a method for generating a pump station unit fault diagnosis system. The method comprises the following steps: collecting fault data of the pump station unit; extracting featurevectors from the collected data information, and selecting an optimal feature vector from the extracted feature vectors; carrying out permutation and combination on the optimal feature vector and a machine learning model through a genetic algorithm, and selecting the optimal combination of the feature vector and the machine learning model; combining the optimal feature vector with a machine learning model to form a diagnosis model for testing; and diagnosing the fault of the pump station unit by using the successfully tested diagnosis model. According to the invention, a genetic algorithm in the field of computer science artificial intelligence is adopted for automatic permutation and combination, the optimal feature vector and machine learning model combination is selected, manual processing and selection are not needed, cost is low, the period is short, and the pump station unit diagnosis system formed through the generation method can rapidly and effectively detect faults of the pump station unit.

Description

technical field [0001] The invention relates to the technical field of pumping stations, in particular to a method for generating a fault diagnosis system for pumping station units. Background technique [0002] With the development of science and technology and the national economy, water pump stations are widely used in industries such as agriculture, electric power, petroleum, mining, and water transfer across river cities, especially in drainage, irrigation, power generation, water supply, and adjustment of shipping water levels. play a pivotal role. The water pump unit is the key equipment of the water pump station, and its operating status directly affects the safe operation of the pump station. After the completion of my country's South-to-North Water Diversion Project, the pump station unit has a long running time, large investment, and high reliability requirements. The resulting interruption of water supply will cause significant economic losses and social impacts....

AI Technical Summary

Problems solved by technology

[0003] Existing fault diagnosis of pumping station units needs to collect fault data, process the fault data, and manually select machine learning models and parameters, which has high cost and long cycle. Updates are a big inconvenience

Images