A state monitoring method for refrigerated dehumidifiers based on genetic fuzzy c-means clustering

Abstract

The invention discloses a state monitoring method of a refrigerated dehumidifier based on genetic fuzzy C-means clustering. The method includes selection of equipment measurement parameters, simulation of working conditions, collection of data samples, calculation of standard class centers, and judgment of state. This enables condition monitoring of the dehumidifier. In the calculation process of the standard class center, the fuzzy C-means clustering method based on genetic algorithm improvement is used. The improvement is mainly reflected in two aspects: on the one hand, the genetic algorithm is used to automatically calculate the initial cluster number of fuzzy C-means clustering, This replaces the traditional manual selection method, reduces the influence of human subjective factors, and improves the accuracy and scientificity of cluster number selection; on the other hand, when the cluster number is obtained, the genetic algorithm is used to calculate the cluster center , to get the global optimal solution, which overcomes the problem of being sensitive to the initialization value and easily falling into the local minimum in the traditional fuzzy C-means clustering solution.

Description

technical field [0001] The invention belongs to the field of HVAC and refrigeration state monitoring and fault diagnosis, and in particular relates to a state monitoring method of a refrigerated dehumidifier based on genetic fuzzy C-means clustering. Background technique [0002] With the needs of social development and production, refrigerated dehumidifiers are widely used in various occasions that require environmental temperature and humidity, such as large warehouses, underground engineering, commercial buildings, electronics and precision instruments, textiles and other fields. Air humidity and temperature adjustment to a certain extent. Medium and large refrigerated dehumidifiers are usually mechatronic equipment, mainly composed of refrigeration, ventilation, temperature control and electronic control, etc., and their working characteristics have the characteristics of large inertia, strong coupling, nonlinearity and multiple interferences. Monitoring the condition o...

AI Technical Summary

Problems solved by technology

[0004] There are two defects in the application of the traditional fuzzy C-means clustering method: first, the initial number of clusters is determined by the λ-cut matrix classification method, and the λ value is artificially selected by experience, and different λ values ​​determine different clusters This may lead to deviations in classification, which will affect its fault monitoring and diagnosis applications; second, the method uses an iterative hill-climbing algorithm to find the optimal solution to the research problem, which is a local search algorithm and is sensitive to the initialization value. Easy to get stuck in local minima

Images