Central air-conditioning system fault detection and diagnosis method based on Bayesian network unit

Abstract

The invention provides a central air-conditioning system fault detection and diagnosis method based on a Bayesian network unit. The method comprises the following steps that firstly, domain knowledge representation and storage are carried out from multiple angles aiming at the diversity of a central air-conditioning system, and a device-level central air-conditioning system fault diagnosis knowledge base is constructed; secondly, a corresponding Bayesian network unit is selected according to the type of a device included in an actual system, whether predefined existence conditions are met or not is judged, and specific device Bayesian network unit instances are generated; then, on the basis of the topological structure diagram of a target central air-conditioning system, the Bayesian network unit instances are integrally connected, to generate a multi-level layered Bayesian fault diagnosis network instance; and finally, fault diagnosis is realized through a probability reasoning process of a Bayesian network by taking actually obtained diagnosis information as a drive. According to the method, a targeted fault diagnosis model is generated based on the reusable knowledge base and specific system information, and fault diagnosis schemes suitable for local conditions can be provided for different central air-conditioning systems.

Description

technical field [0001] The invention belongs to the field of big data analysis and fault diagnosis of building air-conditioning systems, and relates to fault reasoning and analysis under complex and variable diagnostic objects and uncertain diagnostic information, in particular to a central air-conditioning system fault detection and analysis based on Bayesian network units. Diagnostic methods and techniques. Background technique [0002] The central air-conditioning system has the advantages of convenient management, easy noise elimination and anti-vibration, etc., and is currently widely used in our country. This system usually consumes a lot of energy and forms a significant part of the building's energy consumption. The central air-conditioning system has a large number of equipment and a complex structure, and various failures are prone to occur during operation, causing the system to fail to meet the temperature and humidity requirements, resulting in serious energy w...

AI Technical Summary

Problems solved by technology

[0003] The central air-conditioning system has a complex structure, various forms, and great differences, resulting in poor portability of fault diagnosis models between different systems, making it difficult to popularize and apply

It is difficult to build a customized fault diagnosis model for the actual system, the process is cumbersome and time-consuming, and there is duplication of work for the common parts between systems

At present, there is still a lack of a feasible fault diagnosis model generation method, which can be transplanted and applied between different systems at low cost, and can obtain effective diagnosis results

Images