Machine pump health knowledge base searching method based on Bayesian reasoning

Abstract

The invention relates to a machine pump monitoring and fault diagnosis technology, and aims to provide a machine pump health knowledge base searching method based on Bayesian reasoning. The method comprises the following steps: taking machine pump key components, specific fault information and vibration signal characteristics in a machine pump health knowledge base as priori knowledge; establishing Bayesian network search model learning, calculating the score of the Bayesian network by utilizing a scoring function to establish a new database, and inferring the fault probability of the node according to the real-time state of the vibration signal characteristics obtained in the operation process of the machine pump by utilizing the obtained Bayesian network. According to the method, a penalty item can be effectively adjusted during training of some data, excessive fitting is prevented, the topological structure of the Bayesian network can be effectively determined, and the conditional probability density of each node of the Bayesian network model is obtained through priori knowledge, so that the Ankang state and the fault type in the operation process of the pump are obtained. The safety degree and fault diagnosis in the operation process of the pump are analyzed and diagnosed, and safe and efficient operation of the pump is achieved.

Description

technical field [0001] The invention relates to machine pump monitoring and fault diagnosis technology, in particular to a Bayesian reasoning-based search method for a machine pump health knowledge base, which belongs to the field of intelligent control. Background technique [0002] Fans, water pumps, compressors, steam turbines and other mechanical and pump equipment are the key production equipment of modern enterprises such as petroleum, chemical industry, metallurgy, steel and electric power. With the development of intelligence, this type of equipment continues to develop towards high speed, heavy duty, high efficiency and intelligence. At the same time, higher requirements are put forward for the condition monitoring and fault diagnosis system of the equipment. The running status of the pump directly affects the production efficiency of the enterprise, and even causes safety problems in the event of a serious failure. Therefore, the health monitoring of pump equipmen...

AI Technical Summary

Problems solved by technology

However, such monitoring technology requires a lot of manpower for professional data analysis and query, and has no predictive function, and the accuracy is not high

Images