Study on Optimization of Full Flow Test Process of Emergency Feed Pump for Nuclear Power Plant with Automatic Control Technology

Primary water supply is a core component in the second circuit of nuclear power plant, which is directly related to the safety and economy of the primary circuit. Considering the importance of the main supply pump in the power plant and the urgent vibration problem, it is urgent to study and deal with the fault of the main supply pump in order to solve the vibration warning problem that has occurred in the power plant. Effective regulation and improvement can prevent serious safety accidents caused by the spread of oscillation. This will reduce the downtime of nuclear power plants, maintenance spare parts and labor costs. In this paper, process parameters such as fluid flow and pressure are used in the process, combined with vibration state characteristics such as time domain and frequency domain, multi-parameter information with flow and pressure process parameters as working state variables is constructed, and a new method of multi-parameter fusion is constructed. This provides an early warning of the operating status of the equipment. Finally, based on the health information of equipment, this paper uses mechanical-personalized modeling to generate individualized equipment training samples. With personalized training samples as input, the diagnosis model based on deep CNN is trained to realize intelligent diagnosis of equipment.