Method for detecting transient faults of fire pump in quick starting process

Abstract

The invention aims at providing a method for detecting transient faults of a fire pump in a quick starting process. The detected object mainly comprises the fire pump, pipelines, a fire water tank, and the like. The method comprises the following steps: controlling the fire pump to start in a condition of different accelerated speeds and pipeline resistances; synchronously collecting signals, such as vibration, temperature, rotating speed, inlet and outlet pressure, and pipeline flow, of the pump by utilizing a data collecting card; transmitting the signals to a host computer in real time for treatment; calculating lift and efficiency of the pump at a transient state according to collected signals; analyzing and treating the characteristic of each physical quantity, including the peak value and the average value of each signal, in time domain during the starting process of the fire pump; performing EMD decomposition of the signals via an HHT method; performing Hilbert transformation of each IMG component obtained from decomposition to obtain characteristic parameters, such as the amplitude spectrum and the instantaneous frequency of each signal, representing operating state of the pump; and inputting the extracted characteristic parameters into a RBF neural network for fault diagnosis to reveal an transient operating state of the fire pump in the quick starting process.

Description

technical field [0001] The invention provides a method for detecting the transient failure of the fire-fighting water pump during the rapid start-up process. Specifically, when the fire-fighting water pump is started up quickly, the speed, flow rate, etc. The changing signal data is used to calculate the instantaneous head and efficiency of the fire pump, and the HHT method is used to process the pressure, transient head, efficiency, speed and non-stationary vibration signals to obtain characteristic parameters that can characterize the operating state of the fire pump, such as instantaneous frequency Input the extracted feature quantities into the RBF neural network for fault diagnosis, and analyze the change characteristics of various physical quantities during the start-up process of the fire pump, thereby revealing the transient characteristics of the fire pump in the process of changing the operating state, and carry out Troubleshooting. Background technique [0002] T...

AI Technical Summary

Problems solved by technology

The sudden start of a large pump unit may cause overload or current shock to the local power grid, causing damage to the load equipment, or even burning out

At the same time, the start-up process is often accompanied by huge pressure pulsations and shocks, which can easily cause damage to unit equipment and pipelines

Images