Ship area distribution power system fault detection system and method based on time scale measurement

Abstract

The invention provides a ship area distribution power system fault detection system based on time scale measurement. The ship area distribution power system fault detection system comprises a time scale measurement and acquisition module; an information fusion processing module; a joint fault detection module; and a neural network identification and supervision module. A supervision adjustment unit in the neural network identification and supervision module performs parameter adjustment on the time scale measurement and acquisition module, the information fusion processing module, the joint fault detection module and an intelligent operation and protection execution module according to an identified working condition and an information time sequence comparison result. By applying the embodiment of the invention, the accuracy is improved, and online parameter adjustment of the processing modules is realized. The memory function of the LSTM and the feature extraction capability of the CNN are utilized to extract features from information within a period of time before and after a fault instead of training a network by a certain state feature, the robustness of the network is enhanced, forecasting can be carried out in advance when the same fault is encountered next time, so that the system reliability is enhanced.

Description

technical field [0001] The invention relates to the technical field of fault detection of a ship's regional power distribution power system, in particular to a fault detection system and method for a ship's regional power distribution power system based on measurement with a time scale. Background technique [0002] With the rapid development of my country's shipbuilding industry, the capacity of the ship's regional power distribution system is increasing, the internal structure is becoming more and more complex, and its safety and reliability requirements are getting higher and higher. The modern ship regional power distribution power system requires real-time measurement of operating parameters, online detection of fault conditions, rapid response to abnormal conditions, and corresponding measures to avoid the occurrence of ship regional power distribution power system failures; rapid isolation of faults, Reduce the impact of failures on the ship's regional power distribut...

AI Technical Summary

Problems solved by technology

However, the currently used neural network is a shallow network, which cannot fully mine the coupling relationship between data, thus limiting its scope of application in the detection of faults in power distribution systems in ship regions.

At the same time, the existing methods rarely consider the time and space attributes of the operating parameters of the ship's regional distribution power system, which also affects the accuracy of the fault detection results.

Images