Fault Identification Of Aircraft Power System Based On Data Driven

The aircraft power system is the power source of electrical energy for all electrical equipment on the aircraft.Failure to detect and effectively handle failures in time may cause in failure or damage to the associated load equipment,thereby endangering flight safety.Therefore,it is of great significance to study the fault identification of aircraft power system.First,a risk assessment model of aircraft power system based on the fuzzy failure mode and effects analysis method(FMEA)and Technique for order preference by similarity to an ideal solution(TOPSIS)is established.Taking the typical components of the aircraft power system integrated drive generator as an example,the risk assessment research is carried out,and compared with the evaluation results of traditional methods.Then,a data-driven fault diagnosis model for aircraft power system is set up.The basic network structure of the model is a long-term and short-term memory network.Taking the aircraft integrated drive generator as an example,the fault diagnosis research is carried out.The simulation model is used to collect the fault simulation data and make a dataset.After several training,the ideal fault diagnosis model is determined.It is deployed on the server to provide fault diagnosis services for simulation data.Finally,based on the above research results,a maintenance management platform for aircraft power system is developed.The design and implementation of the main function modules,such as role management and privilege management,maintenance task query and task execution,fault diagnosis and fault warning,are briefly introduced,and the final results of the platform are displayed.The failure risk assessment model of aircraft power system in this paper can better characterize the inherent fuzziness of the evaluation index of aircraft power system risk,and avoid small changes in attribute values leading to changes in failure risk level ranking.The aircraft power system fault diagnosis model built does not need to rely heavily on the experience and intuition of experts and to some extent solves the problems of low efficiency and poor dexterity of traditional methods,and the accuracy is 98.57% under the existing experimental conditions and data sets.The maintenance management platform for aircraft power system developed in this paper integrates the related functions of maintenance management and fault diagnosis,can basically meet the needs of daily work,and can effectively improve the application level of maintenance information.