Study On Condition Trend Analysis For Aero-Generator

To be the central component of the power system of the aircraft, the working status of aero-generator directly affects the power supply capability of the aircraft. Because of the long-term work in the complex and harsh environments, safety and reliability is important for the aero-generator. So, this paper shall focus upon the condition trend analysis and residual useful life prediction of aero-generator, and then we can get the information of its condition and residual life in time. The research is helpful to the safe and reliable operation of the aero-generator and enhance the combat capabilities of aircraft. It has broad application prospects.In this paper, we research the condition analysis technology for a certain type of aero-generator and predict its residual useful life. A specific experimental platform of aero-generator is used to perform long-term life prediction experiment and collect the related condition and life test data, and then make sure the condition and life characterization parameters of aero-generator. After pretreatment and thorough analysis of these experimental data, a corresponding GM(1,1) model based on grey system theory and a ARMA(p,q) model based on time series method are designed, and this two models are used respectively to conduct condition analysis research on the condition of the aero-generator. To further improve the accuracy of analytical results, there are two optimization algorithms, genetic algorithm (GA) and particle swarm optimization (PSO), are used to carry on the parameter optimization of the established models. With the research documented in this paper, the results show that the original and the optimized models can realize the function of condition trend analysis of the aero-generator. By the error analysis of the research results, we can see that the prediction error of the ARMA (p,q) model is less than other models’, so it is the best analysis model.Finally, we use the best analysis model, the ARMA (p,q) model based on PSO, to predict the residual useful life of the aero-generator. We get the residual useful life predictive result of the aero-generator after plot the life curve and finish forward prediction of condition trend analysis result. The result shows that the analysis models that we build in this paper can accurate analyze the condition trend of aero-generator and predict its residual useful life better. The research result is scientific and effective and all desired research goals have been completed.