基于阶次分析的发动机故障预测与识别-电子与通信工程专业论文.docxVIP

优秀毕业论文 精品参考文献资料 基于阶次分析的发动机故障预测与识别Engine 基于阶次分析的发动机故障预测与识别 Engine Fault Prediction and Identification Based On Order Analysis Ab stract Car plays an important role in the national lire，the engine as the heart of the Car，it’S failure to occupy a larger proportion of vehicle failure．In order to improve the engine’S safety and predict the fault in advance，the engine fault prediction technology based on vibration signal is developed rapidly．The engine vibration signal has the characteristics of non stability， nonlinear and SO on，the traditional frequency domain analysis Can produce the frequency fuzzy phenomenon,SO the order analysis based on the angle domain is developed． Durability test is all important part in engine development，A company dedicated to engine durability test，and purchased delta analyzer for detecting engine vibration．Enterprises hope to be able to carry out early fault prediction,early warning in early failure，to avoid permanent damage to the engine and cause significant losses．This study is based on the order signal generated by the delta analyzer to detect the vibration of the engine． This paper first analyzes the engine and delta analyzer’S working principle，the order analysis principle and other background knowledge．According to the current shortcomings of the fault prediction and needs of enterprises，design a complete set of solutions． First analyze the principle of fault and the corresponding fault spectrum,find the possible corresponding order for data preprocess． Second，using principal component analysis to reduce dimension of processed data, further compress data,eliminate redundancy and avoid“the curse of dimensionality”． Thirdly，fuzzy C-means and support vector machine is used to classify the reduced dimension data．The results show that this scheme Call aehieve the desired results． Forth,using Python and according to the requirements of enterprises to write the failure prediction software，using the above method． The solution proposed in thi