A Simulation Environment for Self-Propelled Gun Failure Prediction Based on Probabilistic Neural Network

This paper presents the development of a simulation environment in Matlab/Simulink for the analysis of the Sub-System of a kind of self-propelled gun and the design of fault tolerant control laws for this configuration. The environment includes the input module, output module and Simulation Nucleus and modeling of several types of engine failure. The simulation environment capable of implementing failure prediction has been implemented successfully as groundwork for analysis of the engine subsystem of self-propelled gun. Preliminary simulation results show the simulation environment is efficient for the Probabilistic neural network. Both abnormal oil pressure and rotate speed failures have been simulated. The simulation environment provides the necessary tools for the development and testing of advanced fault tolerant control laws.

[1]  Mohamad Musavi,et al.  Improving the performance of probabilistic neural networks , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.

[2]  Ma Songde,et al.  A new radial basis probabilistic neural network model , 1996, Proceedings of Third International Conference on Signal Processing (ICSP'96).

[3]  Donald F. Specht,et al.  Probabilistic neural networks , 1990, Neural Networks.

[4]  A. Zaknich,et al.  A tunable approximately piecewise linear model derived from the modified probabilistic neural network , 2000, Neural Networks for Signal Processing X. Proceedings of the 2000 IEEE Signal Processing Society Workshop (Cat. No.00TH8501).

[5]  Mario G. Perhinschi,et al.  A Simulation Environment for Individual Blade Control Helicopters , 2009 .