Comparison of ANFIS and ANN Techniques in the Simulation of a Typical Aircraft Fuel System Health Management

Abstract

The performance of an aircraft can be improved by predicting the possible complications associated with the system. Prognostics and Health Management (PHM) methodology includes fault detection, diagnosis, and prognosis. In this paper, a comparison of Adaptive Neuro-Fuzzy Inference System (ANFIS) with Artificial Neural Network (ANN) based fault prognosis tool for a typical aircraft fuel system is proposed. The ANFIS is an expert system which works on logical rules. The inputs of both ANFIS and ANN are trained by considering the same input data and generate the corresponding control signal. These methods identify the presence of faults and mitigate them to maintain a proper fuel flow to the engine. Overlooking the presence of any faults in time could potentially be catastrophic which can lead to possible loss of lives and the aircraft as well. These proposed tools work on the logical rules developed as per the engine’s fuel consumption and quantity of fuel flow from the tanks. The results are compared and analyzed which demonstrate the superiority of ANFIS tool compared to ANN.