DEVELOPMENT AND FLIGHT TEST EVALUATION OF A ROTOR SYSTEM LOAD MONITORING TECHNOLOGY

Real-time rotor system load monitoring algorithms have been developed and evaluated in a flight test of an instrumented SH-60 helicopter. Vibratory loads are predicted using an artificial neural network that utilizes input parameters which are typically measured and stored on a flight data recorder. These load prediction algorithms were developed using an extensive SH-60 flight data set which includes approximately 9 flight hours of data and a wide variety of SH-60 usage spectrum maneuvers. The neural network based load monitoring capability was demonstrated in a flight test evaluation where the algorithms were embedded in an on-board processor and flight data recorder for real-time on-board load predictions. Prediction results are compared with strain gauge measurements from two separate flight test aircraft. Assessment of load predictions shows good results and indicates that the neural network algorithms developed using flight data from one aircraft can be used to predict vibratory loads on a second aircraft. The flight test results also demonstrate the potential to use the algorithms for real-time pilot cueing to minimize operation in fatigue damaging flight conditions.