Thrust Estimation System for Military Engine Test Cell Applications

Engine thrust is a critical measurement in military engine acceptance tests, and this measurement may be too noisy or to extract meaningful information. Impact Technologies therefore proposes to develop an accurate real-time military engine thrust estimator to compute “virtual measurements” of engine thrust and other parameters of interest in a streaming, on-line fashion. Thrust estimation is a challenging task due to the complex dependency of thrust on the engine’s health condition and other unknown factors. To accurately capture engine-to-engine variation and engine degradation, a hybrid model is proposed that consists of a physics-based model and an empirical model based on Artificial Neural Networks (ANNs). The model will be used within a Kalman filter estimator structure reconstruct the true dynamic engine performance variables. This scheme will be enhanced by a sensor validation scheme to ensure robustness to sensor faults or noise that can occur in the test cell environment. The proposed Phase I is designed to establish a sound proof-of-concept capability with high prospects for maturation in DoD/commercial engine test cells and on-wing engine applications.