15 January 2020 in IEEE Access
Estimation of Remaining Useful Life (RUL) is a crucial task in Prognostics and Health Management (PHM) for condition-based maintenance of machinery. In order to transmit and store the sensor data for archiving and long term analysis, data compression techniques are regularly used to reduce the requirements of bandwidth, energy and storage in modern remote PHM systems. In these systems the challenge arises of how the compressed sensor data affects the RUL estimation algorithms. A main drawback of conventional statistical modeling approaches is that they require expert prior knowledge and a significant number of assumptions. Alternative regression based approaches and deep neural networks are known to have issues when modeling long-term dependencies in the sequential data. Recently Long Short-Term Memory (LSTM) neural networks have been proposed to overcome these issues and in this paper we create a LSTM network and data fusion approach that can estimate the RUL with compressed (distorted) data. The experimental results indicate that the proposed method is able to estimate RUL reliably with narrower error bands compared to other state-of-the-art approaches. Moreover, the proposed method is able to predict RUL from both the raw and compressed datasets with comparable accuracy.
13 April 2011 in Journal of Applied Mechanics
1 January 2002 in Proceedings of the Institution of Mechanical Engineers, Part G: Journal of Aerospace Engineering
This paper demonstrates the application of a numerical continuation method to dynamic piecewise aeroelastic systems. The aeroelastic system is initially converted into a state space form and then into a set of equations which solve the system as the motion moves between different linear zones in a free-play motion. Once an initial condition is found that satisfies these sets of equations, a continuation method is used to find all other possible solutions of the same period for a variation in any parameter. This process can then be repeated for different order systems, allowing the limit cycle behaviour of the whole system to be built up. The solutions found using this method have been shown to be the same as those found using a more traditional Runge-Kutta type of approach with a considerable time saving and added flexibility through multiple parameter variation.
1 March 1999 in Mechanical Systems and Signal Processing
1 June 2007 in 2007 IEEE International Fuzzy Systems Conference
Structural health monitoring (SHM) is the process of implementing a damage identification strategy for aerospace, civil and mechanical engineering infrastructure. Under this context, feature extraction is the process of identifying damage-sensitive information from measured data. Feature extraction is an essential component of a SHM system needed to convert raw sensor data into useful information about the structural health condition. The need for robust health monitoring and prognosis of components in remote or difficult-to-access locations is driving the advancement of sensing hardware and processing algorithms. In this paper a feature extraction algorithm, referred to as soft computing feature extraction algorithm, is developed to extract damage-sensitive information from measured response data of helicopter rotor-head components. The proposed feature extraction algorithm is based on a combination of discrete wavelet transform theory and fuzzy logic theory. The results of applying the proposed feature extraction approach to tie bar data are presented. Results show that the proposed algorithm is capable of extracting features sensitive to the degradation of tie bar systems.