基于多模型聚合的PHM故障寿命估算方法

An emerging prognostic and health management(PHM) technology has recently attracted a great deal of attention from academia, industries, and governments. The need for higher equipment availability and lower maintenance cost is driving the development and integration of prognostic and health management systems. Taking advantage of advances in sensor technologies, PHM systems enable a pro-active maintenance strategy through continuously monitoring the health of complex systems. The core of PHM technology is prognostics which is able to predict the likelihood of a failure and estimate the remaining useful life (RUL) using the built-in predictive models.Rremaining useful life is time evalution of the fault indicators and is a random variable which represents the degradation process of a system or a component. The estimation of RUL exists uncertainty and remains a challege issue. In order to address this challege and improve the accuracy of RUL estimation for PHM, this research proposes a novel RUL estimation method to strategically integrate the strengths of multiple regression-based models by reducing error and uncertainty of the RUL estimation from each induvidul baseline model. The outputs from multiple baseline models will be fusioned with particle filtering (PF) to generate relatively precise RUL estimaton. In this research, the developed RUL estimation method will be validated using in-depth statistic analysis and will be applied to a real-world application, estimation of engine remaining useful life. The outcome from this research will have a great impact on Chinese PHM technology development and will be widely applied to high-tech industries such as aerospace, high-speed train, and nuclear power stations.

PHM技术是一个涉及人工智能，系统可靠性等多学科的热门研究课题。该技术通过对复杂系统的实时监控，用过去和现在的状态预测未来的状态。根据其预测结果，可以提前将故障排除，保证系统的健康运行，从而提高系统的可靠性和安全性。故障寿命估算是PHM的核心技术。高精度的故障寿命估算是保证PHM技术成功的关键因素。为了提高寿命估算的精确度， 本项目将研究一个基于PF(粒子滤波)技术的多模型聚合的故障寿命估算方法。该方法能从根本上减少单个模型的寿命估算的不确定性和误差，提高PHM故障寿命估算的精确度。本课题将对该多模型聚合的寿命估算方法进行深入的理论研究和深度的性能统计分析。其成果在航空、航天、核电等高端技术领域有着广泛的应用前景。

PHM技术是一个涉及人工智能，系统可靠性等多学科的热门研究课题。该技术通过对复杂系统的实时监控，用过去和现在的状态预测未来的状态。根据其预测结果，可以提前将故障排除，保证系统的健康运行，从而提高系统的可靠性和安全性。故障寿命估算是PHM的核心技术，高精度的故障寿命估算是保证PHM技术成功的关键因素。为了提高寿命估算的精确度，本项目成功研究开发了一个基于PF(粒子滤波)技术的多模型聚合的故障寿命估算方法。该方法能从根本上减少单个模型的寿命估算不确定性和误差，提高PHM故障寿命估算的精确度。本课题对该多模型聚合的寿命估算方法进行了深入的理论研究和深度的性能统计分析，其成果已经推广应用到飞机发动机和智能电网的变压器故障管理和故障寿命估算，取得了较为理想的结果。 该技术的开发成功，在航空、航天、核电等高端技术领域有着广泛的应用前景。