A statistical fault detection and diagnosis method for centrifugal chillers based on exponentially-weighted moving average control charts and support vector regression

► A new FDD strategy combining EWMA control charts and Support Vector Regression (SVR) is proposed. ► EWMA control charts are used to overcome the shortcomings of the available statistic-based chiller FDD strategy, e.g. the t-statistic-based strategy. ► Support Vector Regression (SVR) is adopted to improve the accuracy of reference models. ► A new performance index was proposed to represent the effects of faults on the thermal performance of sub-cooling section. ► The proposed strategy improved the FDD performances significantly, especially on the faults at low severity levels.