Detection and Diagnosis of Multiple-Dependent Faults (MDFDD) of Water-Cooled Centrifugal Chillers Using Grey-Box Model-Based Method

This paper presents the development and use of benchmarking grey-box models for detection diagnosis multiple-dependent faults (MDFDD) a water-cooled centrifugal chiller. Models are developed using data recorded by Building Automation System (BAS) from central cooling plant an institutional building. The forward residual-based fault model identifies symptom, when difference between measured value target variable exceeds corresponding threshold. For diagnosis, most publications start known single establish impact on following variables in system. rule-based backward approach. proposed method if (i) symptom is correct (i.e., has abnormal values), or (ii) incorrect caused impacts generated other faulty due to dependency variables), (iii) both regressor abnormal. testing MDFDD model, some artificial inserted into measurement file, results discussed about potential application.