Evaluation of a Decoupling-Based Fault Detection and Diagnostic Technique - Part I: Field Emulation Evaluation

Abstract

Existing methods addressing automated fault detection and diagnosis (FDD) for vapor compression air conditioning system have good performance for faults that occur individually, but they have difficulty in handling multiple-simultaneous faults. The decoupling-based (DB) FDD method explicitly addresses diagnostics for multiple-simultaneous faults for the first time. This paper is the first part of a two-part evaluation of the DB FDD technique whose intent is to validate the DB FDD performance and demonstrate its applications. The first part focuses on sensitivity and robustness evaluation through controlled field emulation testing. Sensitivity tests with artificially introduced faults show that individual faults can be identified before they cause a 5% of degradation in cooling capacity, EER and sensible heat ratio. Robustness tests for forty-one multiple-simultaneous-fault combinations demonstrate that no wrong diagnosis occurs with only two false alarms and two sensitivity losses for a liquid-line restriction. The second part, accompanying the first one, focuses on field applications in California.