Abstract:Current research on chiller fault detection and diagnosis (FDD) is usually based on the multi-classification method for mixed multiple fault types. The establishment of a multi-classification FDD model requires obtaining extensive test data of normal operation and multiple common faults at different fault levels, so the cost-to-benefit ratio will not be accepted. In addition, because of the information interference between different types of fault data, FDD model is hard to avoid over-fitting or under-fitting, and there may be an intolerable high false alarm rate existing in product application. Therefore, it is difficult to promote FDD in actual chiller product applications. This paper proposes an inverse fault detection and diagnosis (IFDD) strategy, which is based on the characteristics of the target fault (the fault to be concerned) to establish a two-classification fault diagnosis model. Compared with the multi-classification method, the amount of fault test data required by the two-classification methods to realize fault diagnosis can be greatly reduced. With IFDD strategy, the detection and diagnosis are completed simultaneously, while fast calculation speed, high diagnosis accuracy rate and low false alarm rate are achieved. The core of the IFDD strategy is to concentrate resources to capture the characteristics of the target fault, and no longer classify other types of faults and normal operating conditions, so it can avoid the influence of information interference between multiple types of fault data in the fault diagnosis model. In order to verify the feasibility of this strategy, a 200 ton variable-speed screw chiller has been tested on the laboratory, and a database of normal operation, refrigerant leakage, refrigerant overcharge faults, and a new operating condition of insufficient cooling water flow is initially established. Based on the sensors configured by the chiller unit and using the exponentially weighted moving-average (EWMA) control chart for IFDD, the model's fault diagnosis accuracy for the target fault reaches 100%, and no false alarm appears after the transition period.

Study On Fault Indicative Features In The Automated Fault Detection And Diagnosis (Fdd) For Chillers

Important Sensors for Chiller Fault Detection and Diagnosis (fdd) from the Perspective of Feature Selection and Machine Learning

Study on a Hybrid SVM Model for Chiller FDD Applications

Feature selection for chillers fault diagnosis from the perspectives of machine learning and field application

Chiller Fault Detection and Diagnosis Based on Genetic Algorithm and Support Vector Machine with Parameter Tuning

Fault diagnosis strategy for incompletely described samples and its application to refrigeration system

A Robust Pattern Recognition-Based Fault Detection and Diagnosis (FDD) Method for Chillers

An Inverse Fault Detection and Diagnosis (IFDD) Strategy for Practical Application on Chiller Product

Model-based Fault Detection and Diagnosis for HVAC.

A Machine Learning Framework for Detection and Severity Estimation of Faults for Chillers and Air Handling Units in HVAC systems

A Statistical Fault Detection and Diagnosis Method for Centrifugal Chillers Based on Exponentially-Weighted Moving Average Control Charts and Support Vector Regression

An intelligent fault detection and diagnosis model for refrigeration systems with a comprehensive feature selection method{fr}Un modèle intelligent de détection et de diagnostic des défauts pour les systèmes de réfrigération avec une méthode complète de sélection des caractéristiques

Fault Diagnosis Based on Fusion of Residuals and Data for Chillers

Pattern recognition-based chillers fault detection method using Support Vector Data Description (SVDD)

A statistical-based online cross-system fault detection method for building chillers

Partial Domain Adaption Based Prediction Calibration Methodology for Fault Detection and Diagnosis of Chillers under Variable Operational Condition Scenarios

Transfer Learning Based Methodology for Migration and Application of Fault Detection and Diagnosis Between Building Chillers for Improving Energy Efficiency

An Improved Fault Detection Method for Incipient Centrifugal Chiller Faults Using the PCA-R-SVDD Algorithm

Automated FDD of multiple-simultaneous faults (MSF) and the application to building chillers

PCA-SVM-Based automated fault detection and diagnosis (AFDD) for vapor-compression refrigeration systems

A Sensor Fault Detection and Diagnosis Strategy for Screw Chiller System Using Support Vector Data Description-Based D-statistic and DV-contribution Plots