Abstract:Data -driven methods has the inspiring potential for sensors fault detection in heating, ventilation, and air-conditioning (HVAC) intelligence control system. However, the interpretability of pure data -driven methods is relatively poor. Moreover, pure data -driven methods heavily rely on quality of HVAC sensor data that causes pure data driven methods have inconsistent detection efficiency in various operation conditions and sensor locations, resulting in poor general applicability. To solve above problems, an enhanced data -driven model based on energy consumption correlation (ECC) is proposed. First, the collected data from sensors are collected in chronological order and undergo data preprocessing techniques for noise removal. Subsequently, the energy consumption data is utilized to differentiate the operating conditions of the HVAC system. Simultaneously, a mapping relationship is established between sensor data and energy consumption. It uses the mapping relationship to construct the ECC for faulty sensor detection. Finally, the diagnosis of air handling unit (AHU) sensor faults is achieved by employing the hybrid data -driven model based on the ECC. The experiment conducted a comparison between the hybrid data -driven model and two novel pure data -driven methods. The results of the experiment demonstrate that the hybrid data -driven model exhibits excellent fault detection efficiency across different operating conditions and sensor locations. The fault alarm rate of the hybrid data -driven model is 0 - 3.26% under various AHU sensor fault conditions. And the fault diagnosis rate of the hybrid data -driven model is 96.23% and 96.11% under single -operation condition and dual -operation condition, respectively. Furthermore, compared to pure datadriven methods, the hybrid data -driven model is less affected by data quality and possesses a broader applicability.

A Hybrid Model-Based Fault Detection Strategy for Air Handling Unit Sensors

Model-based Fault Detection and Diagnosis for HVAC.

A Novel Model-Based Fault Detection Method for Temperature Sensor Using Fractal Correlation Dimension

A hybrid FDD strategy for local system of AHU based on artificial neural network and wavelet analysis

AHU Sensor Fault Diagnosis in Various Operating Conditions Based on a Hybrid Data-Driven Model Combined Energy Consumption

A Sensor Fault Detection Strategy for Air Handling Units Using Cluster Analysis

A Hybrid Data-Driven Simultaneous Fault Diagnosis Model for Air Handling Units

Multiple faults diagnosis for sensors in air handling unit using Fisher discriminant analysis

A Fault Detection Model for Air Handling Units Based on the Machine Learning Algorithms

Fault Detection and Diagnosis for AHU Sensors with Fixed or Drift Bias

A Model-Based Online Fault Detection Method for Air Handling Units of Real Office Buildings

Sensor Fault Detection and Its Efficiency Analysis in Air Handling Unit Using the Combined Neural Networks

Data Mining Based Sensor Fault Diagnosis and Validation for Building Air Conditioning System

Fault Detection and Diagnosis for Sensors in Variable Air Volume Systems

Fault Detection Methods in Air-Conditioning System Based on Parametric Statistics Characteristics

A system-level incipient fault-detection method for HVAC systems

Model-based Fault Detection and Diagnosis of HVAC Systems Using Support Vector Machine Method

An Enhanced Principal Component Analysis Method with Savitzky–Golay Filter and Clustering Algorithm for Sensor Fault Detection and Diagnosis

Application of Self-tuning Models to Air Handling Units for Fault Detection

Detection and Diagnosis for Sensor Fault in Hvac Systems

A Study on Fault Detection and Diagnosis for VAV Air Handling Units of Real Buildings