A sensor fault detection strategy for air handling units using cluster analysis

Abstract

. (2016). A sensor fault detection strategy for air handling units using cluster analysis. Automation in Construction,[70][71][72][73][74][75][76][77][78][79][80][81][82][83][84][85][86][87][88] A sensor fault detection strategy for air handling units using cluster analysis AbstractSensors are an essential component in the control systems of air handling units (AHUs). A biased sensor reading could result in inappropriate control and thereby increased energy consumption or unsatisfied indoor thermal comfort. This paper presents an unsupervised learning based strategy using cluster analysis for AHU sensor fault detection. The historical data recorded from sensors is first pre-processed to reduce the dimensions using principal component analysis (PCA). The clustering algorithmOrdering Points to Identify the Clustering Structure (OPTICS) is then employed to identify the spatial separated data groups (i.e. clusters),which possibly indicate the occurrence of sensor faults. The data points in different clusters are then checked for temporal separation in order to confirm the occurrence of sensor faults. The proposed sensor fault detection strategy is tested and evaluated with the data collected from a simulation system. The results showed that this strategy can detect single and non-simultaneously occurred multiple sensor faults in AHUs. The fault detection results were not strongly affected by the selection of the user defined input parameters required in OPTICS. (AHUs). A biased sensor reading could result in inappropriate control and thereby increased 10 energy consumption or unsatisfied indoor thermal comfort. This paper presents an 11 unsupervised learning based strategy using cluster analysis for AHU sensor fault detection. 12The historical data recorded from sensors is first pre-processed to reduce the dimensions 13 using principal component analysis (PCA). The clustering algorithm Ordering Points to 14Identify the Clustering Structure (OPTICS) is then employed to identify the spatial separated 15 data groups (i.e. clusters), which possibly indicate the occurrence of sensor faults. The data 16 points in different clusters are then checked for temporal separation in order to confirm the 17 occurrence of sensor faults. The proposed sensor fault detection strategy is tested and 18 evaluated with the data collected from a simulation system. The results showed that this 19 strategy can detect single and non-simultaneously occurred multiple sensor faults in AHUs. 20 The fault detection results were not strongly affected by the selection of the user defined input 21 parameters required in OPTICS. 22

Literature Review for Sensor Impact Evaluation and Verification Use Cases - Building Controls and Fault Detection and Diagnosis (FDD)

“…Outlier detection, PCA 25 Zhao et al [211] Chiller Virtual sensor Virtual fouling monitor sensors 26 Li et al [212] Chiller Sensor fault NA 27 Yan et al [213] AHU Sensor fault NA 28 Wang et al [214] VAV terminal Sensor fault NA 29 Yan et al [215] Chiller Feature selection from available sensors Feature selection with back-tracing sequential forward feature selection 30 Sun et al [216] Chiller Sensor data analysis/mining, sensor fault Data fusion 31 Hu et al [217] Chiller Sensor data analysis/mining, sensor fault Sensitivity of chiller sensor fault detection-based on PCA 32 Dey and Dong [218] AHU Sensor fault NA 33 Yang et al [219] AHU Sensor fault NA 34 Padilla and Choiniere [220] AHU Sensor fault NA 35 Kim et al [221] IAQ Sensor fault validation, PCA on sensor Sensor validation 36 Li et al [222] AHU Feature selection from available sensors NA 37 Kim and Braun [223] Chiller Virtual sensor NA 38 Shahnazari et al [224] VAV terminal Sensor fault, sensor data analysis/mining Sensor-fault-tolerant control 39 Wang et al [225] Chiller Feature selection from available sensors NA 40 Li et al [226] Outdoor unit of VRF Additional and built-in/existing sensors, sensor data analysis/mining Data mining using only built-in/existing sensors 41 Fernandez et al [227] AHU Sensor fault NA 42 Karami and Wang [228] Chiller Feature selection from available sensors, sensor fault NA Najafi [229] AHU Additional and built-in/existing sensor Sensor network architectures not necessarily designed solely for diagnostic purposes 44 Dey et al [230] HVAC terminal unit Sensor layout/location Impact of lacking sensor location 45 Kim and Braun [231] Chiller Virtual sensor NA 46 Pourarian et al [232] FCU Sensor fault FDD algorithm adap...…”

Sensor Fault Detection and Diagnosis Method for AHU Using 1-D CNN and Clustering Analysis

“…The fault detection stage determines the presence of any fault, while the fault diagnosis stage details the root cause of an issue. Generally, FDD methods can be divided into three categories [1, 8, 9]: the model-based, rule-based, and data-driven methods. Among them, the data-driven method attracts more attention and becomes the most popular FDD method in the field of HVAC [7].…”

Cyber-Physical Attack Detection and Recovery Based on RNN in Automotive Brake Systems

“…Various and large amounts of data required to control the vehicles are obtained from the equipped sensors in automotive CPS and are transmitted on vehicular networks. If any sensor is faulty temporarily during the operation due to environmental uncertainty, it may read wrong data [2].…”