John M. House

Subsystem level fault diagnosis of a building's air-handling unit using general regression neural networks

This paper describes a scheme for on-line fault detection and diagnosis (FDD) at the subsystem level in an Air-Handling Unit (AHU). The approach consists of process estimation, residual generation, and fault detection and diagnosis. Residuals are generated using general regression neural-network (GRNN) models. The GRNN is a regression technique and uses a memory-based feed forward network to produce estimates of continuous variables. The main advantage of a GRNN is that no mathematical model is needed to estimate the system. Also, the inherent parallel structure of the GRNN algorithm makes it attractive for real-time fault detection and diagnosis. Several abrupt and performance degradation faults were considered. Because performance degradations are difficult to introduce artificially in real or experimental systems, simulation data are used to evaluate the method. The simulation results show that the GRNN models are accurate and reliable estimators of highly non-linear and complex AHU processes, and demonstrate the effectiveness of the proposed method for detecting and diagnosing faults in an AHU.

A Damper Control System for Preventing Reverse Airflow Through the Exhaust Air Damper of Variable-Air-Volume Air-Handling Units

Traditional air-handling unit (AHU) control systems link the position of the exhaust, recirculation, and outdoor air dampers. Laboratory tests of a variable-air-volume AHU using the traditional damper control approach revealed that outdoor air could enter the AHU through the exhaust air damper. This can negatively impact indoor air quality. This paper examines the conditions that lead to this phenomenon and presents a new control system that can help alleviate the problem. The new control system links only the position of the exhaust and recirculation air dampers. During occupied times, the outdoor air damper is fully open. Simulation results are presented that demonstrate that the new damper control system prevents air from entering the AHU through the exhaust air outlet for all but extreme conditions that are described in the paper. Laboratory and field test results are presented that demonstrate that the new control system prevents air from entering the AHU through the exhaust air outlet for the same conditions that cause significant reverse airflow for the traditional control system. Furthermore, reverse airflow was not observed for any of the conditions examined in the laboratory and field tests when the new control system was used.

Integrated Control and Fault Detection of Air-Handling Units

A new method for integrated control and fault detection of air-handling units (AHUs) is described in this paper. The method uses sensors commonly installed in AHUs and collects much of the key diagnostic information at times when steady-state conditions are imposed on the AHU by the sequencing logic, thereby eliminating the need for a steady-state detector. A model-based fault detection method processes these data and generates residual values that can be further processed to detect faults. For faulty operation, one or more of the residuals is expected to have a value that is significantly different from 0, the expected value for normal operation. In parallel to model-based residual generation, an algorithm calculates state-based performance indices of control loops for the processes used for sequential control of the AHU. The method was assessed through simulations of 16 faults, which consisted of temperature sensor offset faults, stuck and leaking damper faults, and stuck and leaking valve faults. With the exception of the return air temperature sensor offset faults, each fault impacted numerous residuals. In addition, two faults (stuck-closed recirculation air damper and heating coil valve stuck 10% open) caused control performance indices to saturate, which should not happen for a properly designed and operating system. The complementary aspect of the model-based residuals and state-based performance indices is described and illustrated with these faults.

Optimization and sequencing of chilled-water plant based on extremum seeking control

Chilled-water plants with multiple chillers are the backbone of ventilation and air conditioning (VAC) systems for commercial buildings. Optimal operation and sequencing of such plant has great impact on building energy efficiency. Large variations in equipment characteristics and operating conditions make it difficult and expensive to obtain plant models required by model-based control/optimization approaches. For a variable primary flow (VPF) chilled-water plant with parallel chillers, a model-free optimization and sequencing strategy is proposed to maximize energy efficiency in real time, based on a penalty-function based multivariate extremum seeking control (ESC). The ESC takes the total power consumption (chiller compressors + cooling tower fan + condenser water pumps + input-saturation penalty) as feedback, while tower fan airflow, condenser water flows and evaporator leaving chilled-water temperature setpoint as manipulative inputs. A band-pass filter array replaces the high-pass filter in the standard ESC to reduce the cross-channel interference. The chiller sequencing is enabled with input saturation related signals. Simulation study is performed under several testing conditions using a Modelica based dynamic simulation model of a chilled-water plant with two parallel chillers, one cooling tower, one air-handling unit and one zone. Simulation results validate the effectiveness of the proposed strategy with significant energy saving demonstrated.

Extremum Seeking Based Control Strategy for a Chilled-Water Plant With Parallel Chillers

Chilled-water plants with multiple chillers are the backbone of ventilation and air conditioning (VAC) systems for commercial buildings. A penalty function based multivariate extremum seeking control (ESC) strategy is proposed in this paper for maximizing the energy efficiency in real time for a variable primary flow (VPF) chilled-water plant with parallel chillers. The proposed ESC algorithm takes the total power consumption (chiller compressors + cooling tower fan + condenser water pumps + penalty terms if inputs saturation occurs) as feedback, and tower fan air flow, condenser water flows and evaporator leaving chilled-water temperature setpoint as plant inputs (ESC outputs). A band-pass filter array is used in place of the conventional high-pass filter at the plant output so as to reduce the cross-channel interference. Chiller sequencing is also enabled with input saturation related signals. A Modelica based dynamic simulation model is developed for a chilled-water plant with two parallel chillers, one cooling tower, one air-handling unit and one zone. Simulation results under several testing conditions validate the effectiveness of the proposed model-free control strategy, as well as the significant energy saving.Copyright

INTEGRATED CONTROL AND FAULT DETECTION OF AIR-HANDLING UNITS

Abstract A new method for integrated control and fault detection of central air-handling units based on finite state machine sequencing logic is described. The method evaluates thirteen residuals derived from mass and energy balances applied at specific operating conditions imposed on the air-handling unit by the sequencing logic. For faulty operation, one or more of the residuals is expected to have a value that is significantly different from zero, the expected value for normal operation. Computer simulations were used to test the method for several faults common to air-handling units.

Offset-free model predictive controller of a heat pump

This work presents an offset-free model predictive control (OF-MPC) design for energy efficient control of a heat pump. Open-loop system dynamics are examined first to determine the appropriate control structure leading to identification techniques being used to construct linear models of an experimentally-validated heat pump model. Subsequently, a framework consisting of a model predictive controller, cascaded upon a lower level PI controller, is designed. The MPC includes an augmented model (including disturbance states) and an associated Luenberger observer to estimate the prediction disturbance (plant-model mismatch at steady state). Simulation results subject to realistic disturbances and measurement noise demonstrate that energy savings of 2.3% can be achieved with the proposed OF-MPC design compared to a traditional control approach.

Mode switching control for a multi-functional variable refrigerant flow system

The multi-functional variable refrigerant flow system enables simultaneous heating and cooling of the zones it serves by allowing the heat exchangers of both indoor units and outdoor units to operate as evaporators or condensers. To realize automatic and smooth switching between different modes, a mode switching strategy is proposed for a multi-functional variable refrigerant flow system. Whether to turn on or off an indoor unit is determined by the zone temperature and a preset hysteresis band about the temperature set-point. Based on a thermodynamic analysis, a decision variable for determining outdoor unit-heat exchangers mode switching is proposed as the outdoor unit-heat exchangers airside temperature differential normalized by the dimensionless outdoor unit fan speed. Two bumpless transfer strategies are applied to produce smooth mode switching. To evaluate the proposed strategy, a simulation study is performed with a Modelica-based dynamic simulation model of a four-zone multi-functional variable refrigerant flow system. Simulation results validate the effectiveness of the proposed strategy and the performance of the bumpless transfer strategies.

Multi-variable extremum seeking control for a multi-functional variable refrigerant flow system

A multi-variable extremum seeking control is proposed as a model-free strategy to maximize the energy efficiency in real time for a multi-functional variable refrigerant flow system. For the cooling-only, heating-only, cooling-dominated, heating-dominated, and heat-recovery modes, the manipulated inputs of extremum seeking control include compressor suction/discharge pressure set-point, outdoor unit fan speed, bypass flow valve opening, and superheat set-points for evaporating heat exchangers. The feedback of extremum seeking control is the total power of the compressor motor, outdoor, and indoor unit fans. The proposed control strategy is validated with a Modelica-based dynamic simulation model, in terms of both steady-state and transient performance.

Heating, ventilation and air conditioning systems: Fault detection and isolation and safe parking

Abstract This work presents an integrated framework for fault detection and isolation (FDI) and fault tolerant control (FTC) of variable air volume (VAV) boxes, a common component of heating, ventilation and air conditioning (HVAC) systems. To this end, first a statistical model based FDI framework is designed using existing techniques such as principal component analysis (PCA) and joint angle analysis as a benchmark for comparison. Then a novel linear causal model based framework for FDI of multiple actuator and multiple sensor faults is designed and implemented and shown to possess superior FDI capabilities compared to the statistical model based framework. Finally, a safe parking strategy is designed and the ensuing energy savings for the case of stuck dampers demonstrated.