Search

Dynamic simulation tools that could accurately simulate operational data for both the fault-free and faulty dynamic operation of heating, ventilation, and air conditioning (HVAC) systems and equipment are needed for developing and evaluating advanced control and automated fault detection and diagnosis strategies. Among various HVAC subsystems, fan coil units (FCUs) are relatively simple, inexpensive devices that are used extensively in commercial, institutional and multifamily residential buildings. However, very little has been reported in the literature to improve FCU design and operation. There has also been a lack of dynamic simulation tool development focusing on FCUs. The work reported in this study aims at developing and validating a software tool to simulate operational data generated from FCUs that are operated dynamically under both faulty and fault-free conditions. A comprehensive and systematic validation process, using data collected from real FCUs in a laboratory building, is used to validate the tool under both faulty and fault-free operating conditions in different seasons. The validated tool not only is able to predict real-world FCU behaviors under different control strategies, but it is also able to predict symptoms associated with various faults, as well as the effects of those faults on system performance and occupant comfort. [ABSTRACT FROM AUTHOR]

Efficiently, robustly and accurately solving large sets of structured, non-linear algebraic and differential equations is one of the most computationally expensive steps in the dynamic simulation of building energy systems. Here, the efficiency, robustness and accuracy of two commonly employed solution methods are compared. The comparison is conducted using the HVACSIM+ software package, a component based building system simulation tool. The HVACSIM+ software presently employs Powell’s Hybrid method to solve systems of nonlinear algebraic equations that model the dynamics of energy states and interactions within buildings. It is shown here that the Powell’s method does not always converge to a solution. Since a myriad of other numerical methods are available, the question arises as to which method is most appropriate for building energy simulation. This paper finds considerable computational benefits result from replacing the Powell’s Hybrid method solver in HVACSIM+ with a solver more appropriate for the challenges particular to numerical simulations of buildings. Evidence is provided that a variant of the Levenberg-Marquardt solver has superior accuracy and robustness compared to the Powell’s Hybrid method presently used in HVACSIM+. [ABSTRACT FROM AUTHOR]