Search

Ideally, whole-building energy simulation programs model all aspects of a building that influence energy use and thermal and visual comfort for the occupants. An essential component of the development of such computer simulation models is a rigorous program of validation and testing. This paper describes a methodology to evaluate the accuracy of whole-building energy simulation programs. The methodology is also used to identify and diagnose differences in simulation predictions that may be caused by algorithmic differences, modeling limitations, coding errors, or input errors. The methodology has been adopted by ANSI/ASHRAE Standard 140, Method of Test for the Evaluation of Building Energy Analysis Computer Programs (ASHRAE 2001α, 2004). A summary of the method is included in the 2005 ASHRAE Handbook—Fundamentals (ASHRAE 2005). This paper describes the ASHRAE Standard 140 method of test and its methodological basis. Also discussed are possible future enhancements to ASHRAE Standard 140 and related research recommendations. [ABSTRACT FROM AUTHOR]

Having building energy simulation programs with the inherent capability to efficiently perform optimization either during the conceptual design phase or during the operational stage would provide professionals with a tool to reduce energy use in building systems. This paper starts by reviewing two popular simulation techniques for building energy systems—sequential component method (SCM) and equation-oriented method (EOM)—in terms of their advantages and disadvantages. Next, a hybrid simulation strategy, successive approximation method (SAM), is proposed, This strategy is inherently capable of dealing with both design and optimal operation while combining desirable features of both SCM and EOM. SAM is a more coherent strategy that involves two aspects: (1) formulation of the equations following SAM and (2) using the automatic differentiation (AD) technique to compute the Jacobian matrix necessary to solve the equations. Finally, an illustrative example is presented to demonstrate the feasibility of this approach and how its results compare to those of a traditional simulation method. [ABSTRACT FROM AUTHOR]