This paper discusses development of a library of equation-based models for building HVAC system simulation. The main source of the models is the ASHRAE Secondary Systems Toolkit (Brandemuehl 1993), augmented with primary system models from other sources in order to provide a library sufficient to model conunon HVAC systems.

Examples on application of system simulation in preplanning, sizing and comissioning of HVAC systems are briefly given and typical obstacles for a more broader utilization in engineering offices are introduced. Efforts needed to overcome most of the obstades are described. It is shown that currently system simulation is a business for experts. To increase the application of simulation in HVAC engineering offices efforts should be made in training and supporting these experts. 

This study was carried out on a real site gymnasium situated in the center of France. The building is equipped with two remarkable energy saving systems; running together: a ventilated roof and an air-earth exchanger. 

A methodology is presented for creating models which are suitable for use in fault detection and diagnosis schernes in applications where it is impossible to obtain data from the actual plant. Generic qualitative models based on fuzzy rules are used to describe the basic features of the behaviour of a class of plants of similar design. The generic models are identified off-line from training data produced by computer simulation of typical plant designs.

The Research and Development Division at Gaz de France assesses; and improves building heating and cooling equipment by both experimental and modelling/simulation approaches. In this context, the R&D Division uses two tools to model the thermal dynamic behaviour of buildings: the ALLANTM. Simulation modeller connected to the NEPTUNIX solver, and the CSTBat software.  

In computer simulation, accurate modelling of air conditioning equipment is important in the studies of dynamic plant performance, for instance in the selection process of a plant control scherne, in the investigation of plant energy consumptions, or in the detailed design of a plant equiprnent.

The Neutral Model Format for building simulation was proposed in 1989 as a means for documentation and exchange of models. It has attracted much interest and an acceptance as a potential standard, maintained by ASHRAE's TC 4.7 technical committee. So far, the format has only been directed towards component (leaf) models, but many suggestions have been made to extend it to also cater for systems; of component models. A brief review of NMF is given. This paper makes detailed proposals for NMF extensions covering hierarchical.

The main purpose of this paper is to develop fault detection modules for BEMS (Building Energy Management Systems), a software to aid building operators in detecting and diagnosing faults in HVAC systems. The fault detection modules proposed are based on two fundamentally different approaches based on component models: physical models and neural net-works. These modules using these mo approaches are illustrated for a cooling coil of an Air Handling Unit.

The potential of the different passive cooling strategies for popular residential buildings in Israel is evaluated using a simplified simulation program. The program, which can be run on a regular PC computer, is based on a similar one developed by the European community, but takes into account the internal mass of the building more explicitly. The passive cooling strategies considered are: natural ventilation, night cooling, ground cooling using underground pipes and combinations thereof Both air conditioned buildings and free-floating temperature buildings are considered.

There is a continuing need to validate detailed thermal simulation programs of buildings. One way of doing this is to compare program predictions with measured building performance data. This is known as empirical validation. Data from the US National Institute of Standards and Technology passive solar test facility in Gaithersburg, MD, were used to assess predictions of ESP, HTB2 and SERI-RES. The results are tested for significance by means of Monte Carlo sensitivity analysis.