Keywords: neural network, artificial intelligence, solid model, building product modelling, AEC, CAD.

Redevelopment schemes and the refurbishment of old buildings accounts for a large proportion of the building activities conducted in industrial countries. In the interests of the long-term sustainability of the environment and resources it is therefore of particular importance to provide appropriate high-performance, computer-aided planning and simulation tools. The core aspect is a so-called product model, in which not only the three-dimensional geometry of an edifice but also all the relevant product data (materials, physical properties, ecological data, etc.) are stored and made available for the simulation of the entire life-cycle of the building [1,2]. An exact assessment and study of the current state of the building is indispensable for successful planning. For this reason, one of the first steps is to construct a geometric volume model using methods commonly practised in laser-supported engineering surveying. Once a "point cloud" - the resultant data from a laser scan - has been completed, it is analysed and initially transferred to a surface model. This forms the basis for the subsequent creation of a volume model, from which the product model is later derived in a second step. Here, we intend to focus our attention on this last step, in particular on automatic component identification.

Artificial intelligence methods make it possible to evaluate various criteria and, in contrast to the "hard" rules of a decision tree, depict the results as weighting factors for decision purposes. Any incorrect decisions that might occur through "narrowly missing" certain requirements are compensated, so that the overall assessment (probably) permits a correct decision, nevertheless. Artificial neural networks are examined in order to solve the problem concerned. One of the main tasks, therefore, is the training phase.

Three-dimensional volume objects (solids) form the basis for the building component classification process. Per definition they can be arbitrarily formed, so the collection of meaningful parameters is an important task. Once these data are available, they are brought into a form which is accessible by the net. The network, which has the same number of output neurons on the output side as types (categories) of building elements to be differentiated, provides an output value for each category, which accordingly has to be interpreted as net decision.

In order to identify building components, it is necessary to define decision parameters that are equally valid for every object. The approach presented in this paper captures the object geometry directly from the objects. In addition, it attempts to evaluate the objects almost regardless of their position within the global context. In this case, their relationship to neighbouring solids is not assessed.

One discovery that was made when using object data to train the net was that it is more convenient to use ratios than the pure dimensions of the objects. It was sufficient to use a simple network, but naturally a suitable database has to be available for the training phase. Using artificial neural networks to identify building components is a way of reacting flexibly to different samples. Forms of building elements that are hitherto unknown to the system are not excluded from the outset.

1

Neuberg, F.; Ekkerlein, C.; Rank, E.; Faulstich, M.: Integrated Life Cycle Simulation and Assessment of Buildings. In: Xth International Conference on Computing in Civil and Building Engineering (ICCCBE-X), Weimar, Germany 2004

2

van Treeck, C.; Rank, E.: Analysis of Building Structure and Topology Based on Graph Theory. In: Xth International Conference on Computing in Civil and Building Engineering (ICCCBE-X), Weimar, Germany 2004

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £80 +P&P)