A KNOWLEDGE-BASED SYSTEM FOR MIX DESIGN OF CONCRETE CONTAINING POZZOLANIC MATERIALS

摘要

Design of concrete mixtures is the process of selecting the most economical and practical proportions of all the ingredients to produce quality concrete. Because of the nature of the mix design process and all the heuristic knowledge that is associated with it, concrete mix design lends itself well to a knowledge-based application. This paper describes a prototype knowledge-based system called HPCMIX that can be used to load the experimental data, develop several statistical models to determine the responses, evaluate the predicted models, and predict the 28 days strength and slump. It can also be used to design HPC mix proportions that are subjected to specified constraints (e.g. strength and slump). It is capable of selecting proportions of mixing water, cement, supplementary cementitious materials, aggregates and superplasticizer, with consideration being given to the effects of air content as well as water contributed by superplasticizer and moisture conditions of aggregates. As most of knowledge-based systems, this system has explanation facilities, can be incrementally expanded, and has an easy to understand knowledge base. The system was tested on a sample project. The system's selection of the concrete proportions compared favourably with the expert's selection. The system is very useful for civil engineering students as well as practicing engineers.