DEVELOPMENT OF A KNOWLEDGE BASED SYSTEM FOR THE AIR-CONDITIONING OF A BUILDING IN INDIA

Abstract

Knowledge based systems are interactive computer programs that capture, encode the specific knowledge of a particular domain. They mimic the reasoning process of human beings to solve the problems of areas that require rule of thumb judgmental problem solving methods. In this thesis, an effort has been made to solve the engineering problems come across in air-conditioning system design. In the field of air conditioning there are certain problems which cannot be solved by purely algorithmic conventional software i.e. they require experience and judgement on the part of experts. For example selection of a system out of many alternatives available for air conditioning a building, depending upon the requirements, priority of requirements etc. Heat load to be handled by a system in summer has been selected. TR capacity of the air-conditioning system is one of the most important parameters in selecting a system for a particular application. Load calculation makes the air-conditioning system design to be more precise. But economic criteria enable the system designer to select the one which gives optimised system. The other alternatives available can be discarded on the basis of knowledge based system. Specifically the software in `C' language has been prepared for data made available from Indian handbooks. The 1800 line program has been prepared and tested by tracking the problem. In lieu of the volume of the work involved, the knowledge based system handles the load estimation and system selection problems. This software package will provide guide lines`for selection of equipment for air-conditioning systems in an interactive manner for any space located in metropolitans in India. Some data base have been prepared by using information from Indian sources, and the software makes use of these data bases. The program determines the TR requirements for a space and several alternatives can be tried.