Automated Calibration and Use of Stream‐Quality Simulation Model

Abstract

Simulation models have been used successfully in many engineering applications, and their use has expanded and matured in recent years in concert with the evolution of new, more powerful computers. Among the more promising enhancements of simulation methods has been the incorporation of artificial intelligence, or expert systems technology. This integration is particularly beneficial when focused on simulation models that are complex and difficult to use or calibrate due to special expertise required of the user, and/or the large investment of time and money involved. The purpose of the paper is to demonstrate the use of expert system technology in this context. The focus of the paper is on the development of an expert system to aid in the calibration process and subsequent use of a stream‐quality simulation model. An expert system shell (in this case, KES) is used to facilitate the fast and efficient development and testing of the expert system. A biochemical oxygen demand‐dissolved‐oxygen model, graphics software, and a program to perform model calibration are all linked externally to the shell. The user of the expert system is queried for field data necessary for calibration of the stream‐quality model, and is then provided with advice and instructions for performing the calibration and using the simulation model.