![]() | Only 14 pages are availabe for public view |
Abstract Methods of weather prediction using numerical and statistical methods have not eliminated completely the human factor. Human sK i I I is still needed in -forecasting. Expensive and rare human experience can be coded in computer programs that are referred to as Knowledge - based systems. Knowledge-based systems have been used in problem areas difficult for algorithmic solutions, and that depend mainly on human experience. ..-^- The objective of this worK is to build a Knowledge-based system for meso-scale weather forecasting for Alexandria region. For many local forecasters the only available tools may be synoptic charts , and conventional meters, such as thermometers, barometers, and anemometers. The system is designed to assist this type of forecasters. ’ Knowledge is extracted by investigating actual situations, from surface pressure synoptic charts or recorded data, or from books. Conditions for the occurence of certain weather phenomena are put in the form of IF-THEN rules, which are corrected and revised by a meteorology |