THE APPLICATION OF PATTERN-RECOGNITION METHODS FOR FAST ANALYSIS OF THE DYNAMIC STABILITY OF ELECTRICAL-POWER SYSTEMS

Fast methods for estimating the state of electrical power system trans ients for severe disturbances are proposed. The methods are based on p attern recognition theory and expert knowledge, and they can handle st ochastic and incomplete data in conditions of severe time limitations. A method for constructing an expert system for pattern recognition ha s been considered A recognized electrical power system operation condi tion is represented by a point on a personal computer monitor. During training the expert system operates as an intellectual partner of the electrical power system dispatcher, helping him to recognize an emerge ncy situation and take the necessary steps to correct it. The applicat ion of pattern recognition methods and an expert system for transient stability evaluation is shown, with the Vietnam southern electrical po wer system taken as an example.