Predictive Method of the Generator Output Based on the Learning of Performance Data in Power Plant

• Main Authors: YANG, HAC JIN, KIM, SEONG KUN
• Format: Patent
• Language: English; Korean
• Published: 19.05.2017
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORSOR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER; COUNTING; ELECTRICITY; GENERATION; MEASURING; MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES; NUCLEAR ENGINEERING; NUCLEAR PHYSICS; NUCLEAR POWER PLANT; PHYSICS; TESTING

Disclosed is a method for estimating generator output by power plant performance data learning. The method for estimating generator output for use in a power generation plant includes the steps of: measuring generator output, outputted from a controller for controlling a generator, through a turbine expansion line model based on data from a sensor coupled to the generator; verifying the data for the measured generator output through an input data verifying model; and comparing generator output data calculated through the previously set learning model with the previously verified generator output data. The present invention increases the reliability by improving the correction performance estimation model of a conventional technology. 발전플랜트 성능데이터 학습에 의한 발전기 출력 추정 방법이 개시된다. 발전플랜트에 사용되는 발전기 출력 추정 방법은, 발전기를 제어하는 제어기에서 출력된, 발전기에 결합된 센서로부터의 데이터에 기초하여 터빈 팽창선 모델을 통해 발전기 출력을 측정하고, 측정된 발전기 출력에 대한 데이터를 입력 데이터 검증 모델을 통해 검증하고, 미리 설정된 학습 모델을 통해 계산된 발전기 출력 계산 데이터와 앞서 검증된 발전기 출력 데이터를 비교하여 현재 발전기의 추정 출력을 예측한다.