Apparatus for generating a water treatment model simulating an artificial neural network-based water treatment plant and method therefor

• Main Authors: KIM HYUN SIK, KIM YEONG HYEOK, YOO JUN WOO, PARK JUNG WON, NA SANG GUN, LEE YOUNG GEUN
• Format: Patent
• Language: English; Korean
• Published: 14.07.2023
• Subjects: CALCULATING; CHEMISTRY; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; CONTROL OR REGULATING SYSTEMS IN GENERAL; CONTROLLING; COUNTING; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; METALLURGY; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS ORELEMENTS; PERFORMING OPERATIONS; PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL; PHYSICS; REGULATING; SEPARATION; TRANSPORTING; TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE

A device for generating a water treating model comprises: an auto-modeling treating part which sets an injection amount of a chemical product to be injected into a water treating plant and the condition of influent flowing into the water treating plant as input data of a water treating model, and sets the condition of the treated water treated by the water treating plant as output data of the water treating model; a training data collection part which extracts training data including data for training and data for verification corresponding to the set input data and the set output data; and a training part which simulates the water treating plant through training using the training data to generate a water treating model predicting the condition of the treated water in accordance with the injection amount of a chemical product and the condition of the influent with respect to the water treating plant. According to the present invention, it is possible to optimize an injection amount of a chemical product in a water treating plant. 수처리모델을 생성하기 위한 장치는 수처리플랜트에 주입되는 화학제 주입량 및 상기 수처리플랜트에 유입되는 유입수의 상태를 수처리모델의 입력 데이터로 설정하고, 상기 수처리플랜트가 처리한 처리수의 상태를 수처리모델의 출력 데이터로 설정하는 자동모델링처리부와, 상기 설정된 입력 데이터 및 상기 설정된 출력 데이터에 상응하는 학습용 데이터 및 검증용 데이터를 포함하는 학습 데이터를 추출하는 학습데이터수집부와, 학습 데이터를 이용한 학습을 통해 상기 수처리플랜트를 모사하여 상기 수처리플랜트에 대한 상기 화학제 주입량 및 상기 유입수의 상태에 따라 상기 처리수의 상태를 예측하는 수처리모델을 생성하는 학습부를 포함한다.