| Title |
A Study on the improvement of Chemicals Dosing Performance using Neural network in a Purification Plant |
| Authors |
류승기 ; 최도혁 ; 홍규장 ; 문학룡 ; 한태환 ; 유정웅 |
| Abstract |
In general, the water process facilities include the purification plant, the waste water plant and the process of this purification plant is consisted of the intake, coagulation, settling, filtration, disinfection. The coagulation is very important in filtration processing plant and is very related to process of turbidity. The coagulation to the turbidity is, however, not yet to be clarified and the amount of coagulant can not be easily calculated. Moreover the coagulant dosing amount has to be decided adaptively according to the qualities of the raw water. So, the automation of chemicals dosing process and the supervisory system are needed to improve the performance of facilities. In this paper, a neural network is employed to model the coagulation to the turbidity of the treated water and the historical jar-test data are used to train the neural network. And also, an automation system to support the coagulant dosing process using the neural network was implemented and was shown by the field test. This automation system for the operator support system was constructed the environment to supervise the state and management of facilities for the maintenance. |