Fuzzy Model Identification and Control System Design for Coagulation Chemical Dosing of Potable Water

• Main Authors: Pei-Ling Hou, 侯佩伶
• Other Authors: 陳誠亮
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/88493389866285571402

碩士 === 國立臺灣大學 === 化學工程學研究所 === 93 === In Taiwan, which is known for numerous rivers and lakes, the raw water required for water purification plants comes mainly from the surface of rivers or creeks. The raw water is, therefore, significantly subject to the impact of a capricious climate. Especially during torrential storms or typhoons, sandy soil washed out by storms and muddy raw water brought by the floodwater released by reservoirs tremendously challenges the competence of water purification plants in the capacity to treat the water. The process used to remove pellet substances out of water normally involves three major procedures: (1) Coagulation; (2) Flocculation; and (3) Segregation of solids from liquids. These three procedures must be completed under optimal operating conditions. Among these optimal conditions, the performance of the chemical control in the coagulation process represents the most influential factor regarding the performance of the water purification plants. The present study accumulates the figures regarding chemicals added in the Changhsing Water Purification Plant of the Taipei Water Department, which has been deemed a very successful example, during 2001~2004 to come into the regression equation for the feedforward control, in automatic chemical addition forecast mode. Subsequently, the operators adjusted the strategy of chemical additions in line with the turbidity trend and pH value of the sedimentary water and filtered water to set up fuzzy feedback control rules. In the study, the chemical application was tested at the inlet of the Chingtan Weir and Water Separation Well at the Changhsing Plant to monitor different ratios so as to look into the problems of the uneven application of chemicals at the wells of Changhsing Plant and to work out proposals for corrective action. The study further assessed the optimal ratio of chemical application in the two phases to work out the chemical application ratios for the mold of feedforward control. Senior operators adjusted the application of chemicals based on the different trend of water quality at more favorable timings so as to assure the stable quality of the water discharged from the systems. In the fuzzy control strategy in the automatic application of the chemicals, the study suggests that timely countermeasures should be strengthened in the parameters. Through verification of on-the-spot operation, it proves that the data of the on-the-spot operation significantly conforms to the molds concerned. In the future, with water purification plant processing procedures remaining unchanged to go in line with the existent unremitting water quality monitoring equipment, it is advisable to set up a set of comprehensive and effective online programmable automatic systems to control the application of chemicals. Such efforts should virtually upgrade the efficiency of the application of chemicals for water purification plants.