• 제어로봇시스템학회논문지
• /
• 제3권6호
• /
• pp.652-658
• /
• 1997

Coagulant feeding control is very important in the water treatment process. Coagulant feeding is related to the raw water quality such as turbidity, alkalinity, water temperature, pH and so on. However, since the process of chemical reaction has not been clarified so far, coagulant dosing rate has been decided by jar-test. In order to overcome the difficulty mentioned above, Fuzzy Neural Network to fuse fuzzy logic and neural network was proposed, and the scheme was applied to the automatic determination of coagulant dosing rate. This algorithm can automatically identify the if-then rules, tune the membership functions by utilizing expert's experimental data. The proposed scheme is evaluated by computer simulation and interfaced with coagulant feeder operated by magnetic flowmeter, control valve and PLC. It is shown that coagulant feeding according to real time sensing of water quality is very effective.

• 제어로봇시스템학회논문지
• /
• 제9권1호
• /
• pp.57-62
• /
• 2003

It is difficult to determine the feeding rate of coagulant in the water treatment plant, due to nonlinearity, multivariables and slow response characteristics etc. To deal with this difficulty, the genetic-fuzzy system genetic-equation system and the neural network system were used in determining the feeding rate of the coagulant. Fuzzy system and neural network system are excellently robust in multivariables and nonlinear problems. but fuzzy system is difficult to construct the fuzzy parameter such as the rule table and the membership function. Therefore we made the genetic-fuzzy system by the fusion of genetic algorithms and fuzzy system, and also made the feeding rate equation by genetic algorithms. To train fuzzy system, equation parameter and neural network system, the actual operation data of the water treatment plant was used. We determined optimized feeding rates of coagulant by the fuzzy system, the equation and the neural network and also compared them with the feeding rates of the actual operation data.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.123.2-123
• /
• 2001

It is difficult to determine the feeding rate of coagulant in the water treatment plant, due to nonlinearity, multivariables and slow response characteristics, etc. To deal with this difficulty, the neuro-fuzzy system and the genetic-fuzzy system were used in determining the feeding rate of the coagulant. The fuzzy system is excellently robust in multi-variables and nonlinear problems. Therefore it uses basic algorithm, but it is difficult to construct of the fuzzy parameter such as the rule table and the membership function, Therefore we made the neuro-fuzzy system and the genetic-fuzzy system with the fusion of learning algorithms and compared the performance of the two fuzzy systems. To apply these algorithms, we made the rule table, membership function from the actual operation data of the water treatment plant. We determined optimized feeding rate of coagulant using the fuzzy operation, and also compared ...

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.952-955
• /
• 1996

It is difficult to determine the feeding rate of coagulant in water treatment process, due to nonlinearity, multivariables and slow response characteristics etc. To deal with this difficulty, the fusion of genetic algorithms and fuzzy inference system was used in determining of feeding rate of coagulant. The genetic algorithms are excellently robust in complex operation problems, since it uses randomized operators and searches for the best chromosome without auxiliary information from a population consists of codings of parameter set. To apply this algorithms, we made the look up table and membership function from the actual operation data of water treatment process. We determined optimum dosages of coagulant (PAC, LAS etc.) by the fuzzy operation, and compared it with the feeding rate of the actual operation data.

• 제어로봇시스템학회논문지
• /
• 제5권7호
• /
• pp.876-882
• /
• 1999

It is difficult to determine the feeding rate of coagulant in the water treatment process, due to nonlinearity, multivariables and slow response characteristics, etc. To deal with this difficulty, the genetic-fuzzy system was used in determining the feeding rate of the coagulant. The genetic algorithms are excellently robust in complex optimization problems. Since it uses randomized operators and searches for the best chromosome without auxiliary informations from a population consists of codings of parameter set. To apply this algorithms, we made the lookup table and membership function from the actual operation data of the water treatment process. We determined optimum dosages of coagulant(LAS) by the fuzzy operation, and compared it with the feeding rate of the actual operation data.

• 제어로봇시스템학회논문지
• /
• 제10권11호
• /
• pp.1100-1106
• /
• 2004

In this report, a self-organizing fuzzy neural network is proposed to control chemical feeding, which is one of the most important problems in water treatment process. In the case of the learning according to raw water quality, the self-organizing fuzzy network, which can be driven by plant operator, is very effective, Simulation results of the proposed method using the data of water treatment plant show good performance. This algorithm is included to chemical feeder, which is composed of PLC, magnetic flow-meter and control valve, so the intelligent control of chemical feeding is realized.

• 제어로봇시스템학회논문지
• /
• 제10권6호
• /
• pp.551-556
• /
• 2004

Coagulation process is one of the most important processes in water treatment procedures for stable and economical operation, and coagulant dosing of this process for most plants is generally determined by the jar test. However, this method does not only take a long time to analyze and get the result but also has difficulties in applying to automatic control. This paper shows the feasibility of applying neural network to control the coagulant dosing automatically in water treatment plant. To be specific, the predicted results of the neural network model is shown to be similar to that of jar test. The input variables for learning the neural network are turbidity, water temperature, pH, and alkalinity. Combining the neural network and SCD(Streaming Current Detector) for feedforward and feedback control of injecting coagulant, a rapid change of the raw water quality can be accommodated.