Water Treatment Controller Design Using Neural Network Technique
| Autor: | Hsin-Li Chen, 陳信利 |
|---|---|
| Rok vydání: | 2004 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 93 In the operation of water treatment, the relationship between the coagulant dosage and the water quality parameters is non-linear in historical record data, and the affecting factors are quite complex. The normal dosage of coagulation depends on the jar tests and then adjusted to field actual dosage by operator experience in most of treatment plants. The conventional operation method can hardly in time to adjust to the proper dosage. PID controller has been used widely in the industry. In this paper, we adopt a wavelet neural network self-tuning PID controller, which possesses the capability of fast response and self-learning ability. This controller will be applied to the water treatment plant. Prediction can eliminate the human mistake, save tedious labor time, and reduce chemical expenses. Therefore an economic operation and efficiency result can be reached in the operation of water treatment plant. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|