|
二维码（扫一下试试看！） |
| Data-driven Optimal Control for Pulp Washing Process |
| |
| DOI：10.11981/j.issn.1000-6842.2018.04.44 |
| Key Words:pulp washing process; data-driven; neural network; dynamic modeling; operational-pattern |
| Fund Project:陕西省重点科技创新团队计划项目（2014KCT-15）。 |
| Author Name | Affiliation | | SHAN Wen-juan1，* | 1.College of Light Industry Science and Engineering， Shaanxi University of Science & Technology， Xi’an， Shaanxi Province， 710021 | | TANG Wei2 | 2.College of Electrical and Information Engineering， Shaanxi University of Science & Technology， Xi’an， Shaanxi Province， 710021 | | LIU Bing3 | 3. Huawei Technology Co.， Ltd.， Xi’an， Shaanxi Province， 710077 |
|
| Hits: 4109 |
| Download times: 1122 |
| Abstract: |
| Considering the difficulties of modeling， online-measurement of residual soda and Baume degree indexes， and optimal control in pulp washing process， a data-driven operational-pattern optimization method for pulp washing process was proposed in this paper. Firstly， it described the basic concepts about data-driven operational-pattern for pulp washing process. Secondly， it established the data-driven prediction models of residual soda， Baume degree and operation performance integrated evaluation by PCA-BP neural network and multivariate logistic regression. Operational-pattern database was conducted based on large number of industrial operation data and the optimized operation patterns in washing process by the operation performance evaluation model. Thirdly， under the target of high production， low cost and low water consumption， the optimal operational-pattern was obtained by ant colony optimization algorithm from the optimized operational-pattern database. The practical results showed this method was effective in pulp washing process． |
| View Full Text View/Add Comment Download reader HTML |
