|
| Active Learning Fuzzy Rule Extraction from Zinc Leaching Process for Decision-Making |
|
Received:October 18, 2024
Revised:December 25, 2024
Accepted:December 26, 2024
Published Online:April 19, 2025
|
| View Full Text View/Add Comment Download reader |
| DOI：doi：10.20237/j.issn.1007-7545.2025.05.004 |
| KeyWord:neutral leaching; pH control; active learning; SVM; rule extraction |
| Author | Institution |
| CHEN Yu |
杭州科技职业技术学院 |
| HUANG Yudong |
杭州科技职业技术学院 |
| LIU Xuebin |
中南大学 自动化学院 |
| LI Haisheng |
杭州科技职业技术学院 |
| LAO Jiafeng |
杭州科技职业技术学院 |
|
| Hits: 550 |
| Download times: 1450 |
| Abstract: |
| In the neutral leaching process of zinc hydrometallurgy, the stability of pH value directly affects the quality of the product. A fuzzy rule extraction strategy based on active learning classification was proposed to address the challenge of pH value stability control caused by solvent concentration fluctuations and large time delays in traditional manual control methods. The neutral leaching process was subdivided into multiple typical operating conditions by in-depth analysis of the process mechanism. Employing the approximate linear dependency method and active learning algorithm, the information samples representing typical working conditions were accurately screened from a large number of historical data. Then, the support vector machine algorithm was used to extract support vectors under different operating conditions and a fuzzy rule set was constructed. The results show that the qualified rate of pH value control under different working conditions is as high as 93.17%, and the variance is as low as 0.011, which fully verifies the effectiveness and practicability of the algorithm. The results provide a solid and effective support for pH stability control in neutral leaching process. |
| Close |