Self-adaptive regulation and control method for flotation dosing

Abstract

The invention relates to a flotation dosing self-adaptive regulation and control method which is characterized by comprising the following steps: collecting detection data D1 in a flotation process, establishing a comprehensive database, fusing the detection data D1 and historical data to form to-be-integrated data D2 for integration, and constructing a GRNN neural network prediction model. Inputting training data, comparing the training data with the concentrate grade and the metal recovery rate in a historical database, and reasoning by adopting a fuzzy expert system and utilizing a Mamdani fuzzy inference engine to obtain an agent addition amount; taking the obtained agent addition amount and detection data D1 collected in real time as input sample data B of a GRNN neural network prediction model, and obtaining predicted concentrate grade and metal recovery rate data Y; s6, judging whether requirements are met or not, if the requirements are met, executing the current dosage, and if the requirements are not met, returning to the fuzzy expert system for reasoning again, and repeating the operations of S4, S5 and S6. The stability of the flotation concentrate grade is achieved, and the optimal metal recovery rate is obtained.

Description

technical field [0001] The invention relates to the technical field of flotation process optimization control technology, in particular to a flotation dosing self-adaptive regulation method. Background technique [0002] The grade of ore directly affects the economic benefits of iron and steel enterprises, and the flotation production process is the most effective production method to directly improve the grade of concentrate. The given of ore is an important factor affecting the grade of concentrate. At present, neural network prediction technology and flotation foam image recognition technology are mainly used at home and abroad to estimate the grade of concentrate, and according to the change trend of grade, a reasonable given of chemical flow is given. However, the current automatic flotation control only makes decisions based on the current situation, and cannot know whether it meets the site requirements. For example, the invention patent of patent number CN203299674U...

AI Technical Summary

Problems solved by technology

However, the current flotation automatic control only makes decisions based on the current situation, and cannot know whether it meets the requirements of the site

For example, the invention patent of patent No. CN203299674U is aimed at transforming the dosing control system in the flotation process, but it does not further verify the production indicators after dosing, and lacks accuracy; the invention patent of patent No. CN113117896A simply collects data, However, the problem of excessive scale when processing data cannot be solved.

However, in the flotation process, there are contradictions among ore properties, concentra

Images