Thickener underflow concentration prediction method based on DAJYPLS algorithm

Abstract

The invention provides a thickener underflow concentration prediction method based on a DAJYPLS algorithm, and relates to the technical field of hydrometallurgy. The method comprises the following steps: step 1, determining a data set required for establishing a model; 2, performing standardized processing on data in the data set; step 3, establishing a DAJYPLS prediction model based on a JYPLS algorithm; step 4, giving a new sample xnew for the target domain. According to the method, an accurate prediction model can still be established when a small amount of on-site sampling data exists, so that an operator is assisted in control, safe and stable operation of the thickening and dewatering process is guaranteed, comprehensive economic benefits are improved, and meanwhile, the failure rate of the filter press is reduced.

Description

technical field [0001] The invention relates to the technical field of hydrometallurgy, in particular to a thickener underflow concentration prediction method based on a domain adaptive joint output partial least squares (DAJYPLS) algorithm. Background technique [0002] With the rapid development of the country, the continuous advancement of industrialization and urbanization, and the continuous development and utilization of mineral resources, high-grade ores are gradually decreasing, and metal resources are increasingly depleted. The industry has to increasingly face the problem of recovering mineral resources from poor, fine and miscellaneous ores, and the supply of mineral resources is facing unprecedented pressure. How to economically and efficiently utilize low-grade non-ferrous metal mineral resources to produce metal products is of great significance to the sustainable development of our country. The hydrometallurgical process has the advantages of processing low-g...

AI Technical Summary

Problems solved by technology

The above methods are all based on sufficient data samples. For a new batch production process that has just been put into production, due to the short running time of the production process, sufficient process data is usually not available.

If a modeling method similar to the above is used, the established forecasting model often has low forecasting accuracy and cannot meet production needs

According to the traditional method, it is necessary to redesign the experiment to obtain process data, which will consume a lot of manpower and financial resources, and the modeling efficiency is low, which seriously slows down the speed of the new process to achieve production operation optimization, which is not conducive to real-time adjustment of production strategies and expansion of production. scale

Images