Optimization Of SAG Control Rules Based On Deep Learning Clustering Algorithm

With the rapid development of society and the advent of the era of big data,deep learning has become a hot spot in the field of artificial intelligence and machine learning.Unsupervised feature learning has shown great potential.It uses deep structures to mine deep structures and their hidden features.Clustering technology is also hot in the field of deep learning,and it is also an unsupervised learning method,which has attracted a large number of scholars to study in depth.Grinding is a key step in the selection of high-quality ore in the concentrator.The semi-automatic grinding machine is an important component of the SABC process.The operation mode mainly relies on the experience of senior technical workers to make judgments.If the detection is not in place and the judgment is not timely,it is easy to cause production problems and safety accidents.Analyze the experimental data by applying deep learning models and improved FCM(fuzzy clustering)clustering algorithm.The experimental data comes from the operation data of a mine in April 2016.The collected data period is subdivided into several small calculation periods.The data types include 126 sets of variables such as semi-automatic mill power,grinding sound,axial pressure,and grinding concentration.After cluster analysis,based on the comparison of the experience of senior technical workers on site,find out the relationship between power,grinding sound,axial pressure,grinding concentration,and ore amount during the operation of the semi-automatic grinding machine,guide the field operation,and form an expert Control rules,instead of manual control,promote the operation of semi-automatic mills in the optimal operating state,improve the economic efficiency of the concentrator,bring good profits to the enterprise,and promote the development of the mining industry.