Due to the fast material reaction in zinc hydrometallurgy, the traditional national standard photometric method cannot capture the characteristic information of target substances in real time. Herein, a nitrogen protection device is built based on ultraviolet spectrophotometry, supplemented by a programmable logic controller (PLC), to form an automatic control system for the direct detection of target substances (SO₄²⁻, Pb²⁺ and S²⁻) in zinc hydrometallurgy. The baseline straightness comparison results show that the nitrogen atmosphere can effectively improve the stability of the instrument. Furthermore, the detection sensitivity of SO₄²⁻, Pb²⁺ and S²⁻ under the nitrogen atmosphere is higher than that of the air atmosphere, manifesting in sensitivity increases of 16.23%, 18.05% and 17.91%, respectively. Additionally, devices based on PLC systems show advantages over manual control both in states feedback and information backtrack. Moreover, the regulation time and nitrogen consumption during the regulation process are reduced by 80% and 75%, respectively, which effectively reduces the test cost and improves the equipment utilization rate (from four cycles per day to six cycles per day). The device can meet the requirements of different target substances and different process conditions by changing the electronic control parts and air source, so it has great application potential in the automatic direct measurement of target substances in zinc hydrometallurgy.

This study successfully achieved high-precision detection of the clean coal ash content in the coal froth flotation domain by integrating deep learning with the likelihood function. Methodologically, a novel data processing and prediction framework was established by combining a deep learning Keras neural network with the likelihood function from probability statistics. The SIFT algorithm was utilized to extract key feature points and descriptors from the images, and keypoint matching and mean-shift clustering algorithms were employed to accurately obtain information on foam motion trajectories and velocities. For parameter optimization, the maximum likelihood estimation was applied to find the optimal parameter estimates of the likelihood function, ensuring enhanced model accuracy. By incorporating the optimized likelihood function parameters into the Keras deep neural network, an efficient prediction model was constructed for the dosage of flotation reagents, froth velocity, and clean coal ash content. The model’s evaluation involved six performance metrics. The experimental results were highly significant, with R² at 0.99997%, RMSE at 0.04458%, MAE at 0.00170%, MAPE at 0.02329%, RRSE at 0.00994%, and MAAPE at 0.00067%.