Sustainable Mining and Metallurgy

To address the frequent short circuit faults caused by the shorting of anode and cathode plates during the copper electrolysis process, which leads to significant energy loss, this study selected the electrolytic cell voltage signal as the object of analysis. By deeply analyzing the changes in the voltage signal of the electrolytic cell before and after the occurrence of short circuits, a short circuit fault diagnosis method combining Local Mean Decomposition (LMD) and Particle Swarm Optimization Extreme Learning Machine (PSO-ELM) was proposed. First, the Local Mean Decomposition (LMD) technique was used to decompose the original signal into several pure amplitude modulation frequency modulation components (PF), and the relative energy and total energy of each component were calculated, with the first three PF components selected as feature values. To overcome the limitation of the Extreme Learning Machine (ELM) requiring a large number of hidden layer nodes, this study employed the Particle Swarm Optimization (PSO) algorithm for parameter optimization. Subsequently, the extracted feature values were input into the optimized PSO-ELM model to achieve identification of short circuit faults. Experimental verification using field collected data showed that the accuracy of the Extreme Learning Machine (ELM) model combined with Local Mean Decomposition (LMD) and Particle Swarm Optimization (PSO) in the diagnosis of short circuit faults in the electrolytic cell can reach 91.09%, an increase of 6.98% compared to the single ELM diagnostic model and with good stability. Therefore, this model has the potential to be applied in industrial production for short circuit fault identification.