Aluminum Electrolysis Fault Diagnosis System

Aluminum electrolysis process is a nonlinear, coupling, time-varying and large time-delay process. The working current in the aluminum electrolysis process is usually in the tens or hundreds of thousands of amperes, the strong electric field, the strong magnetic field, strong thermal field mutual interference, so the environment is very bad. In the process of aluminum electrolysis, not only exist fault in electrolytic cell, but also exist fault that caused by the fault of motor which is the actuator of aluminum electrolysis. This thesis defined cell fault as the process fault of aluminum electrolysis, actuator fault defined system failure. In view of above situation, this paper proposed the aluminum electrolysis fault diagnosis system. The work arrangement as follows.First, this thesis expounds the significance of aluminum electrolysis fault diagnosis, summarizes the current situation of the development of the aluminum electrolysis fault diagnosis at home and abroad and the existing problems. Aiming at the problems in aluminum electrolysis fault diagnosis, this thesis describes the research purpose and research significance of the topic, put forward my own research method.Second, this thesis summarizes the working principle and production process of electrolytic aluminum. This thesis discusses several common process faults and system faults in the aluminum electrolysis process, and simply introduced these faults.Third, this thesis was established the process fault diagnosis model of the aluminum electrolysis. This thesis expounds the basic principle of neural network, analyzed the possibility and necessity of the application of neural network in fault diagnosis, discusses the selection of feature and process of the input data in the process of aluminium electrolytic. The training process of neural network is optimized by the leapfrog algorithm, the simulation results verified the feasibility of diagnosis model for process fault in aluminium electrolytic.Fourth, this thesis was established the system fault diagnosis model of the aluminum electrolysis. This thesis expounds the superiority of EMD algorithm that is adopted to the signal process of motor fault. It introduces the theory of wavelet analysis, the feature extraction of fault signals used by wavelet analysis, and the combination of wavelet analysis and neural network. The simulation experiments verified the feasibility of diagnosis model for system fault in aluminium electrolytic.Fifth, this thesis used the VB 6.0 as development platform, designed the software for aluminum electrolysis fault diagnosis and system monitoring. Summarizes the general idea and design scheme of the software, and introduced the realization of the function of each part of the process monitoring software in details.Sixth, this thesis summarized the fault diagnosis methods for aluminum electrolysis and done a further analysis. At last this thesis looking forward to the further application for fault diagnosis in aluminum electrolysis.