Modeling And Optimization Of Cobalt Oxalate Particle Size Distributions Based On KPLS Hybrid Model

In the thesis, the modeling and optimizing of cobalt oxalate particle size distributions (PSD) in a synthetic process is studied, which is a subproject of "863" National High-tech Research Projects—the development of the optimal control software system in hydrometallurgical process.With the rapid development of high-tech, the demands for cobalt powder PSD in hard alloy, petrochemicals, magnetic materials and batteries become higher and higher. Cobalt powder PSD relies on the stuff cobalt oxalate PSD to a large extent. At present, the control of cobalt oxalate PSD mainly relies on workers’experience. The process parameters are fluctuating. And meanwhile, cobalt PSD is roughly described by bulk density, it can not meet the precision requirements of industrial development.In the thesis, with deeply understanding of cobalt oxalate synthetic process and according to the practical production, the key parameter and operational variables in the synthetic process of cobalt oxalate are established. And a detailed analysis of the mechanism model of cobalt oxalate synthesis process is made.Due to the complexity of the crystallization process and the uncertainty of the crystallization parameters, the mechanism model is constructed based on some assumptions and simplification resulting in a low precision. In this thesis, a hybrid model of cobalt oxalate PSD is established. It is composed of a simplified mechanism model and a data compensation model. Simplified mechanism model describes the main dynamic characteristics of cobalt oxalate synthetic process. And a kernel partial least squares (KPLS) method is adopted to compensate the model errors. The simulation is conducted, the results indicate effectiveness of the hybrid model.For the shortcomings of the existing cobalt oxalate PSD optimization algorithm, some improvements are made on the real-coded adaptive genetic algorithm (RAGA) mainly in the following aspects:the netlike distribution of initial population creation, closed crossing avoidance and2/4survival competition for the cross results. The results from simulations verify that the improved RAGA inproves the optimization performance.Based on the works above, a prediction and optimization system for cobalt oxalate synthetic process is developed by C#and MATLAB mixed programme. This system platform combines Visual C#which has friendly visual programming interface characteristics and MATLAB which has powerful numerical analysis function. The main functions of the system is to predict cobalt oxalate PSD online according the operating conditions, and provides satisfactory operation guide...