Energy Consumption And Quality Optimization Of Polysilicon Reduction Furnace Based On NICA

Polysilicon reduction furnace is the most important equipment and the most energy-consuming equipment in modified SIEMENS polysilicon production process.The reduction process directly affects the total production cost and production quality.At present,the competition of polysilicon industry is becoming more and more intense.It is an urgent task to improve the energy utilization efficiency and reduce the production cost in the reduction process.However,most of the existing researches at China and abroad focus on the single aspect of improving energy consumption or improving the product quality,and actual production mainly depends on the experience of the workers to adjust the process parameters,and mainly to ensure the quality.In order to improve the competitiveness of polysilicon enterprises,it is necessary to optimize the energy consumption of reduction furnaces by integrating energy consumption and quality in order to improve the competitiveness of polysilicon enterprises in the single direction of reducing energy consumption or improving product quality.In this thesis,the polysilicon reduction furnace as the research object,through the comprehensive analysis of its process and energy consumption,four process parameters affecting energy consumption and quality are determined.A multi-objective optimization model of process parameters,energy consumption per unit product and deposition quality was established,The optimum combination of process parameters is obtained by solving the model.The specific research contents are as follows:1.Combined with the process principle of reduction deposition and the concrete structure of reduction furnace,the ways of energy consumption are analyzed in detail.The idea of improving energy utilization efficiency by improving the deposition rate of silicon is put forward.The four process parameters affecting the deposition rate and deposition quality are analyzed,which are temperature of silicon rod,reaction pressure,inlet velocity and molar ratio of H₂.Finally,the energy consumption per unit product is used as the evaluation index of energy use efficiency in the reduction process,and the calculation formula is given.2.According to the analysis of the production process and the investigation data of the enterprises,the structure and production conditions of the reduction furnace are properly simplified,and the finite element simulation model of the polysilicon deposition process is established based on the numerical simulation method.The effects of silicon rod temperature,reaction pressure,inlet flow rate and H₂ molar ratio on deposition uniformity and energy consumption per unit product are studied.The data sources for the following multi-objective optimization model are also provided for the following multi-objective optimization model.3.A quality evaluation model for evaluating the deposition quality of polysilicon is proposed,and the ELM（Extreme Learning Machine）model for the reduction process parameters,energy consumption per unit product and deposition quality is obtained.According to the theory of multi-objective optimization,a multi-objective optimization model of energy consumption per unit product and deposition quality is established,and the Pareto optimal solution set is obtained by combining NICA（Novel immune clonal algorithm）.Finally,the polysilicon reduction furnace energy optimization system was designed to be used in production.