复杂反应堆存在精细几何结构,传统手工方式建模难以保证计算模型的精度。本文发展了基于辅助面的复杂结构智能分解技术和样条面处理方法,能够将复杂精细模型自动精准地转换为蒙特卡罗计算模型。本文对基于SuperMC建立的精细的国际热核聚变实验堆(ITER)C-Lite模型和手工建立的Benchmark模型典型部件的体积进行对比,结果显示该方法使得ITER的建模精度提升了近10倍。
Traditional manual modeling method is difficult to ensure the accuracy of the calculation model in the case of refined geometry structure in complicated reactors. Two major methods including face shell shrinking based geometry decomposing algorithm and spline surface processing methods were developed in Super Monte Carlo Simulation Program for Nuclear and Radiation Process (SuperMC). With these methods, complex refined models can be accurately converted to Monte Carlo calculation models. The volumecomparison between ITER Benchmark (created by manual) and ITER C-Lite (created with the help of SuperMC) indicated that SuperMC increased modeling accuracy by 10 times for ITER models.