混合辐射场测量中屏蔽设计的组合优化方法

在混合辐射场测量中，各测点的辐射环境和测量目标不同，同一种屏蔽方案不能很好地满足各测点的需要。有必要发展一种辐射屏蔽自动优化设计方法，实现屏蔽效果的定量评价，并针对不同情况进行屏蔽优化。根据代价利益分析方法提出了辐射屏蔽措施的优化评估方法；引入遗传算法建立了组合屏蔽自动设计方法。计算表明该评价方法与文献及MCNP模拟计算结果相符；在典型的裂变中子、裂变r混合辐射条件下，优化设计方案的辐射干扰相当于原方案的44％。组合屏蔽设计方法可以针对特定辐射环境和测点参数进行自动优化设计，对混合辐射场测量的方案设计和数据分析具有积极的作用。

A COMBINED OPTIMIZATION METHOD FOR RADIATION SHIELDING DESIGN IN MIXED RADIATION FIELD MEASUREMENTS

One single shield design was not suitable for .all measuring points with different local radiation characters and measurement objects in mixed-radiation field. A new method is needed for quantitative evaluation and automatically optimizing design of radiation shielding to adapt it to various conditions. Based on cost-benefit principle, we propose an evaluation method for radiation shielding, and implement a combined optimization method based on genetic algorithm. Experiments show that the evaluation method is consistent with results given by various references and MCNP; in typical condition, influence of noise radiation from optimized design is 44 % of the traditional design. The combined optimization method is able to automatically optimize design with certain radiation characters and measuring point parameters, that is useful for project design and data analysis.