对 2 0 0 MW核供热堆装载模式 (燃料组件布置、可燃毒物棒根数和可燃毒物质量分数配置 )进行了优化。利用模拟退火算法和先进格林函数节块法进行多步燃耗优化计算。引入敏感性系数 ,并通过敏感性分析的方法决定优化参数 ,因此在单目标和多目标优化时均取得了明显的效果。对组件布置和可燃毒物质量分数的优化计算结果表明 ,在不改变原有的富集度和组件类型的前提下 ,与参考值相比 ,优化后的循环长度、功率峰因子和卸料燃耗均有明显的改善。该燃料管理方法不仅可用于低温堆而且也可以推广到压水堆 The loading mode (fuel assembly arrangement, number of combustible poisonous rods and flammable toxin mass fraction configuration) of 200 MW nuclear heating reactor was optimized. Multi-step optimization of fuel economy with simulated annealing algorithm and advanced Green’s function nodal method. The sensitivity coefficient is introduced and the optimization parameters are decided by the method of sensitivity analysis. Therefore, both single-objective and multi-objective optimization have achieved remarkable results. The results of the optimization of component layout and flammable toxin mass fraction show that the optimized cycle length, power peak factor and discharge burn-up, compared with the reference value, without changing the original enrichment and component type, Have significantly improved. This fuel management approach can be used not only for cryogenic reactors but also for pressurized water reactors