基于并行技术的人体外照射实时精确剂量评估方法

为了精确评估核辐射环境下人体及器官受照剂量,FDS团队基于中国高精度辐射虚拟人模型Rad-HUMAN发展了体素级人体外照射剂量评估方法。但由于计算量巨大,现有计算流程和计算方法难以满足虚拟仿真对剂量评估实时性的要求。为了解决这一问题,本文借助OpenMP和MPI并行计算技术,发展了基于并行计算的体素级人体外照射剂量评估方法,实现了百万量级体素外照射剂量实时计算。以加速器驱动铅铋冷却反应堆堆顶包容小室内维修作业人体受照剂量评估为例的测试结果表明,该方法满足了体素级剂量实时评估计算的需求,对考虑到器官剂量限制的人体外照射剂量实时精确评估有重要意义。     

基于阵列式吸收发光CT法的辐射场实时测量系统的研究

分析了已有的各种核辐射剂量场测量方法的基础上, 利用闪烁体吸收发光特点并结合计算机实时成像处理技术,提出了对剂量场分布进行实时成像测量的新方法——阵列式吸收发光ＣＴ 法。研制出闪烁光纤阵列构成的探测器及其伺服控制系统, 用高灵敏度的电荷耦合器件（ＣＣＤ）拾取探测器产生的微弱闪烁光信号,并采用定点采集的方法对视频信号进行数据的快速采集。在图像重建方法上, 提出了迭代滤波反投影重建方法和利用非完全投影进行数据修复,对获取的投影数据进行变换和处理,并通过选择滤波函数及其适当的参数, 可获得最佳滤波效果,以重建剂量场的二维场分布,从而建立了剂量场的实时、高精度的成像测量系统。     

海上核动力平台核应急辐射场快速反演

提出了一种海上核应急三维辐射场快速反演方法。利用反距离权重方法,基于测量船上预安装的探测器定时回传的辐射监测值,不断迭代优化反演三维空间辐射场,实现真实场景情况下的光子剂量场与核素活度场耦合。结果表明,该方法能快速反演三维辐射场,满足核应急救援时间紧迫的需要,为核动力舰船等核设施应急提供有力的剂量评估与辐射场反演工具。     

基于最小包围盒策略的人体辐射剂量评估方法研究

在核设施停堆维修过程中,需要对维修工人受到的辐射剂量进行预评估。传统的评估方法是将人体简化为一个质点,无法精确评估不同器官的受照剂量。本文发展了一种基于包围盒策略的人体辐射剂量评估方法,并成功应用到国际热核聚变实验堆ITER极向场线圈维修工作人员辐射剂量评估中,结果表明,该方法不仅能提供更为精确的人体剂量评估结果,还能提供人体器官受照剂量,对考虑到器官剂量限值的工人作业方案制定有重要意义。     

矩阵模型光子束剂量算法的改进

改进了矩阵模型剂量算法,建立了射野矩阵、质心插值公式,提高了计算效率。该方法与AAPM 55#测试结果及直接实验测量值进行比较和检验,结果表明,计算速度快,适合作为放射治疗计划系统的快速剂量算法。     

电子个人剂量计的进展及其在实际使用中应注意的问题

电子个人剂量计可给出实时的剂量和剂量率数据,可报警,监测下限也较其它类型剂量计低很多,已得到广泛应用。本文简要回顾了外照射个人剂量监测中的各种方法,重点介绍评述了电子个人剂量计的进展,详细介绍了多种电子个人剂量计及其性能比较,指出了电子个人剂量计在现场应用时应注意的问题。     

X光透射剂量中模体散射线的物理特性研究

放射治疗的X光透射剂量中含有不同强度的模体散射线,影响其临床应用.本文基于实验方法在不同条件下测量6 MV X线射野中心轴及离轴位置的透射剂量分布,从中提取原射线剂量,计算得到相应的散射线剂量,分析透射剂量中原射线和模体散射线的大小,研究不同条件下透射剂量中模体散射线分布的物理特性.     

核电站三维剂量场评价系统的开发及应用

基于核电站营运单位和设计单位对三维剂量场评价系统的需求分析,得出并解决了三个核心技术问题,即:三维剂量场计算、显示和基于三维剂量场的剂量优化问题,进而开发了核电站三维剂量场评价系统(RPOS)。基于在役核电厂控制区内典型区域的系统测试结果表明:三维剂量场评价系统能够较准确地模拟所测试的核电站典型区域的三维剂量场,并实现测试区域内三维剂量场的虚拟显示以及核电站内工作人员操作的剂量优化。     

基于生物启发神经网络的核辐射场区全覆盖路径规划

核辐射场区全覆盖路径规划对于辐射环境下区域作业者的辐射安全有重要意义。本研究基于生物启发神经网络算法,提出一种进行辐射剂量最优控制的全覆盖路径规划算法。首先,利用福岛核电站部分地形以及蒙特卡罗粒子输运程序分别构建模拟核辐射场区的障碍物分布和辐射剂量场,然后,采用Python语言进行算法仿真试验,模拟核辐射场区的每一个栅格定义为一个神经元,建立起生物启发神经网络,将栅格剂量率与神经元活性耦合实现路径规划的辐射剂量最优控制,分别采用单个、4个和8个移动单元进行仿真试验。结果表明：单个移动单元的规划路径在实现100%覆盖率,4%覆盖重复率的同时,能够优先覆盖低剂量区,延后覆盖高剂量区,实现了过程剂量和累积剂量的最优控制。为提高全覆盖的时间效率和获得更低的单体累积剂量,对算法进行多单元协同搜索的改进,结果表明：4单元和8单元仿真的覆盖重复率分别为5.72%和6.29%,1单元、4单元和8单元仿真完成全覆盖时间分别为30 min、9 min和4 min,时间效率成倍提高;最大单体累积剂量分别为4.11×10^-3 mSv、1.28×10^-3 mSv和0....     

CT图像的体元大小对EGSnrc蒙特卡罗剂量计算的影响

目前的放射治疗计划系统采用经验或半经验常规剂量算法,计算得到的剂量分布不够精确,而蒙特卡罗剂量算法是一种精确的剂量计算方法,但其计算时间冗长问题最终影响临床的应用.本文基于一例病人头部电子计算机体层成像(Computer Tomography,CT)数据,应用EGSnrc(Electron Gamma Shower NRC,EGSnrc)蒙特卡罗程序,研究不同体元大小对剂量分布计算精度和计算时间等的影响.     

量子粒子群融合算法在放射源探测标定中的应用

为有效进行辐射剂量探测,设计了有源在线监控模块。针对监控模块辐射测量值标定这一难题,提出了基于量子粒子群的融合算法实现对测量值与剂量值之间的高精度逼近。使用聚类和数据融合算法实现了拟合数据的自主最佳分段,避免了分段选择的主观性。仿真结果表明,融合算法能有效减少计数率较小时拟合函数的误差,提高各数据段测量值与剂量率之间的拟合精度,与整体拟合函数相比,进一步提高了拟合的精度。     

混合遗传算法在核事故源项反演中的应用

针对国内外普遍关注的核事故源项反演与事故后果评价的问题,运用遗传算法-单纯形法结合烟团模型实现了对源强的反算和释放点位置的快速定位。遗传算法-单纯形法与遗传算法-模式搜索法、遗传算法和单纯形法等3种算法的比较结果表明：遗传算法-单纯形法结合了遗传算法和单纯形法两种算法的优势,也弥补了各自算法的缺陷,其反算值可与期望值较精确符合;扩散模式模块、GA模块和NM模块3者可简单直接的结合,结合所需编写的代码较少,通用性广;GA模块和NM模块的计算花费较少,适用于核电厂对源项的快速估计。     

Development of real-time plasma analysis and control algorithms for the TCV tokamak using Simulink

One of the key features of the new digital plasma control system installed on the TCV tokamak is the possibility to rapidly design, test and deploy real-time algorithms. With this flexibility the new control system has been used for a large number of new experiments which exploit TCV's powerful actuators consisting of 16 individually controllable poloidal field coils and 7 real-time steerable electron cyclotron (EC) launchers. The system has been used for various applications, ranging from event-based real-time MHD control to real-time current diffusion simulations. These advances have propelled real-time control to one of the cornerstones of the TCV experimental program. Use of the Simulink graphical programming language to directly program the control system has greatly facilitated algorithm development and allowed a multitude of different algorithms to be deployed in a short time. This paper will give an overview of the developed algorithms and their application in physics experiments.     

利用混合态分解算法提高重叠关联成像的质量

X射线重叠关联成像技术所采用的常规相位恢复迭代算法均把入射光视为完全相干的单色光(即单态入射光),把样品视为稳定平衡的(即单态样品),把实验过程中成像系统内所有光学元器件的响应视为固定不变的。然而在实际实验中,上述三个条件很难得到同时保证。在非理想的实验条件下,任何单态条件发生变化或者出现多重态混合,其造成的退相干效应均会使衍射图样变得模糊,严重降低利用单态相位恢复迭代算法获得的重建图像质量。为了消除各种来源的混合态效应的负面影响,本文利用多模式拓展重叠关联迭代引擎算法(Mm-e PIE)和子像素上采样拓展重叠关联迭代引擎算法(Us-e PIE)这两种不同的混合态分解算法对实验数据进行处理。重建结果表明,不同来源的混合态效应(退相干效应)的分解结果在重建探针函数和重建样品函数中均得到了体现,两种混合态分解算法均能够显著改善重建图像的质量。其中多模式算法Mm-e PIE在消除入射探针的非相干效应方面更胜一筹。     

泊松克里金算法在核辐射场重构中的应用研究

为验证泊松克里金法在核辐射场重构中的应用效果,本文对泊松克里金算法在辐射场剂量率插值计算中的计算原理进行了研究,结合代理模型对泊松克里金算法进行了求解。通过SuperMC软件建立虚拟辐射场,对泊松克里金算法的参数影响进行了分析,筛选出二次多项式基函数和GAUSS误差相关模型。采用6150AD-b闪烁体探测器对实际放射源进行测量,并利用实测辐射场剂量率数据和福岛周边空间剂量率数据对泊松克里金算法在核辐射场重构中的可行性进行了验证。结果表明：小范围简单辐射场中,样本数据量为网格点数的1/10时,平均相对误差在10%以内;大范围辐射场中,辐射热点区域略有偏移,平均相对误差为141.69%。以上结果表明,该方法在快速、低成本解决未知放射源辐射场的重构问题中有一定的潜力。     

卡尔曼滤波反演核设施核事故中核素释放率的研究

为能在核设施发生核事故时快速连续反演核素释放率,本文结合高斯多烟团大气扩散模型,模拟固定区域的连续监测数据,设计并实现了核设施核事故核素释放率的卡尔曼滤波实时跟踪反演。研究结果表明：与高斯多烟团大气扩散模型结合的卡尔曼滤波器,在约10次滤波后,跟踪到虚设的稳定、线性及非线性变化的释放率真值,反演值标准差随真值的增大而增大;在扩展卡尔曼滤波反演释放高度时,由于截断误差过大,滤波结果不收敛。利用环境监测数据,通过与高斯多烟团大气扩散模型结合的卡尔曼滤波器可用于固定高度和位置、短时连续排放的核事故核素释放率参数反演,是核设施核事故应急可选择的源项反演手段。     

减小辐射场剂量率预测不确定度的数据同化研究

为提高核事故早期辐射场剂量率评估的准确性和可靠度,提出一种基于数据同化理论的评价方法。根据数据同化理论,定义辐射场状态空间,建立了适合辐射场剂量率预测模型的状态转移和误差协方差矩阵。利用同化算法,综合考虑核事故辐射场剂量率预测模型与实测数据,实现辐射场剂量率的最优化表达。在Matlab软件平台上,运用数值模拟实验和福岛实测数据,对该方法进行了双重验证。实验结果表明,数值模拟实验条件下,当放射性物质空气释放率高估1个数量级时,同化后,辐射场剂量率相对均方根误差从1个数量级降低至50%左右。利用福岛核事故监测数据对同化系统进行验证,同化后的结果与实际测量结果相近,相对均方根误差在20%左右。以上结果表明,通过合理运用数据同化方法,可有效减小辐射场剂量率预测的不确定度。     

Particle dispersion modeling in ventilated room using artificial neural network

Due to insufficiency of a platform based on experimental results for numerical simulation validation using computational fluid dynamic method (CFD) for different geometries and conditions,in this paper we propose a modeling approach based on the artificial neural network (ANN) to describe spatial distribution of the particles concentration in an indoor environment.This study was performed for a stationary flow regime.The database used to build the ANN model was deducted from bibliography literature and composed by 261 points of experimental measurement.Multilayer perceptron-type neural network (MLP-ANN) model was developed to map the relation between the input variables and the outputs.Several training algorithms were tested to give a choice of the Fletcher conjugate gradient algorithm (TrainCgf).The predictive ability of the results determined by simulation of the ANN model was compared with the results simulated by the CFD approach.The developed neural network was beneficial and easy to predict the particle dispersion curves compared to CFD model.The average absolute error given by the ANN model does not reach 5％ against 18％ by the Lagrangian model and 28％ by the Euler drift-flux model of the CFD approach.     

Improvement of accident dose consequences simulation software for nuclear emergency response applications

This study integrated the nuclear power plant simulation software, PCTRAN, with an atmospheric diffusion model to efficiently evaluate a nuclear power plant accident and its off-site dose consequences. PCTRAN, with its user-friendly interface, provides a fast simulation scheme that can simulate many kinds of nuclear power plant accidents. Once accident initiation events are activated in the software, the plant parameters are calculated and displayed via animations on the user interface. Based on the simulated plant conditions, the radioactive materials considered in the software may be released from the plant to the environment. In this study, a dispersion algorithm, including a modified atmospheric diffusion model and its programming method, is proposed such that PCTRAN satisfies the application requirements to be used to plan nuclear emergency responses. First, the modified atmospheric diffusion model handles the variations of meteorological conditions (wind direction, wind velocity, and stability category) during a nuclear power plant accident simulation. Furthermore, the proposed programming method promotes calculation capability and efficiency by reducing the computational burden. For demonstration purposes, a postulated accident event was simulated for the Maanshan Nuclear Power Plant in Taiwan. The overall accident evolution, whole plant response, and off-site dose consequences could be predicted much earlier than what actually occurs. The thyroid and whole body dose rates (and their accumulations) as a function of accident time are displayed on the map within the emergency planning zone (EPZ). The influence of the accident on the off-site area can thus be estimated earlier, and the emergency classification can be determined by referring to the emergency action levels (EALs) for a quick nuclear emergency response.     

Assessment of fusion facility dose rate map using mesh adaptivity enhancements of hybrid Monte Carlo/deterministic techniques

Three mesh adaptivity algorithms were developed to facilitate and expedite the use of the CADIS and FW-CADIS hybrid Monte Carlo/deterministic techniques in accurate full-scale neutronics simulations of fusion energy systems with immense sizes and complicated geometries. First, a macromaterial approach enhances the fidelity of the deterministic models without changing the mesh. Second, a deterministic mesh refinement algorithm generates meshes that capture as much geometric detail as possible without exceeding a specified maximum number of mesh elements. Finally, a weight window coarsening algorithm decouples the weight window mesh and energy bins from the mesh and energy group structure of the deterministic calculations in order to remove the memory constraint of the weight window map from the deterministic mesh resolution. The three algorithms were used to enhance an FW-CADIS calculation of the prompt dose rate throughout the ITER experimental facility and resulted in a 23.3% increase in the number of mesh tally elements in which the dose rates were calculated in a 10-day Monte Carlo calculation. Additionally, because of the significant increase in the efficiency of FW-CADIS simulations, the three algorithms enabled this difficult calculation to be accurately solved on a regular computer cluster, eliminating the need for a world-class super computer.