部署实物保护系统（Physical Protection System, PPS）是关键设施的重要防护措施。 实物保护系统如何评估关键设施在特定威胁下的防护性能并识别系统的漏洞，是目前 迫切需要解决的问题。针对定性评估中主观因素的影响、定量评估中三维空间中脆弱 路径搜索问题以及实物保护系统的脆弱性评估系统的设计开发展开研究。本文主要工 作包括：

（1）针对现有实物保护系统脆弱性定性评估方法安全指标构建不完善、评估结果 受主观因素影响大、评估过程存在模糊和不确定性的问题，提出了基于二维云模型的 实物保护系统脆弱性定性评估方法（Cloud Assessment of Physical Protection System, CAPPS）。CAPPS 的步骤是：1）构建实物保护系统安全风险评估指标体系，采用博弈 论方法将主客观权重相结合，克服经典评估方法主观因素较强的不足；2）基于二维多 规则正向云发生器获取所有评估指标的云模型数字特征；3）采用基于 JS散度的云相似 度计算方法得出实物保护系统评估结果。仿真实验结果表明：CAPPS 方法可以准确发 现待评估系统的脆弱指标点，与对比方法相比具有更高的区分度。

（2）对实物保护系统而言，来自空中的威胁日益严重。现有的实物保护系统定量 评估方法只考虑了二维脆弱路径的搜索问题，而没有考虑三维脆弱路径的搜索问题。 针对实物保护系统定量评估中三维脆弱路径搜索的问题，提出了一种基于麻雀搜索优 化的实物保护系统脆弱性定量评估方法（Sparrow Assessment of Physical Protection System, SAPPS）。SAPPS 的步骤是：1）使用栅格法对实物保护系统部署区域进行三维 环境建模；2）为解决三维空间搜索脆弱路径不精确的问题，引入 EASI 模型 （Estimation of Adversary Sequence Interruption, EASI）的累积拦截概率对适应度函数进 行修正，实现三维环境下有效脆弱路径搜索；3）采用麻雀搜索算法，结合 Cubic 混沌 映射生成麻雀种群，避免算法后期陷入局部最优解。仿真实验结果表明：SAPPS 方法 可有效搜索易受攻击的入侵路径，与 MAPPS、DAPPS 和 HAPPS 三种方法相比拦截概 率显著提升。

（3）基于上述两项研究结果，开发了实物保护系统脆弱性评估软件工具。使用该 软件对某型一体化安防监控系统进行了脆弱性评估，获取了准确客观的脆弱性评估结 果，证明了该评估工具的有效性。

Deploying physical protection system is an important protective measure for critical facilities. How to evaluate the protection performance of key facilities under specific threats and identify the loopholes of the system in the physical protection system is an urgent problem to be solved at present. Research on the influence of subjective factors in qualitative assessment, the problem of vulnerable path search in three-dimensional space in quantitative assessment, and the design and development of vulnerability assessment system for physical protection system. The main work of this paper includes:

(1) Aiming at the existing qualitative assessment methods, the construction of security indicators is not perfect, the assessment results are greatly affected by subjective factors, and the assessment process is fuzzy and uncertain, etc., a qualitative assessment method for the vulnerability of physical protection systems based on a two-dimensional cloud model is proposed (Cloud Assessment of Physical Protection System, CAPPS). The steps of CAPPS are: 1) Constructing a security risk assessment index system for the physical protection system,, and use the game theory method to combine the subjective and objective weights to overcome the shortcomings of the classic assessment method with strong subjective factors; 2) Based on twodimensional multi-rule forward cloud generater the cloud model digital features of all evaluation indicators are obtained by the controller; 3) The evaluation result of the physical protection system is obtained by using the cloud similarity calculation method based on JS divergence. The simulation results show that the CAPPS method can accurately discover the vulnerable index points of the system to be evaluated, and has a higher degree of discrimination than the comparison method.

(2) For physical protection system, threats from the air are becoming increasingly serious. Existing quantitative evaluation methods for physical protection systems only consider the Subject : Research on Vulnerability Assessment Method of threedimensional Physical Protection System Specialty : Electronic Information Name : Yan Wenhao (Signature) Instructor : Li Aiguo (Signature) Wu Ruina (Signature) search problem of two-dimensional vulnerable paths, without considering the search problem of three-dimensional vulnerable paths. Aiming at the problem of three-dimensional vulnerable path search in quantitative assessment of physical protection system, a quantitative assessment method of physical protection system vulnerability based on sparrow search optimization (Sparrow Search Assessment of Physical Protection System, SAPPS) is proposed. The steps of SAPPS are: 1) Using the grid method to model the 3D environment of the physical protection system deployment area; 2) In order to solve the problem of inaccurate vulnerable path search in 3D space, the cumulative interception probability of EASI model (Estimation of Adversary Sequence Interruption, EASI) is introduced to modify the fitness function to realize effective vulnerable path search in 3D environment; 3) Sparrow search algorithm is used. In order to avoid falling into the local optimal solution in the later stage of the algorithm, the algorithm combines with Cubic chaotic map to generate sparrow population. Simulation results show that, SAPPS method can effectively search for vulnerable intrusion paths, and the interception probability is significantly improved compared with MAPPS method, DAPPS method and HAPPS method .

(3) Based on the above two research results, a vulnerability assessment software tool for physical protection system was developed. Using this software tool, a vulnerability assessment was conducted on a certain type of integrated security and surveillance system, and accurate and objective vulnerability assessment results were obtained, proving the effectiveness of this assessment tool.