Passive structural systems against rockfalls, as net fences and embankments, are among the most effective mitigation measures for high energy events. Although largely adopted, the design of these systems has not been codified yet. A profitable time-dependent reliability approach has been recently introduced by the Authors, accounting for different possible probability distributions of velocity, mass, and height of the impacting block. Two failure modes were considered, related to the energy absorption capacity and the intercepting height of the systems. As the current design approach is based on partial safety factors, several sensitivity analyses are herein presented, with a particular focus on the height, firstly to define a suitable combination of equivalent partial safety factors, and, secondly, to investigate the parameters related to the geometry and the kinematics of the block which mostly affect the factors. Moreover, for a given failure probability, two shallow neural network were built, one for each failure mode, considering the possible range of all the input parameters, creating thus two input-output relationships that can be used to evaluate the partial safety factors for the height and the energy.

诸如网状围栏和路堤之类的防止落石的被动结构系统是缓解高能事件的最有效措施。尽管已被广泛采用，但这些系统的设计尚未编纂。作者最近引入了一种有利时效的可靠性方法，该方法考虑了撞击块速度，质量和高度的不同可能概率分布。考虑了两种失效模式，分别与能量吸收能力和系统的拦截高度有关。由于当前的设计方法是基于局部安全系数的，因此在此介绍了几种灵敏度分析，其中特别着重于高度，首先是定义等效的局部安全系数的合适组合，其次，研究与主要影响因素的块的几何和运动学有关的参数。此外，对于给定的故障概率，建立了两个浅层神经网络，其中一个针对每种故障模式，考虑了所有输入参数的可能范围，因此创建了两个输入-输出关系，可用于评估安全系数的局部安全系数。高度和精力。