The development of Structural Health Monitoring (SHM) systems and integration in our structures is a necessity. It has proven to provide a robust and low-cost solution for monitoring structural integrity and can predict the remaining life of our structures. One of the most important aspects of SHM systems is the design and implementation of sensor networks. This study proposes a new hybrid model for optimizing sensor placement on convex and non-convex structures. We propose a novel framework in which two detection mechanisms are considered: pitch-catch and pulse-echo to provide coverage for a given surface. These two mechanisms will complement each other to minimize the number of sensors used while maintaining a high coverage. This combination also allows for better coverage of the corners and regions in the proximity of geometrical discontinuity (such as holes and openings). The monitored area is discretized into a set of control points. For a control point to be covered, it should satisfy the user-defined coverage level which is the number of sensing paths crossing that point. These sensing paths are provided by two modes of communications (pitch-catch and pulse-echo) between the actuator-sensor pairs. The model, which is solved using a genetic algorithm (GA), provides flexibility by allowing the user to input different parameters such as the attenuation distance of the propagating waves and the sensing path limits of both coverage configurations that can be determined through experimentation. The efficiency of the proposed model is then demonstrated by simulating different geometrical shapes. Significant improvement in the coverage of the monitored area, reaching 34.6%, was achieved when compared to the coverage provided by some preliminary solutions such as uniformly placing the sensors on the plate under study. Also, the advantage of combining both configurations (pitch-catch and pulse-echo) in the same model was investigated. It was shown that the latter highly impacted the coverage in the blind zones (corners and edges) where a single configuration is not effective. Afterward, experimental validation was carried out to evaluate the model’s accuracy in damage localization within the optimized sensor networks. The results demonstrated the proficiency of the model developed in distributing the sensors on the tested specimens.

结构健康监测 (SHM) 系统的开发和我们结构中的集成是必要的。事实证明，它为监测结构完整性提供了强大且低成本的解决方案，并且可以预测我们结构的剩余寿命。SHM 系统最重要的方面之一是传感器网络的设计和实现。本研究提出了一种新的混合模型，用于优化凸面和非凸面结构上的传感器放置。我们提出了一种新颖的框架，其中考虑了两种检测机制：一发一收和脉冲回波，以提供给定表面的覆盖范围。这两种机制将相互补充，以最大限度地减少使用的传感器数量，同时保持高覆盖率。这种组合还允许更好地覆盖几何不连续性（例如孔洞和开口）附近的角落和区域。监控区域被离散化为一组控制点。对于要覆盖的控制点，它应该满足用户定义的覆盖水平，即通过该点的传感路径数。这些传感路径由执行器-传感器对之间的两种通信模式（一发一收和脉冲回波）提供。该模型使用遗传算法 (GA) 求解，允许用户输入不同的参数，例如传播波的衰减距离和可通过实验确定的两种覆盖配置的感测路径限制，从而提供了灵活性。然后通过模拟不同的几何形状来证明所提出模型的效率。与一些初步解决方案（例如将传感器均匀放置在所研究的板上）提供的覆盖范围相比，监测区域的覆盖范围显着提高，达到 34.6%。此外，还研究了在同一模型中结合两种配置（一发一收和脉冲回波）的优势。结果表明，后者严重影响了单一配置无效的盲区（角落和边缘）的覆盖范围。之后，进行了实验验证，以评估模型在优化传感器网络内的损伤定位精度。结果证明了开发的模型在将传感器分布在测试样本上的能力。