Enhancing water pressure sensing in challenging environments: A strain gage technology integrated with deep learning approach

Abstract

The manuscript introduces a novel approach to design and construct a pore water pressure sensor utilizing strain gage technology integrated with deep learning principles. This sensor type is specifically tailored for measuring pressure at the vertex of pile bases in structures with substantial load-bearing capacity. While existing pressure sensors employing strain gage technology are available, this research addresses a unique measurement model suited for deep-water environments characterized by high corrosiveness and heavy loads. Consequently, the manuscript proposes design innovations aimed at optimizing the sensor’s form and dimensions to accommodate these demanding conditions. Computational simulations are conducted to perform relevant calculations, with results validated through rigorous analysis and experimentation against real-world datasets. Moreover, the study incorporates a pioneering deep learning-based data acquisition model to enhance output values, a feature currently underutilized in sensor technology. The findings demonstrate the viability of the proposed water pressure sensor model in various challenging working environments. This research underscores the potential for proactive manufacturing of sensors in diverse configurations, emphasizing adaptability and efficiency.