Toward Workable and Cost-Efficient Monitoring of Unstable Rock Compartments with Ambient Noise

Ambient Vibration-Based Structural Health Monitoring (AVB–SHM) studies on prone-to-fall rock compartments have recently succeeded in detecting both pre-failure damaging processes and reinforcement provided by bolting. The current AVB–SHM instrumentation layout is yet generally an overkill, creating cost power issues sometimes requiring advanced signal processing techniques. In this article, we paved the way toward innovative edge-computing approach tested ambient vibration records made during bolting of a ~760 m3 limestone column (Vercors, France). First, established some guidelines for comparing several basic, computing-efficient, seismic parameters (i.e., Fast Fourier Transform, Horizontal to Vertical Spectral Ratios). All three performed well revealing unstable compartment’s fundamental resonance frequency. HHSR appeared as most consistent spectral estimator, succeeding higher modes. Only mode should be trustfully monitored with HVSR since peaks may artifacts. Then, first application novelty detection algorithm case study showed following: feasibility automatic removing adverse thermomechanical fluctuations column’s dynamic based machine learning, systematic clear, permanent change behavior after grout injection hardening around bolts (i1 i2). This implementation represents significant workload reduction, compared physical-based algorithms or numerical twin modeling, shows better robustness regard gaps. We believe that monitoring systems combining basic techniques could help facilitate remote natural structures such compartments.