Structural Health Monitoring (SHM) is the process of implementing a damage detection and characterization strategy for engineering structures using continuous or periodic measurements. It combines sensing, data acquisition, signal processing, and decision-making to assess the condition of a structure over its lifetime.
Structural Health Monitoring (SHM) is the process of implementing a damage identification strategy for aerospace, civil, and mechanical engineering structures. The goal is to detect, locate, and quantify damage β and, ideally, to predict the remaining useful life of the structure β using measurements from a network of sensors installed on or within the structure. SHM is fundamentally interdisciplinary, drawing on structural dynamics, signal processing, machine learning, sensor technology, and reliability engineering. The field has grown rapidly over the past three decades, driven by the aging of critical infrastructure, the increasing use of composite materials, and the availability of low-cost sensing and computing.
The core paradigm of SHM is often expressed as a four-level hierarchy: (1) detection β is damage present? (2) localization β where is it? (3) quantification β how severe is it? (4) prognosis β what is the remaining life? Most practical SHM systems today achieve levels 1 and 2, while levels 3 and 4 remain active research areas. Damage is typically identified through changes in the structure's modal properties (natural frequencies, mode shapes, damping), through local measurements (strain, acceleration, acoustic emission), or through statistical pattern recognition applied to large sensor datasets. The choice of approach depends on the structure type, the expected damage mechanisms, the environmental and operational variability, and the required level of confidence.
SHM is broadly divided into two paradigms. Model-based (physics-based) SHM compares measured responses to predictions from a numerical model (typically finite element); damage is inferred from the residual β the difference between model and measurement. This approach requires an accurate model and is sensitive to modeling errors. Data-driven SHM relies on statistical or machine learning methods to identify patterns in the measurement data, often using baseline (healthy-state) data as a reference. Modern SHM increasingly combines both paradigms in a hybrid approach. Applications span a wide range β from bridges, buildings, and offshore platforms to wind turbines, aircraft, and pipelines. In earthquake engineering, SHM is particularly valuable because it can provide rapid post-earthquake assessment, distinguishing between structures that are safe to reoccupy and those requiring inspection. The 2011 Christchurch earthquake, the 2016 central Italy earthquakes, and recent events in Japan have all highlighted the value of pre-installed monitoring systems for rapid post-event decision-making. In Iran, SHM is an emerging field, with pilot deployments on bridges, dams, and important buildings, and growing interest in the integration of SHM with earthquake early warning and rapid response systems.