Feedback Control for Structural Health Monitoring in a Smart Aggregate Based Sensor Network

Abstract

The concept of smart aggregates, a distributed intelligent multi-purpose sensor network for civil structures, has been implemented to address three important issues including early-age concrete strength monitoring, impact detection and evaluation, and structural health monitoring. This paper presents mainly the employment of smart aggregates' active sensing property to form feedback in a sensor network to reduce damage-location detection time for lower power cost. Firstly, the concept of smart aggregates and the principle of a smart-aggregate-based sensor network are outlined. Next, the data pretreatment methods, including the sensor observation estimation model and the wavelet-packet-based signal processing algorithm, are proposed. A crucial concept using the damage index is also introduced. Moreover, the concept of the geometry structure matching method with the knowledge of an expert system is presented to determine which sensor is the optimal actuator. Finally, the data pretreatment algorithm and the geometry structure matching method are evaluated for a two-story concrete frame instrumented with smart aggregates as a testing object by means of actual experiments. The testing results demonstrate that the proposed algorithms are feasible and perform well in selecting optimal actuators of the sensor network for detecting damage locations.