For aerospace composite materials and structures, damage due to impact events may not be visible to surface inspection but still can cause significant loss of structural integrity. Therefore, an investigation was performed to develop a real-time health monitoring system for the identification and prediction of the location and force history of foreign object impact on composite panel structures with distributed built-in piezoceramic sensors. The smart health monitoring system is composed of two main subsystems: a measurement subsystem and an identification subsystem. The measurement subsystem with distributed built-in sensor network was used to collect and preprocess sensor data, and then the identification subsystem was implemented to reconstruct the force history and determine impact location with the acquired prefiltered sensor data. Thereupon, the identification subsystem consists of a structure system model, an inverse model operator (IMO) and a response comparator. The identification subsystem was created to identify the impact location and reconstruct the force history on composite structures without the need for the information about actual mechanical properties, geometries and boundary conditions of a structure, and without building a specific neural network with exhaustive training such as neural-network techniques, also without the need of constructing a full-scale accurate structural model. Consequently, a novel dynamic mechanical model based time-series model structure approach is used into the identification subsystem, where the entire impact identification procedure is much faster than that of the traditional model-based techniques. The smart health monitoring system was tested with various impact situations, for all of the cases considered, which verified the accuracy of impact load and position predictions, and the estimation errors fell well within the prespecified limit.