Optimization of mechanical characteristics of models of laminate composites using embedded optical fiber strain sensors

The problem of analysis and prediction of the mechanical behavior of modern composite materials and structures at the stage of their design, production and in-service conditions is of great importance. One of the most promising solutions in the field of real-time monitoring of the mechanical state of composite structures is associated with smart materials and smart systems based on the sensor elements. The data obtained during operation on the state of the structure can be used both for monitoring the mechanical state of structures and for refining mathematical models to predict the failure processes. This paper is devoted to the approach according to which indications of the embedded fiber-optical strain sensors (FOSS) with Bragg gratings are used to refine the mechanical characteristics of a laminate composite material. The essence of the approach is to estimate the difference between the deformation response predicted using the model with the data obtained in real time with the help of the FOSS. To refine the model parameters in accordance with the information received from the FOSS, an algorithm is proposed, according to which the inverse problems are solved in order to ensure that the numerical and experimental results having the specified accuracy. The optimization parameters are the elastic material constants, which, in the final analysis, should ensure that the simulation results and the FOSS measurements are consistent in the control points. To optimize the parameters for the regression model, various minimization algorithms are used. The algorithm implementation is demonstrated on the example of the test problems of two types of composite samples with a concentrator (notches): with quasi-isotropic and transversal-isotropic plies.