چكيده به لاتين
Experimental residual stress measurement methods such as successive hole-drilling or slitting are the most common intralabratory techniques because of the appropriate precision, standard procedure and convenience of the performance. However, despite the mentioned advantages, the uncertainty of the obtained stresses, the severe influence of the error and noise on the final results and lack of knowledge about the minimum safe interval to repeat the tests are some major drawbacks of these methods.
In the present research, by utilizing the concept of the eigenstrain, a novel approach was developed to overcome the preceding downsides. In the first step, the repeatability of the slitting experiment was investigated on the metals to obtain the uncertainty of the solution; a new parameter named as the residual stresses relief distance (RSRD) was introduced using the concept of the eigen-strain and the FEM simulations depicted that the parameter determines the minimum safe slitting interval. In the next phase, four strain gages were installed on the aluminum part containing known residual stress field and the slitting experiments were performed which ensured the validity of the computations. By extending the previous method for the orthotropic materials, the minimum safe slitting distance for polymer matrix composites was achieved and by carrying out four slitting tests, the measured elastic strains likewise the aluminum part had good agreement with each other. Eventually, the residual stresses were calculated based on conventional and eigen-strain based approaches where the effect of the noise on the final results was obvious. In order to rectify this problem, the least square method was developed for the pulse functions and it was shown that in spite of its influence on decreasing the resolution, it contributes to the diminishment of the noise effect. Other techniques such as Tikhonov regularization and data reduction were also examined on the both conventional and eigen-strain based methods as well and it was depicted that the eigen-strain based method is capable of diminishing the influence of the noise in the absence of any other noise-reduction method.
