انتخاب تابع موجك مناسب در تشخيص خرابي ساختمان پيش ساخته پانلي مبتني بر نتايج آزمايشگاهي و روش عددي

Selecting the appropriate wavelet function in the damage detection of precast panel building based on experimental results and numerical method

بيشتر سازه‌هاي عمراني به مرور زمان و تحت شرايط محيطي و بارهاي خارجي دچار آسيب مي‌شوند و همواره وقوع خرابي امري اجتناب‌ناپذير است. لذا تشخيص آسيب، موضوع بسياري از مطالعات بوده است. در اين زمينه، تبديل موجك كه يك ابزار رياضي توانمند پردازش سيگنال است، در حوزه‌ي پايش سلامت، نظر بسياري از پژوهشگران را به خود جلب كرده است. پژوهش حاضر به پايش سلامت ساختمان پيش‌ساخته‌ي پانلي كامل بر مبناي نتايج آزمايشگاهي و به كمك روش تبديل موجك پيوسته پرداخته است و آسيب‌هايي كه ممكن است در سازه‌هاي مذكور رخ دهند، ارزيابي شده است. با توجه به نتايج آزمايشگاهي و تحليل نتايج توابع موجك مختلف، موقعيت خرابي به كمك تابع 5 c‌o‌i‌f با پارامتر مقياس 8 با موفقيت شناسايي شده است. همچنين اثر مودهاي بالاتر در نمايش شدت خرابي محسوس‌تر بوده است، به طوري كه در مود پيچشي سازه، بيشينه‌ي ضرايب موجك بزرگ‌تر و شدت خرابي بيشتر نمايان شده است.

Most building structures are damaged over time under environmental conditions and external loads. In this regard, the occurrence of damage is common and the detection of damage is the subject of much research. In this regard, wavelet conversion, which is a powerful mathematical tool for signal processing, has attracted the attention of many researchers in the field of health monitoring. In this study, free vibrations of a four-story building with specified boundary conditions and monitored the health of the building based on experimental results using the continuous wavelet analytical method are studied and the damage that may occur in these structures were evaluated and analyzed. The finite element software is used to Model of the Building by the sandwich model. In this four-story building, eight-layer sandwich panel (polystyrene, concrete, steel, concrete) is used symmetrically. The fourteen natural frequencies of the sandwich structure were compared with the experimental model and the main modes of the structure were obtained to influence the health of the structure. An error of less than 2.5% reveals a good match between the results of the two models. Precast panel health monitoring results show that based on the experimental results, the damage location using the coif5 function with scale parameter 8 has been successfully identified and showed a higher perturbation of the coefficients at the damage locations than the other functions. Thus, the relative maximum and minimum jumps in the wavelet coefficients occurred at the location of the damage and considering the maximum or minimum wavelet coefficients generated at the damage location as the center of damage, the damage center can be identified with an error of less than 8%. Also, effects of higher modes are more pronounced in the damage intensity index as in the torsional modes of the structure, the maximum wavelet coefficients are greater and the intensity of the damage more pronounced.