پديد آورندگان :
فصيحي، فاضل دانشگاه پيام نور تهران - بخش فني و مهندسي , كي منش، محمودرضا دانشگاه پيام نور تهران - بخش فني و مهندسي , صحاف، علي دانشگاه فردوسي مشهد - دانشكده مهندسي , قره، سهيل دانشگاه پيام نور تهران - بخش فني و مهندسي
كليدواژه :
ضريب بار هم‌ارز , شبكه عصبي مصنوعي , اجزاي محدود , آباكوس , روسازي‌هاي انعطاف‌پذير
چكيده فارسي :
ﯾﮑﯽ از ﻣﺸﮑﻼت اﺻﻠﯽ در زﻣﯿﻨﻪ روﺳﺎزي راهﻫﺎ، ﻋﺪم ﺷﻨﺎﺧﺖ دﻗﯿﻖ رﻓﺘﺎر روﺳﺎزي ﺗﺤﺖ ﺑﺎرﻫﺎي ﻋﺒﻮري و در ﻧﺘﯿﺠﻪ ﻋﺪم اﻣﮑﺎن ﺗﻌﯿﯿﻦ ﺿﺮﯾﺐ ﺑﺎر ﻫﻢارز در ﺗﺒﺪﯾﻞ ﺑﺎرﻫﺎ ﺑﻪ ﺑﺎر ﻣﻌﺎدل اﺳﺖ. ﺗﺤﻘﯿﻘﺎت ﺑﺴﯿﺎري در اﯾﻦ زﻣﯿﻨﻪ اﻧﺠﺎم ﺷﺪه ﮐﻪ ﮐﺎﻣﻞﺗﺮﯾﻦ آنﻫﺎ، روش ﻣﺒﺘﻨﯽ ﺑﺮ آزﻣﺎﯾﺸﺎت ﺟﺎﻣﻊ اﺷﺘﻮ اﺳﺖ. ﺿﻌﻒ اﺻﻠﯽ ﺿﺮاﯾﺐ ﺑﺎر ﻫﻢارز در اﯾﻦ روش، ﻣﺤﺪودﯾﺖ ﻧﺘﺎﯾﺞ ﺑﻪ ﻣﺤﻮرﻫﺎي ﺑﺮرﺳﯽ ﺷﺪه ﺑﻮده ﮐﻪ ﺑﺎﻋﺚ ﻋﺪم اﻣﮑﺎن ﺗﻌﯿﯿﻦ دﻗﯿﻖ ﺿﺮاﯾﺐ ﺑﺎر ﻫﻢارز ﺑﺮاي ﺗﻤﺎﻣﯽ ﻣﺤﻮرﻫﺎي ﻣﻮﺟﻮد اﺳﺖ. اﯾﻦ ﻋﻠﺖ را ﻣﯽﺗﻮان ﯾﮑﯽ از دﻻﯾﻞ ﺑﺮوز ﺧﺮاﺑﯽﻫﺎي زودرس و ﺻﺮف ﻫﺰﯾﻨﻪﻫﺎي ﺑﺎﻻي ﺗﻌﻤﯿﺮ و ﻧﮕﻬﺪاري راهﻫﺎ داﻧﺴﺖ. اﻣﺮوزه، ﺑﺎ ﭘﯿﺸﺮﻓﺖ ﻋﻠﻢ ﻧﺮماﻓﺰارﻫﺎي ﺑﺴﯿﺎري در زﻣﯿﻨﻪ ﺗﺤﻠﯿﻞ روﺳﺎزيﻫﺎ اﯾﺠﺎد ﺷﺪ ﮐﻪ ﻣﯽﺗﻮان از آنﻫﺎ در ﺗﻌﯿﯿﻦ اﯾﻦ ﺿﺮﯾﺐ اﺳﺘﻔﺎده ﮐﺮد. ﻣﺸﮑﻞ اﺻﻠﯽ ﻣﻮﺟﻮد در ﺗﻤﺎﻣﯽ آﻧﺎن، ﻧﯿﺎز ﺑﻪ دادهﻫﺎي ورودي ﻣﺘﻌﺪد، زﻣﺎنﺑﺮ ﺑﻮدن ﻓﺮاﯾﻨﺪ ﺷﺒﯿﻪﺳﺎزي و اﻣﮑﺎن ﺑﺮرﺳﯽ ﺗﻨﻬﺎ ﯾﮏ ﻣﻘﻄﻊ در ﻫﺮ زﻣﺎن ﻣﯽﺑﺎﺷﺪ. از ﻃﺮف دﯾﮕﺮ ﺷﺒﮑﻪﻫﺎي ﻋﺼﺒﯽ ﻣﺼﻨﻮﻋﯽ، ﺑﻪ ﻋﻨﻮان ﯾﮑﯽ از ﺷﺎﺧﻪﻫﺎي ﻋﻠﻢ ﻫﻮش ﻣﺼﻨﻮﻋﯽ داراي ﻣﺰاﯾﺎي زﯾﺎدي اﺳﺖ ﮐﻪ از آن ﺟﻤﻠﻪ ﻣﯽﺗﻮان ﻣﺤﺪود ﮐﺮدن ﺗﻌﺪاد دادهﻫﺎي ورودي، ﺳﺮﻋﺖ ﺑﺎﻻي ﻓﺮاﯾﻨﺪ ﻣﺪلﺳﺎزي، ﺗﻮاﻧﺎﯾﯽ ﻣﺪلﺳﺎزي ﻫﻢزﻣﺎن ﭼﻨﺪﯾﻦ روﺳﺎزي ﺑﺎ ﺷﺮاﯾﻂ ﻣﺨﺘﻠﻒ را ﻧﺎم ﺑﺮد. ﻟﺬا در اﯾﻦ ﭘﮋوﻫﺶ ﭘﺲ از اﻃﻤﯿﻨﺎن از ﺻﺤﺖ ﻧﺤﻮه ﻣﺪلﺳﺎزي روﺳﺎزيﻫﺎي اﻧﻌﻄﺎفﭘﺬﯾﺮ ﺑﺎ اﺳﺘﻔﺎده از ﻧﺮماﻓﺰار اﺟﺰاي ﻣﺤﺪود آﺑﺎﮐﻮس، ﺑﻪ ﻃﺮح ﺷﺒﮑﻪ ﻋﺼﺒﯽ ﻣﺼﻨﻮﻋﯽ ﺟﻬﺖ ﺗﻌﯿﯿﻦ ﺿﺮﯾﺐ ﺑﺎر ﻫﻢارز ﭘﺮداﺧﺘﻪ ﺷﺪه اﺳﺖ. در ﻧﻬﺎﯾﺖ ﺷﺒﮑﻪ ﺑﻬﯿﻨﻪ از ﻧﻮع اﻧﺘﺸﺎر ﺑﺮﮔﺸﺘﯽ ﭘﯿﺶﺧﻮر ﺑﺎ آراﯾﺶ 7-13-1 و ﺗﺎﺑﻊ اﻧﺘﻘﺎل ﺳﯿﮕﻤﻮﯾﺪ ﺑﻌﻨﻮان ﺷﺒﮑﻪ ﺑﻬﯿﻨﻪ اﻧﺘﺨﺎب ﮔﺮدﯾﺪه اﺳﺖ.
چكيده لاتين :
Lack of accurate knowledge of pavement behavior under moving loads is the one of the most important disadvantages in calculation of Equivalent Axle Load Factor (EALF) in roads pavement. Among the many researches, the most comprehensive method is based on the AASHTO road test. As the main weakness of this method, the results are limited to the experimented axles, which makes it impossible to determine the EALF for all existing axles, hence reducing the accuracy of the results, causing premature failure, and leading to higher maintenance costs. Today, although numerous software packages are available for EALF calculation, they require various parameters, are time-consuming, and can only simulate one section at a time . On the other hand, artificial neural networks, as an artificial intelligence subcategory, have many advantages such as reduced input data, increased modeling process speed, ability of parallel modeling of several pavements with different conditions, etc. In this paper, after verifying the simulation of flexible pavements in ABAQUS, a model based on Artificial Neural Network (ANN) was presented to calculate EALF using the back-propagation architecture. Finally, from among the reviewed ANN configurations, the network with the 7-13-1 architecture incorporating the sigmoid function was selected as the optimum network.
