Font Size: a A A

Study On The Antifreeze Performance Of Rubber Aggregate Concrete Based On Neural Network

Posted on:2019-06-12Degree:MasterType:Thesis
Country:ChinaCandidate:L Y ZhangFull Text:PDF
GTID:2322330542974885Subject:Structural engineering
Abstract/Summary:PDF Full Text Request
Rubber aggregate concrete is made of rubber granules instead of concrete.The existing research shows that the incorporation of rubber particles can effectively improve the antifreeze and thaw cycle performance of concrete.To further explore the size and content of rubber particles affect the performance of concrete freezingthawing resisting cycle regularity,the rubber aggregate concrete freezing-thawing resisting cycling test,on the basis of through the establishment of the artificial neural network model,analysis of many influence factors,summed up the basic rule of concrete freezing and thawing cycle performance.Experimental study: the rubber particle size of 2.36mm-4.0 mm,0.43mm-0.60 mm,and 0.18mm-0.25 mm was used as substitute material,and the rubber aggregate concrete was prepared according to the ratio of 5 %,10% and 15 %.Using fast freezethaw cycling test,and the different rubber aggregate concrete under freeze-thaw cycles cube compressive strength,splitting tensile strength and dynamic modulus of elasticity,analyzes the change laws of various performance indicators.Test results show that the mixed rubber can improve concrete after freeze-thaw cycles,the relative dynamic elastic modulus,compressive strength and splitting tensile strength,the best rubber particle size of 0.18 mm to 0.25 mm,best rubber content is 10%.Based on neural network theory,uses the typical BP neural network model,with rubber particle size,the rubber content,water/cement ratio,cement dosage and block of freeze-thaw cycles as input,respectively with rubber aggregate concrete after freeze-thaw cycles of relative dynamic elastic modulus,the cube compression strength against relative splitting tensile strength and the cube as the output,set up three of the 5-5-1 three layer BP neural network model.The results show that the prediction results of BP neural network model agree well with the test results.Then use has built the BP network model,complete the content of rubber and rubber aggregate concrete antifreeze performance relationship,prediction results show that the mixed rubber can improve concrete after freeze-thaw cycle of compressivestrength,splitting tensile strength and relative dynamic elastic modulus and other properties.
Keywords/Search Tags:rubber concrete, freeze-thaw cycle, network model, rubber content
PDF Full Text Request
Related items