Study And Prediction Of Properties Of Styrene Methyl Copolymers/SBS Composite Modified Asphalt And Mixture

Styreneic Methyl copolymer(SMC),the normal temperature modifier is a new type of ambient temperature modified material developed by the Scientific Research Institute of the Ministry of Transport.It is from the waste rubber and waste plastic extract methyl styrene polymer materials such as methyl styrene block copolymer as the main raw materials of polymer solution,due to the modification process is simple and the advantages of the green environmental protection,cause the attention of some scholars in our country in recent years,at present,the performance of modified asphalt and its mixture of SMC research and application has achieved good effect,but and SBS composite modified asphalt and its mixture performance research is not thorough,SMC modifier can improve the low temperature performance of the mixture,but the mechanical properties and durability of the composite modified asphalt mixture need to be further studied.In this dissertation,SMC room temperature modifier on styrene-methyl copolymer/SBS composite modified asphalt(SMC/SBSCMA)and its mixture of performance research as the main purpose,4%SBS unchanged,study SMC dosage on the high and low temperature performance of composite modified asphalt and mixed mechanical properties and fatigue performance improvement;At the same time,SMC can reduce the mixing and compaction temperature of asphalt construction,which can fully reflect the characteristics of saving resources,protecting the environment and reducing investment.In addition,based on the machine learning model,a new idea is proposed to predict the dynamic modulus,phase Angle and fatigue life of composite modified asphalt mixture.The main research work and contents are as follows:Firstly,the conventional and high and low temperature properties of SMC/SBSCMA with different SMC content were studied.The viscosity and temperature curves were obtained by Brookfield rotational viscosity test.The mixing temperature and compaction temperature of SMC/SBSCMA with 8%,10%,12%and 14%content were 50℃～60℃lower than those of SBS modified asphalt(SBSMA),which had positive significance for energy saving and emission reduction,low-carbon and environmental protection.The PG grade of SMC/SBSCMA with 10%and 12%content can reach PG64-34,while the PG grade of SBSMA is PG64-28.SMC/SBSCMA has a wider temperature application range,and its low temperature performance is better than SBSMA,which can meet the minimum design temperature of-34℃pavement.Considering the conventional performance and high and low temperature performance of SMC/SBSCMA,it is considered that the content of 10%is relatively optimal.By means of scanning electron microscopy(SEM),Fourier transform infrared spectroscopy(FTIR)and scanning electron microscopy energy dispersive spectroscopy(SEM-EDS),it is found that SMC and SBS modified asphalt can be fully fused after mixing,and only simple physical reaction occurs,no new material is formed.Secondly,the viscoelastic properties of SMC/SBSCMA mixture were studied.Based on the dynamic modulus test,based on the Sigmoidal model,and according to the approximate Kramers-Kronig relationship,the frequency domain master curve models of SMC/SBSCMA mixture with different mixing quantities and SBSMA mixture were constructed on the premise that all the master curve models had the same shift factor.The discrete spectrum was obtained based on the Prony series,and the time domain principal curve of each asphalt mixture was further obtained.By comparing the frequency domain and time domain curves of each asphalt mixture,it can be concluded that:Under the condition of low temperature and high frequency,SMC/SBSCMA mixture has lower dynamic modulus value,better flexibility and prominent hysteresis phenomenon,and the larger the SMC content,the smaller the dynamic modulus value,indicating that SMC/SBSCMA mixture has better low-temperature resistance to vehicle load impact,vibration resistance and noise reduction effect than SBSMA mixture.At the same time,the larger the SMC content,The better the crack resistance at low temperature;The addition of SMC modifier reduced the relaxation modulus and increased the creep flexibility of the mixture.At a short time of low temperature,the relaxation modulus of 12%SMC/SBSCMA mixture was the smallest and the creep flexibility was the largest,followed by8%SMC/SBSCMA mixture,indicating that SMC/SBSCMA mixture had good crack resistance at low temperature,but the comprehensive high temperature performance,The content of SMC should be between 8%and 10%.Thirdly,the study of SMC/SBSCMA dynamic modulus and phase Angle machine learning constitutive model.The machine learning constitutive models of SMC/SBSCMA mixture dynamic modulus and phase angle were constructed under the influence factors of different SMC content,different temperature and different load frequency.And expand the multi-model comparison evaluation,at the same time,compared with the predicted|E*|value of Sigmoidal type(S)constitutive model,with the root mean square error(RMSE)and the determination coefficient(R~2)model of performance evaluation and analysis to determine the basic form of constitutive model.Results show that Support Vector Machine(SVM)constitutive model can not only consider the influence of all parameters,but also has the highest accuracy.The prediction results of dynamic modulus|E*|(RMSE=302.1494,R~2=0.9897),and prediction of phase angle(RMSE=1.5239,R~2=0.9229),the prediction effect is best,It has the characteristics of fast modeling and strong engineering practicability.Fourthly,the fatigue resistance of SMC/SBSCMA mixture was studied.Of different content of SMC SMC/SBSCMA mixture specimen,the implementation of semicircle bending(SCB)asphalt mixture fatigue test,compare and analyze different fatigue specimens with different content of SMC,different oil-stone ratio different stress ratio and the influence of load frequency on the fatigue performance,and establish the phenomenological fatigue equation,to find the optimal fatigue factor conditions.The evolution of fatigue cracking behavior of SMC/SBSCMA mixture was carried out by image processing technology.The results show that the traditional fatigue equation reflects the law of fatigue life-stress ratio(stress amplitude)evaluated by a single factor,and the coefficient of determination of all fitted equations is R~2≥0.877;The comprehensive analysis shows that the maximum fatigue life is obtained when the SMC content is 10%,the oil-stone ratio is 4.3 and the load frequency is 15Hz.Finally,the fatigue life prediction model of Support Vector Machine based on particle swarm optimization(PSO＿SVM)is studied.According to the characteristics of SCB test and PSO＿SVM model,combined with the traditional fatigue equation prediction model to determine the influencing parameters of the model,a PSO＿SVM prediction model of SMC/SBSCMA mixture fatigue life was proposed,which took SMC content,oil-stone ratio,stress ratio and load frequency as the influencing factors.70%of SCB test data were used as training samples.Particle swarm optimization(PSO)was used to optimize the optimal hyperparameter penalty factor C,widthεand kernel function parameter g of Support Vector Machine(SVM)model to construct the PSO＿SVM prediction model.Compared with other machine learning models,the PSO＿SVM prediction model is verified to be feasible and has better performance in predicting the fatigue life of SMC/SBSCMA mixture by analyzing the error between the measured and predicted values of 30%randomly selected test samples.In addition,the coefficient of determination index R~2of the PSO＿SVM model reached 0.9792,which was larger than the minimum coefficient of determination of the traditional phenomenological fatigue equation of 0.877,and the evaluation result was more accurate.The PSO＿SVM model has been verified and can be used to predict the fatigue life of SMC/SBSCMA mixtures with SMC content,oil-stone ratio,stress ratio and load frequency as the influencing factors.It has high simulation accuracy and rapid engineering application characteristics,and provides a new method and means for predicting the fatigue life of SMC/SBSCMA mixtures.