An Intelligent Optimization Method of Durable Concrete Mix Proportion Based on Data mining

• Main Authors: Yin, Yingjun, Wu, Xianguo, Wang, Lei, Chen, Bin, Chen, Hongyu, Zhang, Limao, Dai, Xiaosong, Jin, Wu, Yang, Tingyou, Liu, Pengcheng, Wang, Fan, Wang, Kunyu, Wang, Hongtao, Yuan, Fuyin, Tu, Liuqing, Yan, Tiwen, Hu, Yi, Li, Tiejun
• Format: Patent
• Language: English
• Published: 27.08.2020
• Subjects: ARTIFICIAL STONE; CALCULATING; CEMENTS; CERAMICS; CHEMISTRY; COMPOSITIONS THEREOF, e.g. MORTARS, CONCRETE OR LIKE BUILDINGMATERIALS; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; CONCRETE; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; LIME, MAGNESIA; METALLURGY; PHYSICS; REFRACTORIES; SLAG; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR; TREATMENT OF NATURAL STONE

The invention belongs to the technical field of concrete multi-objective mix proportion optimization, and discloses a method for optimizing the mix proportion of base RF-NSGA-II durable concrete, which mainly comprises the following steps: step 1, taking the frost resistance and impermeability of concrete as research objects, establishing an original sample set based on concrete materials and mix proportion factors; step2, respectively establish prediction models of relative dynamic elastic modulus and chloride ion permeability coefficient of concrete by using random forest regression algorithm (RF), carry out model performance evaluation, and obtain two trained prediction functions; Step 3: Two non-linear prediction functions are taken as multi-objective functions for optimizing the mix proportion, application constraints of various influencing factors are combined, and NSGA-II is used for global optimization to obtain the optimal mix proportion. By utilizing the established RF-NSGA-II model, the invention not only realizes high-precision prediction of the relative dynamic elastic modulus and chloride ion permeability coefficient of concrete, but also realizes multi-objective intelligent optimization of concrete mix proportion.