In order to study the cracking scale and main fracture precursor characteristics of polypropylene fiber recycled concrete, 20 groups of specimens with coarse aggregate replacement rates of 25% and 50% were designed and manufactured. Through the uniaxial compression acoustic emission test, the acoustic emission energy and main frequency of polypropylene fiber recycled concrete fracture process were extracted. Using energy as a medium, the correspondence between cracking scale and acoustic emission characteristics is discussed. In this paper, K-means clustering method is used to perform a two-dimensional cluster analysis of large scale cracking and small scale cracking, and then a support vector machine is used to obtain the boundary between large scale cracking and small scale cracking. The evolution law of acoustic emission signal in the fracture process of polypropylene fiber recycled concrete was analyzed. It was found that the acoustic emission event rate decreased, the large scale cracking signal increased, and the low-energy small scale cracking signal disappeared. The concept of main fracture precursor response coefficient is introduced. The calculation results show that the precursor response coefficient of the disappearance characteristics of low energy small scale cracking signal is the largest, whose precursor response time is the earliest.

In order to study the effect of different mix proportions on the compressive strength of polypropylene fiber reinforced recycled concrete, 30 sets of specimens with coarse aggregate particle sizes of 5~10 mm and 10~20 mm at 4∶6, 5∶5, and 6∶4 were designed and conducted. Through uniaxial compression tests, the stress-strain curves, elastic modulus, peak strength, and other parameters of each set of specimens were obtained. The test results show that: As the coarse aggregate particle size decreases, the peak stress shows an overall downward trend; Polypropylene coarse fibers have the best strengthening effect on the elastic modulus, and the mixing of coarse and fine polypropylene fibers has the best strengthening effect on the peak stress of recycled concrete. A statistical damage constitutive model for polypropylene fiber reinforced recycled concrete was established, and it was found that the specimen with a ratio of 5∶5 for coarse aggregates of 5~10 mm and 10~20 mm exhibits the most severe damage, and the mixture of coarse and fine fibers had the best inhibitory effect on damage. Finally, support vector machine regression was used to predict the post peak strain of recycled concrete. The study show that using a linear kernel function and using the post peak compressive machine strain as the input variable had the best prediction effect.