A reasonable arrangement of filling pipelines can solve the problems of low line magnification, a high flow rate, large pipe pressure, etc., in deep well filling slurry transportation. The transportation pressure loss value of filling slurry is the main parameter for the layout design of filling pipelines. At present, pressure loss data are mainly obtained through the loop pipe experiment, which has problems such as a large amount of labor, high cost, low efficiency, and a limited amount of experimental data. In this paper, combined with a new generation of artificial intelligence technology, the random forest machine learning algorithm is used to analyze and model the experimental data of a loop pipe to predict the pressure loss of slurry transportation. The degree of precision reaches 0.9747, which meets the design accuracy requirements, and it can replace the loop pipe experiment to assist with the filling design.
Balancing high fluidity and stability is a critical challenge in deep-shaft cemented paste backfill (CPB) with high-concentration tailings. This study investigates the synergistic regulation mechanism of a combined admixture system comprising hydroxypropyl methylcellulose (HPMC) thickener and polycarboxylate (PCE) or Melamine-Formaldehyde Resin (MFR) superplasticizers on CPB rheology, mechanical strength, and microstructure. Results indicate that HPMC significantly enhanced anti-segregation performance via intermolecular bridging, substantially increasing yield stress and plastic viscosity. Upon PCE introduction, the steric hindrance provided by its side chains effectively disrupted HPMC-induced flocs and released entrapped water. Consequently, yield stress and plastic viscosity were reduced by up to 22.1% and 64.3%, respectively, with PCE exhibiting markedly superior viscosity-reducing efficiency compared to MFR. Mechanical testing revealed that PCE co-addition did not compromise early-age strength but enhanced 3, 7, and 28-day unconfined compressive strength (UCS) by refining pore structures and promoting the uniform distribution of hydration products. Microstructural analysis unveiled a competitive adsorption mechanism: preferential PCE adsorption dispersed particle agglomerates, while non-adsorbed HPMC formed a viscoelastic network within the pore solution, constructing a stable “dispersion-suspension” microstructure. This work provides a theoretical basis for optimizing high-performance backfill formulations.