Powder Technology最新文献

{"title":"Machine learning rapid prediction for quality indicators in advanced manufacturing processes of titanium alloys","authors":"Ge Yu ,&nbsp;Yifu Long ,&nbsp;Deyu Yue ,&nbsp;Meng Li ,&nbsp;Yafeng Yang ,&nbsp;Shaofu Li ,&nbsp;Xizhong An","doi":"10.1016/j.powtec.2026.122166","DOIUrl":"10.1016/j.powtec.2026.122166","url":null,"abstract":"<div><div>Laser powder bed fusion (L-PBF) additive manufacturing (AM) of titanium alloys involves complex interactions between thermal and fluid phenomena that strongly influence molten pool geometry, surface morphology, and defect formation. However, conventional experimental and numerical optimization of process parameters is computationally expensive and unsuitable for rapid quality prediction. This study establishes a machine learning (ML) – Shapley additive explanation (SHAP) framework for fast and interpretable prediction of key quality indicators, including molten pool depth, surface roughness, and pore type of L-PBF titanium alloys, using datasets generated from a validated discrete element method (DEM) - computational fluid dynamics (CFD) numerical simulations. Seven ML algorithms (Regression, KNN, BPNN, DT, RF, SVR, and XGBoost) are systematically evaluated with min-max normalized inputs and 5-fold cross-validation. Among them, XGBoost demonstrates the best balance of prediction accuracy, generalization, and efficiency across both regression and classification tasks. To enhance interpretability and reduce the impact of experimental design, SHAP is applied to quantify feature contributions, nonlinear dependence trends, and parameter interaction effects. The SHAP results reveal that molten pool depth is primarily governed by laser power, whereas surface roughness and pore type are dominated by laser scanning velocity. Dependence and partial dependence analysis further uncover monotonic relationships and class switches, such as the strong positive sensitivity of depth to energy input, the suppression of shrinkage pores driven by velocity, and the linear promotion of lack-of-fusion pores by hatch spacing. SHAP interaction matrices confirm that coupled effects between laser power and scanning velocity control both molten pool morphology and pore transition mechanisms, while combinations involving hatch spacing exhibit only weak interactions. Overall, this interpretable ML-SHAP framework not only achieves rapid and accurate prediction of material quality indicators but also provides mechanistic insights consistent with underlying L-PBF physics. The developed approach offers a practical and physically informed tool for intelligent parameter optimization and quality control in additive manufacturing of titanium alloys.</div></div>","PeriodicalId":407,"journal":{"name":"Powder Technology","volume":"473 ","pages":"Article 122166"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146076481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel size-class-weighted DEM calibration framework for predicting the moisture-dependent flowability of soybean","authors":"Xiangwei Liu ,&nbsp;Changyu Wang ,&nbsp;Yuqing Feng ,&nbsp;Yandi Wang ,&nbsp;Ruizhi Dong ,&nbsp;Heiko Briesen ,&nbsp;Yuan Tan","doi":"10.1016/j.powtec.2026.122162","DOIUrl":"10.1016/j.powtec.2026.122162","url":null,"abstract":"<div><div>During the storage and handling of food grains, moisture variation causes particles to swell and agglomerate, which strongly influences their flow behavior. As the most suitable numerical modeling tools for such particulate systems, the discrete element method (DEM) still struggles to balance geometric accuracy, computational efficiency and experimental calibration effort at different moisture levels. This study develops a size-class–weighted DEM calibration framework that efficiently captures the combined influence of moisture-induced particle morphology on flowability, using soybeans as a representative food grain. The framework enables rapid construction of quasi-spherical particle models composed of six sub-spheres, reproducing shape changes across a broad moisture range (0.50–21.62 wt%) while reducing calibration workload by ∼75%. Calibration was performed using angle-of-tilting tests, from which a flowability index was derived. Validation through angle-of-repose tests achieved up to 99.6% agreement with experiments, confirming the framework's reliability. The calibrated and validated model revealed a non-monotonic relationship between moisture and flowability, with the lowest flowability observed at 17.12 wt% moisture before improving at higher levels. Overall, this work provides a physically grounded and computationally efficient approach for predicting moisture-dependent flowability in soybeans and other grains with similar morphology, enhancing the understanding and control of particulate behavior in food processing systems.</div></div>","PeriodicalId":407,"journal":{"name":"Powder Technology","volume":"473 ","pages":"Article 122162"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146076423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Discrete element model for collisions of multiple wet particles","authors":"Rajarshi Chattopadhay,&nbsp;Robert H. Davis","doi":"10.1016/j.powtec.2026.122201","DOIUrl":"10.1016/j.powtec.2026.122201","url":null,"abstract":"<div><div>Collisions of granular particles coated with thin viscous liquids are simulated using a hybrid time- and event-driven algorithm to maintain chronological order of hard-sphere collisions while ensuring numerical stability in highly dissipative systems. A novel approach was adopted for many-particle simulations using a rotating polar coordinate system that decomposes the relative pairwise motion into normal and tangential components along the line-of-centers. This approach provides better accuracy than Cartesian coordinates for the same timestep size, with only small increase in computational time per step. Collisions of three wetted spheres were analyzed by varying Stokes number (<span><math><mrow><mi>S</mi><msub><mrow><mi>t</mi></mrow><mrow><mi>n</mi></mrow></msub></mrow></math></span>, ratio of particle inertia to viscous forces), fluid film thickness, capillary number (ratio of viscous to interfacial tension forces), and collision angles to reveal several outcome regimes: complete agglomeration, partial coagulation, and full separation. Simulations were extended to a billiards configuration, where a single striker sphere impacts an ordered cluster. Here, multiple simultaneous fluid interactions enhance viscous dissipation and broadens the range of partial agglomeration outcomes, whereas asymmetric collision promotes fragmentation. Finally, a system of randomly distributed 1000 spheres in a periodic domain, with some particles having initial velocity, was studied for varying <span><math><mrow><mi>S</mi><msub><mrow><mi>t</mi></mrow><mrow><mi>n</mi></mrow></msub></mrow></math></span> and packing fraction (<span><math><mi>ϕ</mi></math></span>). All cases exhibit rapid decay of kinetic energy, with lower <span><math><mrow><mi>S</mi><msub><mrow><mi>t</mi></mrow><mrow><mi>n</mi></mrow></msub></mrow></math></span> and higher <span><math><mi>ϕ</mi></math></span> displaying steeper slope, followed by a plateau corresponding to the conserved center-of-mass kinetic energy. The coordination number (number of simultaneous fluid overlaps per particle) gradually increases with time as more fluid overlaps occur, saturating at higher values for dense systems (high <span><math><mi>ϕ</mi></math></span>) and low <span><math><mrow><mi>S</mi><msub><mrow><mi>t</mi></mrow><mrow><mi>n</mi></mrow></msub></mrow></math></span>.</div></div>","PeriodicalId":407,"journal":{"name":"Powder Technology","volume":"473 ","pages":"Article 122201"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146076420","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deposition of coal dust particles in the human upper respiratory tract during unsteady breathing: Insights from CFD-DPM simulation and XGBoost-SHAP analysis","authors":"Bingyou Jiang ,&nbsp;Haoyu Wang ,&nbsp;Yu Zhou ,&nbsp;Jie Wang","doi":"10.1016/j.powtec.2026.122173","DOIUrl":"10.1016/j.powtec.2026.122173","url":null,"abstract":"<div><div>Dust particles in coal mine roadways typically exhibit a broad particle size distribution (PSD) that evolves along the transport path. Current dust risk assessments primarily focus on environmental/individual external concentrations and time-weighted average (TWA) exposure. However, the true determinants of upper respiratory tract (URT) health effects are the deposition mass (<em>DM</em>) and flux within functional regions following inhalation. A CFD-DPM-based numerical model was developed and validated to simulate transient dust transport and deposition at measurement points (MPs) under varying respiratory flow rates, and the XGBoost-SHAP method was applied to analyze deposited mass per breath (<em>DM</em>_<em>PB</em>) in nasal functional regions. Results show that increased respiratory flow significantly elevates vortex volume proportion (<em>V</em>_<em>Ω</em>^<em>⁎</em>), average wall shear stress (<em>AWSS</em>), and secondary flow intensity (<em>Se</em>) in the nasal cavity. Compared with exhalation, inhalation yields lower peak <em>V</em>_<em>Ω</em>^<em>⁎</em>, while <em>Se</em> follows a sinusoidal pattern over time. PSD is the dominant factor for deep nasal deposition, and high exposure concentration does not necessarily increase <em>DM</em>_<em>PB</em>, particularly at high breathing flows. The cut-off particle size for penetration into lower respiratory tract (LRT) decreases exponentially with respiratory flow. SHAP analysis reveals that nasal deposition prediction is mainly driven by the aerodynamic diameter (<em>dₐ</em>), with dust concentration (<em>C</em>_<em>MP</em>) as a secondary factor for anterior deposition and impaction parameter (<em>IP</em>) as a secondary factor for middle and posterior deposition. These findings indicate concentration alone is insufficient for assessing respiratory deposition risk; PSD must also be considered to better capture deposition patterns and health implications across nasal regions.</div></div>","PeriodicalId":407,"journal":{"name":"Powder Technology","volume":"473 ","pages":"Article 122173"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146076461","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Partitioning populations of real and manufactured particles using spherical harmonics representations","authors":"Gabriel Laberge,&nbsp;Marc Laforest,&nbsp;Serge Prudhomme","doi":"10.1016/j.powtec.2026.122216","DOIUrl":"10.1016/j.powtec.2026.122216","url":null,"abstract":"<div><div>The objective of this paper is to review the discriminating power of spherical harmonic representations of the surfaces of particles, specifically for the task of partitioning a population of particles into clusters. By comparing the results of partitioning tasks on both classical shape descriptors and spherical harmonics, and by performing these tasks on both real populations of grains and populations of manufactured (analytical) grains, it is possible to come to some conclusions about the geometrical content of spherical harmonics. Spherical harmonics are compared to the classical shape descriptors of sphericity, convexity, corners, roundness, and the indices of elongation and flatness. Eight types of manufactured particles, each with various numbers of free parameters, are used to generate artificial populations that are to be partitioned. The types of manufactured particles are called spheres, prolates, oblates, cubes, cylinders, peanuts, diamonds and popcorn, but only the last two contain non-smooth surfaces either at points or along curves. Detailed experiments are performed to partition both manufactured and real populations using a Gaussian mixture model and the Bayesian information criterion to determine the number of such subpopulations. The experiments show that classical shape descriptors provide geometrical characterizations that can only separate out the most spherical particles from other particles, while spherical harmonics are effective throughout the range of smooth to non-smooth particles. Clustering is also performed for four different real populations of particles and contingency matrices are presented. The paper also includes several new benchmarks for the verification of calculations of spherical harmonics from micro-tomographic scans.</div></div>","PeriodicalId":407,"journal":{"name":"Powder Technology","volume":"473 ","pages":"Article 122216"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146185360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Application of similarity criteria in mechanistic study of spray-based dust suppression at open-pit truck unloading stations: Insights from scaled experiments and numerical simulations","authors":"Fabin Zeng ,&nbsp;Zhongan Jiang ,&nbsp;Guoliang Zhang ,&nbsp;Bin Yang ,&nbsp;Ming Wang ,&nbsp;Yapeng Wang ,&nbsp;Ya Chen ,&nbsp;Mingli Si ,&nbsp;Wei Ma ,&nbsp;Yee-Chung Jin","doi":"10.1016/j.powtec.2026.122239","DOIUrl":"10.1016/j.powtec.2026.122239","url":null,"abstract":"<div><div>During unloading at open-pit coal mine truck unloading stations, strong impact airflows and high dust concentrations make these sites dominant sources of dust emissions. Yet the multi-scale coupling among “impact airflow-dust dispersion-spray capture” insufficiently understood, limiting the transfer of spray design parameters from experiments to engineering applications. This study develops a dynamic similarity framework for spray-based dust suppression grounded in similarity criteria theory, identifying key dimensionless groups {Stk, Reₚ, Re_g, Re_d, <em>δ</em>, Pe_D, <em>d</em>_p/<em>D</em>_w, Λ, We_l} that govern gas-solid-liquid interactions. Through integration of a 5:1 scaled experimental platform and CFD-DPM numerical simulations, dynamic consistency is achieved. The results show that dust-droplet interactions undergo three distinct stages: capture, coalescence, and escape. Dust suppression efficiency increases with airflow rate, particle size and interaction distance, but exhibits a rise-fall trend with liquid flow rate; droplet size decreases with gas pressure and increases with liquid pressure. Nozzle spacing and ambient wind velocity primarily control the radial and axial development of the spray field. Considering both dust control and moisture constraints, optimal parameters are <em>P</em>_G ≥ 0.6 MPa, <em>P</em>_L ∈ (0.2, 0.3) MPa, and <em>L</em>s ∈ (30, 40) cm, yielding total and respirable dust reduction rates of 90% and 86%, respectively. The proposed framework clarifies the multi-scale dynamics of spray dust suppression at truck unloading stations and provides a transferable similarity-based basis for the scientific design and optimization of dust control systems in open-pit coal mines.</div></div>","PeriodicalId":407,"journal":{"name":"Powder Technology","volume":"473 ","pages":"Article 122239"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146185359","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Moving radial basis function method for solving the population balance equation for growth, coagulation, and deposition processes","authors":"Kaiyuan Wang ,&nbsp;Xue Gong","doi":"10.1016/j.powtec.2026.122206","DOIUrl":"10.1016/j.powtec.2026.122206","url":null,"abstract":"<div><div>Particle growth, coagulation, and deposition are fundamental processes in various particulate systems, with the size distribution evolution governed by the population balance equation (PBE). The radial basis function method is a promising numerical technique for solving the PBE. However, existing developments remain insufficient to fully resolve particle growth and coupled particle dynamics. This study proposes a moving radial basis function (MRBF) method to overcome these numerical challenges. The MRBF method employs moving center points rather than fixed ones to maintain consistency with particle size growth. Moreover, the shape parameter is designed as a time-varying parameter to adapt to the continuously evolving particle size distribution (PSD). By combining the MRBF approximation with the collocation method, the original PBE is transformed into a system of ordinary differential equations (ODEs), which are then solved using standard ODE solvers. The accuracy of the MRBF method is validated through various numerical cases and aerosol chamber data. The comparison results show that the MRBF method provides highly accurate predictions for the PSD evolution across all numerical cases. Relative deviations between the method predictions and the analytical solutions are on the order of 10⁻⁵ to 10⁻⁴. The MRBF method also demonstrates excellent agreement with the experimental data from aerosol chamber tests under different initial conditions. Additionally, this method achieves high computational efficiency, with all selected simulations completing within 1 s.</div></div>","PeriodicalId":407,"journal":{"name":"Powder Technology","volume":"473 ","pages":"Article 122206"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146185609","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel method using DEM–FEM and wavelet-based energy analysis for tumbling mill shell vibration signal generation and load feature extraction","authors":"Huaiwen Zou ,&nbsp;Xiaoli Wang ,&nbsp;Qingfei Xiao ,&nbsp;Jiayi Zhou ,&nbsp;Yifan Wang ,&nbsp;Stephen Pooley","doi":"10.1016/j.powtec.2026.122183","DOIUrl":"10.1016/j.powtec.2026.122183","url":null,"abstract":"<div><div>Grinding is a critical operation in mineral processing, whose load is closely related to grinding efficiency, energy consumption, and ore recovery. Therefore, accurate detection of mill load is essential for achieving optimal mill performance. Among various monitoring techniques, shell vibration signals constitute promising information that is highly sensitive to load variations and immune to interference from neighboring equipment. However, several challenges arise in industrial applications. Industrial mills must run continuously due to production and cost requirements, which hinders the design and adequate testing of the sensor to maintain its reliability. In addition, only small adjustments of the operating parameters are permitted, so it is difficult to obtain vibration data that cover a broad range of operating conditions. Moreover, the charge motion cannot be directly observed, which obscures the establishment of a correlation between the measured vibrations and the true load. To address these issues, a coupled discrete element method and finite element method (DEM–FEM) framework that produces shell vibration signals under different load conditions is developed. By using these simulated signals together with laboratory data, a feature-extraction workflow combining spectral analysis, wavelet transform, Hilbert-envelope analysis, and locally weighted scatterplot smoothing (Lowess) is designed to obtain robust impact-energy features. Experimental tests confirm a clear quantitative correlation between these features and the actual mill load. The proposed framework clarifies the mechanisms of shell-vibration formation and provides a solid basis for selecting effective vibration features, enabling reliable and non-intrusive load detection in industrial tumbling mills.</div></div>","PeriodicalId":407,"journal":{"name":"Powder Technology","volume":"473 ","pages":"Article 122183"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146185702","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Numerical analysis and multilayer perceptron prediction of proppant distribution in multi-cluster perforated horizontal wells","authors":"Wanzhuo Ping ,&nbsp;Mengmeng Zhou ,&nbsp;Wenzhaoyang Xie ,&nbsp;Xianzhi Song ,&nbsp;Zhengming Xu","doi":"10.1016/j.powtec.2026.122225","DOIUrl":"10.1016/j.powtec.2026.122225","url":null,"abstract":"<div><div>In multi-cluster perforation, the uniform distribution of proppant is crucial for fracture support and oil-gas production. However, in field operations, due to the significant accumulation of proppant in the wellbore, efficient distribution in multi-cluster perforation is challenging, and there has been a lack of quantitative methods to evaluate the proppant distribution uniformity. The Computational Fluid Dynamics (CFD) and Discrete Element Method (DEM) are coupled and used in this study, which is validated with experimental data, showing an error margin of less than 5%. This study reveals the phenomenon of recycle flow of proppant in horizontal wells with multi-cluster perforations and their distribution patterns. The findings indicate that under cycle flow, proppant forms reverse transport and collide with the proppant in the main flow direction. Some proppants are carried upward due to fluid entrainment, while others distribute downward under the combined effects of collision force, fluid resistance, and gravity. This leads to underfilled or unfilled perforation clusters, while localized accumulation occurs in the wellbore. In addition, this study proposes a novel method based on a multilayer perceptron (MLP) to simultaneously predict proppant transport efficiency in multi-cluster perforations by inputting normalized parameters. The dataset was divided in a 7:3 ratio for training, and the predicted results showed high consistency with actual data. The method quantifies the uniformity of proppant distribution, providing guidance for optimizing oil and gas production.</div></div>","PeriodicalId":407,"journal":{"name":"Powder Technology","volume":"473 ","pages":"Article 122225"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146185835","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Performance and backfilling process design of coal-based solid waste paste under conditions of constant temperature, humidity, and variable pressure","authors":"Haiyue Qin ,&nbsp;Fangtian Wang ,&nbsp;Denghong Chen ,&nbsp;Wenhua Hao ,&nbsp;Tiankuo Tang ,&nbsp;Chao Liu ,&nbsp;Xu Wang ,&nbsp;Chao Li ,&nbsp;Depei Lu","doi":"10.1016/j.powtec.2026.122208","DOIUrl":"10.1016/j.powtec.2026.122208","url":null,"abstract":"<div><div>To address the challenges of high-water-level subsidence, intense strata pressure in deep mining, and low mining-backfilling synergy efficiency in the coal mining areas of the Yellow River Basin, this study, with the backfilling of goafs at the Ningdong Base as the engineering context, conducted mix proportion experiments on paste backfill materials using coal-based solid waste suitable for variable-pressure environments. Quicklime (20%) was identified as the core activating component, with the optimal mix proportion determined as follows: paste concentration of 78%, and a ratio of coal gangue: fly ash: cement: quicklime = 5:8:3:4. The results indicate that the backfill material exhibits satisfactory fluidity, with an average slump of 74.6 mm and spread diameter of 175.8 mm, meeting the archaeological requirements for long-distance pipeline transportation underground. Its 7-day unconfined compressive strength (UCS) exceeds 4.0 MPa, ensuring timely load-bearing capacity to effectively control early-stage deformation of the stope roof. Furthermore, after curing under 3.0 MPa pressure, the later-age strength of the backfill reached 6.74 MPa, representing a 57% increase over the strength achieved under standard curing. This significant enhancement confirms its superior adaptability to dynamic strata loads and long-term stability in deep mining environments. Additionally, numerical simulation was employed to optimize the mining-backfilling process. The simulation clarified that the interaction between the backfill ratio and backfilling step distance is the key factor controlling roof subsidence and the distribution of front abutment pressure. Directly applicable optimal backfilling process parameters were proposed: a backfill strength of 2.0 MPa, a“three-mining-sequences followed by one backfilling” pattern, and a backfill ratio of 70%. This scheme can effectively control roof subsidence and stress concentration, providing crucial process support for achieving stability in the stope surrounding rock. This study leverages the advantages of on-site material sourcing and cost-effectiveness, offering the mining area an integrated solution for solid waste utilization and green mining that is technically feasible, economically viable, and environmentally compatible.</div></div>","PeriodicalId":407,"journal":{"name":"Powder Technology","volume":"473 ","pages":"Article 122208"},"PeriodicalIF":4.6,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146185363","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}