Design and study of mine silo drainage method based on fuzzy control and Avoiding Peak Filling Valley strategy.

Coal is a non-renewable fossil energy source on which humanity relies heavily, and producing one ton of raw coal requires the discharge of 2-7 tons of mine water from the ground. The huge drainage task increases the cost of coal mining in coal mines significantly, so saving the drainage cost while guaranteeing the safe production of coal mines is a problem that needs to be solved urgently. Most of the fuzzy controllers used in the traditional dynamic planning methods applied to mine drainage are two-dimensional fuzzy controllers with limited control effect, so the traditional two-dimensional fuzzy controllers are improved by introducing the rate of change of gushing water to form a three-dimensional fuzzy controller with three-dimensional control of instantaneous section-water level-rate of change of gushing water, and at the same time, the optimized dynamic planning method is designed by combining the Avoiding Peak Filling Valley strategy and the optimal dy-namic planning method is used in conjunction with the un-optimized dynamic planning method. The optimized dynamic planning method is applied to the same coal mine water silo gushing water experiments; experimental comparison found that the pumping station system before the optimi-zation of the electricity consumed is 52,586 yuan/day, while after the optimization of the electricity consumed is reduced to 41,692 yuan/day, the cost per day consumed compared with the previous reduction of 20.69%, a year can be saved 3,969,730 yuan. Therefore, the mine water bin drainage method based on fuzzy control and Avoiding Peak Filling Valley strategy proposed in this paper can be used as an improvement method of the existing mine drainage method, which can further ex-pand the economic benefits of coal mines and realize safe production while realizing cost savings.

Study on reasonable anchorage length based on failure mechanism of the bolt anchorage system.

In addition to analysing the mechanism of failure of the prestressed rock anchor anchor system and investigating the appropriate depth for fixing the rock anchors, theoretical equations were derived to calculate the rock anchors' axial force, ultimate capacity, and the interfacial shear force in the elastic phase. These equations are then used to analyse the pressure distribution within the rock bolt anchorage section and to investigate the effect of interfacial shear strength, shear stiffness, and anchorage length on interface failure. Drawing on the findings from both field-based rock bolt pull-out tests and numerical simulations, analyzed the failure mechanism of the anchor system, and proposed a reasonable anchor length design method for rock bolt. The results show that there is a strong dependence between ultimate load carrying capability of rock bolts and interfacial shear stress and shear rigidity, and that increasing the anchorage length and reducing the interface shear stiffness can avoid the stress concentration phenomenon. The primary factor leading to the anchor system failure is secondary interface failure. The evolution law of interface damage is that the damage occurs first at the initial position. As the interface damage location changes, the peak shearing stress moves towards the bottom of the anchored section. The engineering application results verified the feasibility of a reasonable anchorage length calculation method and rock bolt design process. The findings of this paper can be used as a basic reference for determining rock bolt anchorage support parameters during the design and construction of underground engineering projects.

Research on prediction of compressive strength of fly ash and slag mixed concrete based on machine learning.

Every year, a large amount of solid waste such as fly ash and slag is generated worldwide. If these solid wastes are used in concrete mixes to make concrete, it can effectively save resources and protect the environment. The compressive strength of concrete is an essential indicator for testing its quality, and its prediction is affected by many factors. It is difficult to predict its strength accurately. Therefore, based on the current popular machine learning supervised learning algorithms: Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Support Vector Machine (SVR), three models established a nonlinear mapping between multi-factor features and target feature concrete compressive strength. Using the three completed training models, we validated the test set with 206 example sets, and the Root Mean Square Error (RMSE), fitting coefficient (R2), and Mean Absolute Error (MAE) were used as evaluation metrics. The validation results showed that the values of RMSE, R2, and MAE for the RF model were 0.1, 0.9, and 0.21, respectively; the values of XGBoost model were 0.05, 0.95, and 0.15, respectively. The values of SVR were 0.15, 0.86, and 0.3, respectively. As a result, Extreme Gradient Boosting (XGBoost) has better generalization ability and prediction accuracy than the other two algorithms.