Recognizing and distinguishing coal and gangue are essential in engineering, such as in coal-fired power plants. This paper employed a convolutional neural network (CNN) to recognize coal and gangue images and help segregate coal and gangue. A typical workflow for CNN image recognition is presented as well as a strategy for updating the model parameters. Based on a powerful trained image recognition model, VGG16, the idea of transfer learning was introduced to build a custom CNN model to solve the problems of massive trainable parameters and limited computing power linked to the building of a brand-new model from scratch. Two hundred and forty coal and gangue images were collected in a database, including 100 training images and 20 validation images for each material. A recognition accuracy of 82.5% was obtained for the validation images, which demonstrated a decent performance of our model. According to the analysis of parameter updating in the training process, a principal constraint for obtaining a higher recognition accuracy mainly resided in a shortage of training samples. This model was also used to identify photos from a washing plant stockpiles, which verified its capability of dealing with field pictures. CNN combined with the transfer learning method we used can provide fast and robust coal/gangue distinction that does not require harsh data support and equipment support. This method will exhibit brighter prospects in engineering if the target image database (as with the coal and gangue images in this study) can be further enlarged.

This paper explores advanced machine learning approaches to enhance the prediction accuracy of compressive strength (CoS) in geopolymer composites (GePC). Geopolymers, as sustainable alternatives to Ordinary Portland Cement (OPC), offer significant environmental benefits by utilizing industrial by-products such as fly ash and ground granulated blast furnace slag (GGBS). The accurate prediction of their compressive strength is crucial for optimizing their mix design and reducing experimental efforts. We present a comparative analysis of two hybrid models, Harris Hawks Optimization with Random Forest (HHO-RF) and Sine Cosine Algorithm with Random Forest (SCA-RF), against traditional regression methods and classical models like the Extreme Learning Machine (ELM), General Regression Neural Network (GRNN), and Radial Basis Function (RBF). Using a comprehensive dataset derived from various scientific publications, we focus on key input variables including the fine aggregate, GGBS, fly ash, sodium hydroxide (NaOH) molarity, and others. Our results indicate that the SCA-RF model achieved a superior performance with a root mean square error (RMSE) of 1.562 and a coefficient of determination (R²) of 0.987, compared to the HHO-RF model, which obtained an RMSE of 1.742 and an R² of 0.982. Both hybrid models significantly outperformed traditional methods, demonstrating their higher accuracy and reliability in predicting the compressive strength of GePC. This research underscores the potential of hybrid machine learning models in advancing sustainable construction materials through precise predictive modeling, paving the way for more environmentally friendly and efficient construction practices.

Tailings ponds are the indispensable facilities in the mine production and operation. Once the dam is destabilized and damaged, it will pose a serious threat on the life and property of the downstream population and could also potentially cause an environmental disaster. With an engineering background, this paper dynamically and numerically simulates the evolution process of tailings flow from dam failure and the influence scope of any resulting disaster in context. The evolution characteristics of leaked tailings flow are analyzed at various downstream riverbed slopes and debris blocking dam settings. In addition, parameters such as flow rate, impact force and deposition range of leaked tailings flow at downstream arrival are studied, as well as their correlations. The results indicate that the flat terrains upstream and downstream of passage zone show a relatively larger area of inundation by tailings flow. Both the maximum and final downstream inundated ranges increase with the elevating slope of downstream riverbed, and the leaked tailings are deposited mainly in the nearby villages in front of the dam and the flat terrains of the downstream passage zone. Additionally, rational establishment of debris blocking dams on the downstream side is effective in diminishing the damage of tailings flow to the downstream section. This study can also provide an important basis for the quantitative evaluation of post-disaster influence scope for tailings pond as well as for the design of dam body.

A number of large underground oil storage spaces will be constructed in deep salt mines in China in the coming years. According to the general geological survey, the first salt cavern oil storage base of China is planned to be built in Jintan salt mine. In this research, the geological feasibility of the salt mine for oil storage is identified in detail as follows. (1) The characteristics of regional structure, strata sediment, and impermeable layer distribution of Jintan salt mine were evaluated and analyzed. (2) The tightness of cap rock was evaluated in reviews of macroscopic geology and microscopic measuring. (3) According to the geological characteristics of Jintan salt mine, the specific targeted formation for building underground oil storage was chosen, and the sealing of nonsalt interlayers was evaluated. (4) Based on the sonar measuring results of the salt caverns, the characteristics of solution mining salt caverns were analyzed. In addition, the preferred way of underground oil storage construction was determined. (5) Finally, the results of closed well observation in solution mining salt caverns were assessed. The research results indicated that Jintan salt mine has the basic geological conditions for building large-scale underground oil storage.