5 results about "Steam coal" patented technology

This invention discloses a method for training and optimizing a large-scale model in the thermal coal industry based on knowledge distillation, belonging to the field of large-scale model training and optimization technology. The method includes the following steps: collecting and preprocessing multimodal industry data to obtain a data matrix; inputting the data into a teacher's visual Transformer network to generate a sparse probability label matrix; constructing an improved TinyViT student network; embedding multiple optimization units in self-attention blocks; completing weight initialization; performing pre-training distillation to generate a composite distillation signal; inputting the signal into the student network for fine-tuning training to obtain a set of fine-tuned student weights; performing dynamic weighted training on hard samples; optimizing the training effect on long-tailed samples; and obtaining a lightweight model through quantized perceptual training and channel pruning. The model is then deployed to edge computing nodes for real-time inference. This invention improves the training efficiency of large-scale models in the thermal coal industry while optimizing the model's inference speed and accuracy, and can be effectively deployed in edge computing environments.

The application provides a coal power adjustable capacity prediction method. The method comprises the following steps: establishing a prediction scene set of typical power coal variety prices to obtain predicted prices of the typical power coal varieties; establishing an objective function of minimizing the coal blending and burning power generation cost, and establishing a coal blending optimization model according to the power coal consumption constraint, the boiler calorific value, the boiler ash content and the boiler sulfur content constraint, solving the coal blending optimization model to obtain an optimal coal blending strategy; establishing a linear regression model of the blocked capacity of the coal power unit during the peak output period and the power coal quality according to the historical data of the adjustable capacity and the power coal quality of the coal power unit in the power system during the peak output period; predicting the corresponding blocked capacity through the linear regression model; calculating the coal power adjustable capacity of each coal power unit in the power system to obtain the adjustable capacity coefficient of the power system, and evaluating the coal power generation capacity according to the adjustable capacity coefficient.

This invention discloses an optimized scheduling method for a power system incorporating biomass indirect co-firing units, belonging to the field of power system optimization. The method includes: establishing a power system scheduling model with the objective function of minimizing the sum of power generation cost, coal-fired unit start-up cost, carbon trading cost, and curtailment penalty; and using the following optimization variables: the quality of thermal coal input to the boiler of each coal-fired unit at each time, the quality of biomass input to the pyrolysis furnace of each coal-fired unit at each time, the start-up status of each coal-fired unit, the net carbon dioxide emissions of each coal-fired unit at each time, and the renewable energy curtailment of each renewable energy unit at each time. The optimized scheduling plan is obtained by solving the power system scheduling model under preset constraints. These preset constraints include the operational constraints of the pyrolysis furnace, boiler, and turbine within the coal-fired units. This invention can simultaneously improve the renewable energy absorption capacity and reduce carbon emissions while satisfying the operational constraints of the pyrolysis furnace, boiler, and turbine within the coal-fired units.

This invention discloses a coal resource allocation calculation model and method in the field of coal resource planning and allocation, aiming to solve the technical problems of inaccurate coal resource allocation calculation, poor coordination, low matching efficiency, and lagging revenue accounting in existing coal resource allocation calculations. The model achieves accurate calculation of the raw coal allocation and washing decision-making module for the Dingwugeng coal group, the allocation module for coal from the four major coal preparation plants, and the production-sales coordination and dynamic optimization module. Combined with data acquisition and preprocessing, dynamic ratio adjustment, vector matching, and profit calculation algorithms, it realizes accurate calculation of the raw coal washing volume and thermal coal allocation volume for the Dingwugeng coal group, the coal allocation volume for the four major coal preparation plants, and the coal supply matching volume between the mine and the coal preparation plant. This invention reduces the calculation error by more than 30%, increases resource utilization by 15%-20%, and improves economic benefits by 10%-15%, making it suitable for refined and dynamic resource allocation and production decision-making in coal enterprises.