Search

A single particle fusion combustion model considering the effect of ash fusion is established. The comparison of the predicted data with experimental data reported in the literature validates the model. The accuracy of the prediction results is improved with the model. The fusion combustion characteristics of the single particle are investigated. Results show that the char could not convert to ash completely due to the pore space is taken up by the fused ash, thereby resisting the further diffusion of oxygen at 1673–1873 K. The burnout rate decreases linearly with the particle density and the attenuation slope increases with the temperature. The burnout rate decreases from nearly 100% to about 88% as the particle size increases from 1 mm to 10 mm. The fused ash decreases the char combustion rate. The effect increases with particle density, particle size as well as temperature. The total combustion time decreases from nearly 19 s to about 3 s as the surface-area-to-volume ratio increases from 0.6 to 6 mm−1.

Short-term load forecasts are important tools for electrical utilities to balance electricity supply and demand. The exponential increase in behind-the-meter solar panel installation in California has made it more difficult to accurately predict electrical load. This study developed three forecast models based on multiple linear regression, random forest, and gradient boosting that incorporated solar capacity to predict hourly load in southern California 24-h in advance. Air temperature was the most important meteorological variable and holiday, month, solar capacity and the load from the previous week were the most important non-meteorological variables. All three models were more accurate when load was lower, such as early morning, late at night and during the winter. In contrast, all models had larger errors during the middle of the day and in summer when load was higher. The mean error, based on the forecasts for each hour of the day, was 3.5% for the random forest, 3.4% for the multiple linear regression, and 3.1% for the gradient boosting. Overall, the gradient boosting model performed the best and had errors