Prediction of Compressive Strength of Lime-Fly Ash Stabilised Coal Gangue Mixture Using an Improved SVM Based on the ESHO Algorithm

The continuous development of artificial intelligence has brought new research directions to the field of new materials. The strength of ordinary coal gangue solid waste itself is relatively low, and the experimental methods for determining the compressive strength of different coal gangue mixtures consume a large amount of time and resources. Therefore, the research proposed a prediction model for the compressive strength of coal gangue mixture that combines the improved hippocampus optimization algorithm with the support vector machine. This model improves the Seahorse optimization algorithm by using the Tent mapping method and the sine and cosine optimization algorithm, and then uses the improved seahorse optimization algorithm to optimize the support vector machine and construct the corresponding index system. Experiments show that the maximum prediction accuracy of the fusion model is 99.2%, which is 1.4% higher than that of the second-best least squares support vector machine. After 40 days of curing, the prediction result is slightly larger than the actual compressive strength, and the average prediction error is 0.002. At the same time, improve computing efficiency. From this, it can be concluded that the method proposed by the research institute can effectively predict the compressive strength of coal gangue mixture, improve the utilization rate of coal gangue solid waste, and reduce the environmental pollution caused by the accumulation of coal gangue.