Construction of an Electric Energy Prediction Model Based on Improved IHHO-RF Algorithm

With the rapid development of artificial intelligence technology, machine learning, as its core branch, can effectively solve prediction problems by learning and extracting patterns from large amounts of data. Algorithms play a key role in this process. AI technologies will show great potential in a wider range of application scenarios in the future. Therefore, an improved Harris Eagle Algorithm (IHHO) is proposed, which is combined with the construction of a coupling algorithm based on random forest model (IHHO-RF) and applied to predict the power output of cyclic power plants. Firstly, on the original Harris Eagle algorithm, a Harris Eagle algorithm based on the best point set theory, stochastic shrinkage index and adaptive inertia weight was proposed, and 10 Benchmark functions were used to test the effectiveness of the algorithm. Secondly, the IHHO algorithm is combined with random forest to form a parallel ensemble algorithm, and it is applied to the construction of the prediction model of the power output of the cyclic power plant. The results show that the prediction accuracy of the improved IHHO-RF model has significant advantages compared with the HHO-RF and other models, and the parallel ensemble algorithm constructed with the random forest model can achieve more accurate prediction tasks and has the ability to be widely promoted.