Forecasting of İstanbul Solar Power Using Dendritic Neuron Model Artificial Neural Network

Artificial neural networks are widely used in solving many problems and provide successful results. These models, which are frequently encountered especially in estimation problems, attract the attention of researchers with the high accuracy rates they achieve. The dendritic neuron model artificial neural network, which is formed by adding the dendrites in the biological neuron to the artificial neural network model, draws attention with its better performance compared to other artificial neural network methods used in time series analysis. In this study, the dendritic neuron model artificial neural network was preferred due to its superior performance characteristics, and the particle swarm optimization algorithm was used for training the dendritic neuron model artificial neural network. To evaluate the forecasting performance of this preferred artificial neural network, 15-minute daily data belonging to the 1200 kWp solar power plant located in the İkitelli Drinking Water Treatment Plant of the Istanbul Water and Sewerage Administration was used and performance comparisons were made with many artificial neural network models in the literature. It was observed that the performance of this tested method had the best performance among the other methods.