This paper introduces and investigates novel hybrid deep learning models for solar power forecasting using time series data. The study analyzes the efficacy of various models, including combinations of convolutional neural networks (CNN), long short-term memory (LSTM), and transformer (TF) models, in capturing complex patterns in solar power data. The research compares these hybrid models with single CNN, LSTM, and TF models using different optimizers. Three evaluation metrics—mean absolute error (MAE), mean squared error (MSE), and root mean square error (RMSE)—are employed to assess performance. The results show that the CNN–LSTM–TF hybrid model, using the Nadam optimizer, outperforms other models with an MAE of 0.551%. In contrast, the TF–LSTM model performs poorly with an MAE of 16.17%, highlighting the challenges in making reliable predictions for solar power. This study provides valuable insights for optimizing and planning renewable energy systems, emphasizing the importance of selecting appropriate models and optimizers for accurate solar power forecasting. The research also marks the first time transformer networks have been used in hybrid models for solar power forecasting, contributing significantly to the field.This paper introduces and investigates novel hybrid deep learning models for solar power forecasting using time series data. The study analyzes the efficacy of various models, including combinations of convolutional neural networks (CNN), long short-term memory (LSTM), and transformer (TF) models, in capturing complex patterns in solar power data. The research compares these hybrid models with single CNN, LSTM, and TF models using different optimizers. Three evaluation metrics—mean absolute error (MAE), mean squared error (MSE), and root mean square error (RMSE)—are employed to assess performance. The results show that the CNN–LSTM–TF hybrid model, using the Nadam optimizer, outperforms other models with an MAE of 0.551%. In contrast, the TF–LSTM model performs poorly with an MAE of 16.17%, highlighting the challenges in making reliable predictions for solar power. This study provides valuable insights for optimizing and planning renewable energy systems, emphasizing the importance of selecting appropriate models and optimizers for accurate solar power forecasting. The research also marks the first time transformer networks have been used in hybrid models for solar power forecasting, contributing significantly to the field.