<p>An accurate forecast of power generation by different renewable sources is important for efficient scheduling. Previous research has mainly focused on forecasting individual energy sources while neglecting the interconnections between them; they have not been able to simultaneously accurately forecast all energy sources. Hence, this paper introduces a new Hybrid Sequential Convolutional Dual Drift Adaptive-based Support Vector model for power generation prediction, namely solar photovoltaic, solar thermal, and wind power. The process starts with the collection of overall time series data, encompassing both the power generation factors and corresponding weather conditions. This information is pre-processed and strongly feature-extracted to determine major patterns, such as (i) energy correlations with a sequential convolutional neural network, (ii) nonlinear temporal extraction using a dual drift adaptive long short-term memory, and (iii) linear temporal characteristics using an adaptive support vector regression. The framework employs extensive data preparation to make the model more robust, and the outputs are concatenated and passed through a fully connected layer to improve predictive accuracy. To validate the efficiency, various advanced prediction models are compared. Experimental outcomes show that the proposed model improved the power generation predictions of Solar photovoltaic, with the coefficient of determination value of 0.991, 0.946, and 0.973, respectively, for both Solar photovoltaic and Solar Thermal as well as Wind Power, attesting to the accuracy of the model.</p>