<p>The integration of photovoltaic rooftops into industrial buildings is a key pathway to achieving carbon neutrality goals. The core challenge lies in maximizing photovoltaic power generation efficiency. Traditional Maximum Power Point Tracking methods often fall into local optima and lack dynamic responsiveness in complex factory environments. Existing prediction approaches suffer from issues such as power oscillation and slow tracking speed. To address these challenges, this study proposes a Maximum Power Point Tracking control model for photovoltaic rooftop integration, which combines a Hybrid Particle Swarm Optimization algorithm with an Improved Whale Optimization Algorithm. The model continuously calculates real-time carbon footprints and verifies the optimal operating point that meets both energy efficiency requirements and low-carbon standards. Experimental results show that the global maximum power point tracking time of the proposed model under dynamic local shadow condition is only 0.37–0.41&#xa0;s, the average accuracy rate is 95.67%, the value of the area under the curve is 0.93, the F1 score is higher than 80.26%, and the power output characteristics are smoother and more stable in different seasons. The above results show that the model has excellent accuracy and applicability in the integrated design of photovoltaic roof in complex industrial environment, effectively solves the failure problem of maximum power point tracking in dynamic environment, provides a new idea for intelligent management and regulation of energy system in plant, and significantly improves energy utilization efficiency and system operation stability.</p>