An Improved Strength Pareto Evolutionary Algorithm 2 for Constrained Multiobjective Optimization Problems in Continuous Annealing Furnace

This research proposes a constraint multiobjective optimization algorithm that combines the auxiliary task method with the strength Pareto evolutionary algorithm 2 (ATM‐SPEA2). The algorithm is designed to optimize the heating process in the continuous annealing furnace (CAF), ensuring that the annealing temperature meets the process requirements, while also minimizing energy consumption and maximizing production capacity. First, a multiobjective production optimization model for the CAF is established grounded in thermodynamic theory. Subsequently, to address the limitations of traditional constrained multiobjective optimization algorithms that are prone to local optima and uneven distributions, the ATM‐SPEA2 algorithm is employed, which dynamically expands the search area for Pareto nondominated solutions through an auxiliary task to aid in finding the optimal solution. Finally, the ATM‐SPEA2 algorithm is applied to the production optimization of the CAF for adjusting parameters such as furnace temperature, strip speed, and gas flow to form the most effective combination of process parameters. Experimental results show that the proposed algorithm outperforms traditional constrained multiobjective optimization algorithms, and offers significant guidance for actual production with its optimized parameters.