In order to solve the problems of the method involving the optimization of the traditional combustion chamber structure, which has a long computation cycle, high computation cost, and can easily fall into the local optimal solution, this paper refers to the concept of a fuzzy neural network in machine learning. This study proposes a method of combustion chamber structure optimization that uses a fuzzy neural network to prejudge the results of the fitness function before calculating it in order to reduce the periodicity of computation and improve computational accuracy. The validation results show that the combustion chamber structure optimization method proposed in this paper can effectively reduce the computational cost under the premise of guaranteeing optimization accuracy. Using the test function, compared with the traditional genetic algorithm, the average number of iterations at convergence is reduced by 28.59%, and the average number of calculations of the fitness function is reduced by 25.59%. When optimizing the combustion chamber structure, the peak pressure of the optimal combustion chamber structure is increased by 10.32%, the computational count is reduced by 23.33%, and the time consumed is reduced by 23.91% compared with the traditional genetic algorithm.
This paper's license is marked as closed access or non-commercial and cannot be viewed on ResearchHub. Visit the paper's external site.
Connect with your self-custody wallet
Connect with your Coinbase account