Neural Network Prediction of Locomotive Engine Parameters Based on the Dung Beetle Optimization Algorithm and...

Area of Science:

• Mechanical Engineering
• Artificial Intelligence
• Combustion Engines

Background:

• Altitude significantly affects diesel engine power and emissions, particularly for plateau dual-source locomotives.
• Optimizing engine performance under varying altitudes is crucial for efficiency and environmental impact.
• Existing optimization methods may not fully address the complexities of high-altitude engine operation.

Purpose of the Study:

• To analyze and optimize the performance of diesel engines for plateau locomotives operating at different altitudes.
• To develop an intelligent optimization framework combining neural networks and multi-objective algorithms.
• To validate the effectiveness of AI in improving diesel engine performance under high-altitude conditions.

Main Methods:

• Simulated diesel engine performance using GT-Power to generate a dataset.
• Employed a random sampling method to create 400 operating points for model training.
• Utilized a DBO (Dolphin Bay Optimization) algorithm to optimize neural network prediction models.
• Applied the NSGA-II (Non-dominated Sorting Genetic Algorithm II) for multi-objective optimization.

Main Results:

• Neural network prediction models optimized by the DBO algorithm achieved correlation coefficients exceeding 95%.
• The NSGA-II algorithm successfully performed multi-objective optimization of engine parameters.
• The intelligent optimization approach demonstrated significant improvements in diesel engine performance across various altitudes.

Conclusions:

• Artificial intelligence optimization algorithms are effective in enhancing diesel engine performance at different altitudes.
• The proposed method offers a viable solution for optimizing plateau locomotive engines.
• This research highlights the potential of AI in addressing challenges in internal combustion engine performance optimization.