Intelligent Air Traffic Flow Optimization and Allocation Algorithms

This article aims to study a new type of intelligent air traffic flow optimization and allocation algorithm, which improves the existing flow management mode by introducing artificial intelligence technology. Firstly, this article analyzes the main problems in current air traffic flow management, and then proposes a flow optimization algorithm framework called LDDO (LSTM-DDPG Optimizer) algorithm based on deep deterministic policy gradient (DDPG) and long short-term memory (LSTM) network, aiming to achieve dynamic optimization and real-time allocation of air traffic flow. The effectiveness of the proposed algorithm in improving airspace utilization efficiency and reducing flight delay time was verified in the experimental stage. The results of this study not only provide a new flow management tool for aviation traffic management institutions, but also provide theoretical and technical support for further research on intelligent aviation traffic management systems. In the experimental stage, this study designed four experiments to evaluate the application effects of Optimization Theory Algorithm (OTA), Machine Learning Algorithm (MLA), and research algorithms in aviation traffic management. In the flight punctuality experiment, the LDDO algorithm achieved an on-time rate of 85%. In the experiment of reducing airspace congestion index, the LDDO algorithm reduced the congestion index to 80. In the response time efficiency evaluation experiment, the average response time of the LDDO algorithm was 140 seconds. In the experiment of reducing flight delay time, the LDDO algorithm reduced the average delay time by 20 minutes. From the above data conclusion, it can be seen that the LDDO algorithm has good performance in intelligent air traffic flow optimization.