基于改进多目标粒子群算法的航空器改航研究

针对影响航行安全的动态危险天气,提出了一种基于改进多目标粒子群算法的改航路径规划方法。首先,获取实时动态气象数据,利用栅格法对改航环境进行建模并采集危险天气区域初始边界点的历史气象数据。然后采用灰色预测模型对上述初始边界点坐标进行位置预测,进而建立改进后的实时动态环境模型。最后利用改进环境模型的多目标粒子群算法对改航路径进行动态规划。在考虑改航路径角度和距离等约束条件的基础上,确定了改航路径危险系数和距离最优的双目标函数。对中东部沿海某次短时危险天气下的航空器改航进行仿真分析,仿真结果表明改进后算法具有一定的有效性和可行性。

On the aircraft flying-goal diversion based on the improved multi-objective particle swarm optimization

In view of the dynamically dangerous weather which may affect the flying safety,this paper aims to find a way for predicting and forecasting the location of the dynamically dangerous weather area and find an alternative route for the aircraft. For this purpose,it would be necessary to acquire the real-time dynamic meteorological information and data by collecting the historical meteorological data of the initial boundary points in the dangerous weather areas under study to modulate the alternative navigation environment concerned. And,then,it is necessary to predict and lay out the coordinates of the initial boundary points mentioned above by using a grey prediction model. It is only by building up such an improved real-time dynamic environment model and applying it to the particle swarm optimization goal that would be possible for the pilot to build up the coordinates of the above mentioned initial boundary points. And,secondly,considering the actual situation of the aircraft re-routing,so as to make the aircraft safely and efficiently implement their own re-routing process,it is necessary to reconsider the constraints,such as the angle and distance of the re-routing pathways respectively,and overcome the risk factors of the re-routing pathway and the distance between the re-routing pathway and the starting point objective function. Thus,finally,we use the improved multi-objective particle swarm optimization( MPSO) algorithm to simulate and analyze the aircraft rerouting under a short-term dangerous weather in the eastern and central coast of China,and get the aircraft rerouting path under this algorithm,then,we compare it with traditional methods. The said improved algorithm should be efficient enough to forecast and predict the dangerous weather areas in the course of rerouting by taking the actual changes of the dangerous weather area into consideration. Practically speaking,the dynamic environment model has been built up with high accuracy and feasibility to make the prediction more closely in line with the demands of the actual operation in the process of rerouting it for dangerous weather area. And,therefore,the algorithm has been made on the actual basis fit for the bi-objective function for the aircraft in the course of re-routing. In addition,the algorithm can converge to Pareto optimal boundary promptly and steadily,so as to help to find the solution set of the optimal diversion approaches in line with the actual situation of the aircraft diversion. Thus,in comparison with the other traditional alternatives,the algorithm we have chosen is in a position to be advantageous in lots of application values in the air transportation and traffic management.