我国西部地区诸多矿口至干线间存在几十公里的高落差、大风沙区域，为提高货运空轨列车在该种复杂环境中的适应性与智能性，研究了适用的运行轨迹优化方法；分析了货运空轨的动态运行过程，考虑线路落差大、线型复杂多变，以及列车易受环境干扰偏离计划等特点，提出了一种改进的复杂线路列车运行曲线生成方法，以时间-能耗综合最优为目标，利用改进多目标差分进化算法对操纵模型求解；在得到Pareto最优解集的基础上，建立干扰场景下的整体调整模型，及时避免或消减延误的恶化与传播，最后采用某试验线数据进行仿真验证；研究结果表明：与标准操纵方法相比，改进操纵方法下净能耗的下限得到优化，在某虚拟区间内改进操纵较标准操纵牵引能耗降低84.13%，制动回收减少43.15%，净能耗相差5.20 kW·h，在试验线多区间干扰场景中净能耗相差41.82 kW·h，能够解决货运空轨列车在复杂环境下的轨迹优化问题，有效保障运输计划兑现率并降低运营成本。

There are several tens of kilometers of high drop and large sandstorm areas between many mines and main roads in western China. In order to improve the adaptability and intelligence of the freight suspended monorail train in this complex environment, an applicable trajectory optimization method was studied. The dynamic operation process of freight rail was analyzed. Considering the characteristics of large line drop, complex line type, and train deviation from plan due to environmental interference, an improved method for generating long ramp train operation curve was designed with the goal of comprehensive optimization of time and energy consumption. On this basis, the overall adjustment model in the interference scene was established to avoid or reduce the deterioration and propagation of delay in time. Finally, the data of a test line was used for simulation verification. The results show that: Compared with the standard control method, the lower limit of net energy consumption under the improved control method is optimized. In a virtual interval, the traction energy consumption of the improved control method is reduced by 84.13%, the brake recovery is reduced by 43.15%, the net energy consumption difference is 5.20 kW·h, and the net energy consumption difference is 41.82 kW·h in the multi-interval interference scene of the test line. It can solve the trajectory optimization problem of freight suspended monorail train in complex environment, effectively guarantee the redemption rate of transportation plan and reduce operating costs.