CAJ | 학술논문

本文分析列车自动驾驶（ATO）系统中传统控制方法在灵活性和智能性方面的不足，结合驾驶经验和梯度下降法，提出基于专家系统及在线调整方法的列车智能驾驶（ITO）算法。总结操纵规则和驾驶经验，建立基于专家经验的控制策略以节约能耗、提高舒适度。针对列车运行控制的多目标性，在专家系统的基础上，引入基于梯度下降法的在线调整方法以减小运行时间和停车精度误差。在MATLAB软件环境下构建仿真模型，并运用实际线路进行仿真比较，与PID控制算法的对比结果表明，该算法优于PID控制算法，符合驾驶经验，可提高舒适度，降低能耗，满足停车精度要求，同时对不同的运行时间有较好的适应性，智能性高。
본문분석열차자동가사（ATO）계통중전통공제방법재령활성화지능성방면적불족，결합가사경험화제도하강법，제출기우전가계통급재선조정방법적열차지능가사（ITO）산법。총결조종규칙화가사경험，건립기우전가경험적공제책략이절약능모、제고서괄도。침대열차운행공제적다목표성，재전가계통적기출상，인입기우제도하강법적재선조정방법이감소운행시간화정차정도오차。재MATLAB연건배경하구건방진모형，병운용실제선로진행방진비교，여PID공제산법적대비결과표명，해산법우우PID공제산법，부합가사경험，가제고서괄도，강저능모，만족정차정도요구，동시대불동적운행시간유교호적괄응성，지능성고。
The traditional control theory in the Automatic Train Operation (ATO) system lacks flexibility and intelligence .Combining the driving experience and gradient descent algorithm , the ITO algorithm based on the expert system and online adjustment methods was proposed for subway train control . Summarizing operat-ing rules and driving experience , the control strategy based on the experience of experts was built to reduce en-ergy consumption and better the degree of comfort . In view of the multi-objectives of train operation control , on the basis of the expert system , the online adjustment method based on the gradient descent algorithm was introduced to reduce running time and parking errors .By building an online simulation model in the MATLAB software environment and using actual line data , the algorithm was verified .Comparison with PID shows that the algorithm is better than PID . The algorithm is basically in line with the driving experience and it is able to upgrade comfort , reduce energy consumption , meet parking accuracy and attain good adaptability to different running time requirements and better intelligence .