Energy-Efficient Optimization Driving Method For Trains In Urban Rail Transit

Urban Rail Transit has become the major part of Urban Rail Transportation (URT) System with its safe and comfortable feature as well as its mass rapid transportation capability. Based on the statistics from China association of URT, there are over15cities have urban rail transit system, and the total number of operation lines exceed63until the end of August2012. The operating mileage reaches1777kilometers. Therefore, reducing energy consumption is a very important problem for sustainable development of URT.In order to reduce energy consumption of URT trains, in this thesis, we research on energy-efficient optimization driving methods for trains within two successive stations. First, the possible energy-efficient control modes on different grades and the number of control modes switches are proved based on the maximum principle. Then a geometrical analysis based energy-efficient driving trajectory analysis method is proposed to analyze the control switching sequence. Through the trajectory analysis for different grades, general energy-efficient driving control sequence and running status sequence are proposed. Based on these, energy-efficient optimization methods for single train and two tracking trains are proposed, respectively. In our method, the energy-efficient driving trajectory for each gradient is determined by3or4decision parameters. Since the length of the distance between two successive stations is very short and the gradients are limited, all the optimization driving models we built belong to small-scale nonlinear programming problem, thus, it is particularly suitable for solving by sequence quadratic programming (SQP) algorithm. The simulation studies demonstrate that the energy consumption could be reduced and the solving time is very short, which means our new methods are not only suitable for energy-efficient trajectory's off-line per-planning, but also suitable for energy-efficient trajectory's online updating. The innovations of the dissertation are as follows:1. Based on the maximum principle, the possible energy-efficient control modes on different grades and the number of control modes switches are proved.2. Aiming to determine train energy-efficient control sequence, a geometrical analysis based train energy-efficient driving trajectory analysis method is proposed. In this new method, new energy consumption models for different gradients are designed in which only one variable is related to energy consumption. Thus, it is easy to determine the energy-efficient trajectory and the corresponding control sequence on each gradient.3. Aiming to depict energy-efficient control modes and running status with a unified form, general energy-efficient driving control sequence and running status sequence are proposed. The general energy-efficient driving control sequence is composed by5control modes in order, and the general energy-efficient running status sequence is composed by2runing status (speed changing or holding) in order.4. Aiming to update the train's energy-efficient trajectory online, a novel energy-efficient optimization driving method for a single train is proposed. In this method, energy consumption and running time ans distance can be easily formulated based on the general energy-efficient running status sequence, therefore, the energy-efficient driving optimization problem is built to a small-scale nonlinear programming problem which could be solved by SQP algorithm quickly.5. The process of two trains'tracking between two stations is unclear. Aiming to solve this problem, the tracking dynamic process and the energy saving space for the following train are deeply analyzed with the consideration of the leading train's running information; energy-efficient driving scenarios for two trains'tacking are proposed.6. Aiming to saving energy with the guarantee of running time, a novel energy-efficient driving optimization method for two tracking trains are proposed. The tracking process of the following train can be seen as a switching process among the energy-efficient driving scenarios. Then these scenarios are classified into4energy-efficient optimization problems, and the general energy-efficient control sequence and running status sequence are improved according to the different problems. After that, the whole tracking optimization process is modeled by four small-scale nonlinear programming problems according to different situations. It is particularly suitable for solving by SQP algorithm. Simulation results demonstrate the effectiveness of the proposed method.