Torque distribution for distributed drive of three-module virtual track trains

To optimize the stability of distributed-drive virtual rail rubber-tired trains, a driving force distribution model was constructed based on the train's power distribution characteristics. This distribution model first allocated the total driving force among the modules and then distributed the driving force among the wheels within each module. The first allocation distributed the total driving force proportionally to each module vehicle according to the acceleration consistency principle. The second allocation used a hierarchical control method to distribute the driving force among the four wheels of each vehicle unit within each module. The upper-layer expected yaw moment controller output the expected yaw moment and total driving force of the module during vehicle operation. The lower layer designed a stability objective function and used quadratic programming method to solve the objective function with the inequality constraint that the driving force did not exceed the road surface adhesion, as well as the equality constraints that the sum of the driving forces of the four wheels was equal to the total driving force of a single vehicle, and the sum of the driving yaw moments of the four wheels was equal to the expected yaw moment of the vehicle. The driving torque required by each wheel during vehicle operation was obtained based on the solution results. By establishing a dynamic model of a three-module virtual rail train and building a joint simulation platform, the feasibility of this strategy was verified, and a comparative analysis was conducted on the distribution performance between this strategy and the average driving torque distribution method. The research results show that the established driving torque distribution model can effectively achieve the distribution of driving torque. Before the first allocation, the longitudinal and lateral hinge forces at the hinge point are 6.4 kN and 7.7 kN, respectively, which are larger than those at the second hinge point, indicating that this distribution strategy has no effect on the mechanical balance of the inter-vehicle hinge system. During curve driving, the driving torques allocated to the wheels on the curve side is higher than that allocated to the wheels on the outer side of the curve. The ratio of the driving torque between the inner and outer sides of the curve for the head car, middle car, and tail car is 6.6%, 16.6%, and 24.9%, respectively, indicating that the tail car has sufficient additional yaw moment to pass through the curve and is easier to pass through the curve. Compared with the average driving torque distribution method, this strategy can achieve the torque difference between the inner and outer wheels at the curve, ensuring additional yaw moment to pass through the curve. This strategy provides a new research idea for the study of driving torque distribution in multi-axle articulated vehicles and offers theoretical support for the development of virtual rail trains towards higher transportation capacity and longer marshalling.