Model Framework And Energy Management Control For Distributed Driving Electric Vehicles Platoon

A one-dimensional platoon of vehicles is considered as the research object in this paper,and a model framework of a general-purpose platoon of vehicles that serves the analysis and control of the platoon is established,then an adaptive energy management and driving strategy for electric vehicles in platoon is proposed;finally the general-purpose and portability simulation platform laying the foundation for achieving cooperative control of real vehicles is built.The main work of this thesis includes the following aspects:(1)Consider the platoon of vehicles as a special one-dimensional Multi-agent system.First,the basic element-based platoon framework decomposes the platoon into three basic elements: the longitudinal dynamics of a single vehicle,the topology of the inter-vehicle communication,the inter-vehicle distance.Such perspective naturally decomposes a platoon into three interrelated sub-components,namely 1)Node dynamics(ND),2)Information flow topology(IFT),3)Formation geometry(FG).The node dynamics with strong nonlinear characteristics is transformed into a linear model for analysis by employing the feedback linearization technique;the graph topology is used to effectively express the topological structure of the information interaction between vehicles in the platoon in the form of a matrix.Through this way of decoupling,a general model of platoon of vehicles is established,which provides a unified method for quantifying the influence of information flow topology,vehicle dynamics,and controller parameters on the performance metrics of a platoon.(2)Based on the dynamic model of distributed direct-drive electric vehicle,a comprehensive control strategy based on double-layer architecture is proposed.The upper is dynamic control strategy layer and the lower is energy optimization control strategy layer.The decoupling control between the dynamic and energy optimization is realized through the control allocation.linear mappings between the upper virtual control and the lower actual control.The upper controller is mainly responsible for the conventional trajectory tracking control of the vehicle,and the linear mapping relationship between the generalized control variables output by the upper-level model predictive control and the actual control input in the lower layer is defined according to the control allocation theory.An adaptive energy-efficient control allocation strategy is adopted in the lower controller.Based on the motor working efficiency,the torque of the four motors is redistributed to achieve the energy optimization effect.(3)A self-adapting energy management and driving strategy for electric vehicle in platoon is proposed here.This strategy keeps the vehicle to run along a predetermined road with a varying slope,while ensuring a safe distance between adjacent vehicles in the platoon.The discrete-time formulation of the strategy based on the driving and regenerative braking efficiency of a BLDC in-wheel motor experimentally measured,the actual road slope and the formation of the platoon is reformulated in a discrete-distance format.The energy optimization problem is solved by solving the dynamic programming technique to obtain the optimally varied vehicle velocity and globally optimal in-wheel motor actuation torque distributions to reduce the energy consumption of the electric vehicle through the entire trip.(4)By analyzing the features and functions of the actual intelligent vehicles,a simulation platform for model vehicles is designed.According to the construction of the model vehicles used in the platform,the mechanical structure,control framework and the structure of the communication system similar to that of an actual intelligent vehicle are designed respectively to construct the control platform with universality and algorithm portability,and it is possible to transplant algorithms that are validated on the model vehicles to practical ones.