Energy Management in Refrigerated Electric Small Transport: A Hierarchical Approach

Abstract

As the delivery industry undergoes a transition with the introduction of small electric transport vehicles, it is crucial to maximize their energy efficiency. Especially for vehicles with cooling chambers, innovative control strategies tailored for the simple onboard hardware are necessary. The presented approach utilizes two-layer optimization, based on an offline optimization layer and a simple real-time controller. Offline optimization prior to the driving mission uses dynamic programming to ensure optimal operation. The approach involves performing backward and forward recursion, computing the cost-to-go matrix at each step by sweeping through the stage variable. The optimal control trajectory is obtained through forward dynamic programming in the second sweep. The optimized offline trajectory is stored onboard the vehicle, defining the control trajectory for a simple real-time control. This approach allows for a holistic optimal energy management using the limited computational resources in existing transport vehicles. Actual vehicle data and realistic driving missions are used for validations. The potential of the proposed control architecture is evaluated in simulations against mixed-integer predictive control and a rule-based proportional-integral controller. Results indicate energy savings ranging from 12.6 to 14.2% and enhancements in temperature regulation from 1.1 to 2.8%.

Keywords

• Energy management
• hierarchical
• dynamic programming
• model predictive control
• electrical vehicles
• mixed-integer