Simulation and Bench Testing of a GM 5.3L V8 Engine 2017-01-1259

The Hybrid Electric Vehicle Team of Virginia Tech (HEVT) is currently modeling and bench testing powertrain components for a parallel plug-in hybrid electric vehicle (PHEV). The custom powertrain is being implemented in a 2016 Chevrolet Camaro for the EcoCAR 3 competition. The engine, a General Motors (GM) L83 5.3L V8 with Active Fuel Management (AFM) from a 2014 Silverado, is of particular importance for vehicle integration and functionality. The engine is one of two torque producing components in the powertrain. AFM allows the engine to deactivate four of the eight cylinders which is essential to meet competition goals to reduce petroleum energy use and greenhouse gas emissions. In-vehicle testing is performed with a 2014 Silverado on a closed course to understand the criteria to activate AFM. Parameters required for AFM activation are monitored by recording vehicle CAN bus traffic.

Simulink models of the engine plant and SoftECU are developed to represent the operation of the engine system. AFM logic in the engine SoftECU is modeled based on known parameters and validated with test data. The engine model is tested as part of a full hybrid vehicle model with two different control strategies: Fuzzy logic and a regen-only strategy. Fuel energy consumption is modeled as 597 Wh/mi when using a Fuzzy logic-based strategy, reduced from 601 Wh/mi with a strategy that only used regenerative braking. The donor engine is then run on a test stand to determine startup requirements and validate the selected transmission. All performed tests are significant for collection of data and information that can help improve the hybrid control strategy. Improving the hybrid vehicle control strategy allows the motor to load or assist the engine output torque to maintain AFM mode while using the stock L83 engine control module. All of these activities are essential for building a safe and functional vehicle and control strategy which will meet all EcoCAR 3 and HEVT goals.