Search Results

This study investigates the use of machine learning methods for the selection of energy storage devices in military electrified vehicles. Powertrain electrification relies on proper selection of energy storage devices, in terms of chemistry, size, energy density, and power density, etc. Military vehicles largely vary in terms of weight, acceleration requirements, operating road environment, mission, etc. This study aims to assist the energy storage device selection for military vehicles using the data-drive approach. We use Machine Learning models to extract relationships between vehicle characteristics and requirements and the corresponding energy storage devices. After the training, the machine learning models can predict the ideal energy storage devices given the target vehicles design parameters as inputs. The predicted ideal energy storage devices can be treated as the initial design and modifications to that are made based on the validation results.

A Range-Extended Electric Vehicle (REEV) uses battery as the primary energy source and engine as the secondary source to extend the total range of the vehicle. Deep Orange 11 program at Clemson University is proposing a REEV for solving the mobility needs in the year 2040. Designing the Energy Management System (EMS) of such a vehicle is a critical aspect of the problem statement of this program to improve the vehicle economy and bring down the cost of operation of the vehicle. This paper proposes a reinforcement learning based algorithm for designing the EMS of such a vehicle. Q-learning is a model-free algorithm which seeks to improve the cumulative reward by finding the best policy over the course of operation. A rule-based strategy is first used to establish a baseline model of engine operation during the operation of vehicle over an EPA drive-cycle (FHDS).

This paper presents a hierarchical control architecture to coordinate a group of connected vehicles on signalized intersection roads, where vehicles are allowed to change lane to follow a prescribed path. The proposed hierarchical control strategy consists of two control levels: a high level controller at the intersection and a decentralized low level controller in each car. In the hierarchical control architecture, the centralized intersection controller estimates the target velocity for each approaching connected vehicle to avoid red light stop based on the signal phase and timing (SPAT) information. Each connected vehicle as a decentralized controller utilizes model predictive control (MPC) to track the target velocity in a fuel efficient manner. The main objective in this paper is to consider mandatory lane changes. As in the realistic scenarios, vehicles are not required to drive in single lane. More specifically, they more likely change their lanes prior to signals.

Energy management of hybrid vehicle has been a widely researched area. Strategies like dynamic programming (DP), equivalent consumption minimization strategy (ECMS), Pontryagin’s minimum principle (PMP) are well analyzed in literatures. However, the adaptive optimization work is still lacking, especially for reinforcement learning (RL). In this paper, Q-learning, as one of the model-free reinforcement learning method, is implemented in a mid-size 48V mild parallel hybrid electric vehicle (HEV) framework to optimize the fuel economy. Different from other RL work in HEV, this paper only considers vehicle speed and vehicle torque demand as the Q-learning states. SOC is not included for the reduction of state dimension. This paper focuses on showing that the EMS with non-SOC state vectors are capable of controlling the vehicle and outputting satisfactory results. Electric motor torque demand is chosen as action.

The Organic Rankine Cycle (ORC) has proven to be a promising technology for Waste Heat Recovery (WHR) systems in heavy duty diesel engine applications. However, due to the highly transient heat source, controlling the working fluid flow through the ORC system is a challenge for real time application. With advanced knowledge of the heat source dynamics, there is potential to enhance power optimization from the WHR system through predictive optimal control. This paper proposes a look-ahead control strategy to explore the potential of increased power recovery from a simulated WHR system. In the look-ahead control, the future vehicle speed is predicted utilizing road topography and V2V connectivity. The forecasted vehicle speed is utilized to predict the engine speed and torque, which facilitates estimation of the engine exhaust conditions used in the ORC control model.

Most studies on supervisory controllers of hybrid electric vehicles consider only fuel economy in the objective function. Taking into consideration the importance of the energy storage system health and its impact on the vehicle’s functionality, cost, and warranty, recent studies have included battery degradation as the second objective function by proposing different energy management strategies and battery life estimation methods. In this paper, a rule-based supervisory controller is proposed that splits the torque demand based not only on fuel consumption, but also on the battery capacity fade using the concept of severity factor. For this aim, the severity factor is calculated at each time step of a driving cycle using a look-up table with three different inputs including c-rate, working temperature, and state of charge of the battery. The capacity loss of the battery is then calculated using a semi-empirical capacity fade model.

This paper presents the transient power optimization of an organic Rankine cycle waste heat recovery (ORC-WHR) system operating on a heavy-duty diesel (HDD). The optimization process is carried on an experimentally validated, physics-based, high fidelity ORC-WHR model, which consists of parallel tail pipe and EGR evaporators, a high pressure working fluid pump, a turbine expander, etc. Three different ORC-WHR mixed vapor temperature (MVT) operational strategies are evaluated to optimize the ORC system net power: (i) constant MVT; (ii) constant superheat temperature; (iii) fuzzy logic superheat temperature based on waste power level. Transient engine conditions are considered in the optimization. Optimization results reveal that adaptation of the vapor temperature setpoint based on evaporation pressure strategy (ii) provides 1.1% mean net power (MNP) improvement relative to a fixed setpoint strategy (i).

This paper presents an Organic Rankine Cycle (ORC) system model for heavy-duty diesel (HDD) applications. The dynamic, physics-based model includes: heat exchangers for parallel exhaust and EGR circuits, compressible vapor working fluid, distribution and flow control valves, a high pressure pump, and a reservoir. A finite volume method is used to model the evaporator, and a pressure drop model is included to improve the accuracy of predictions. Experimental results obtained on a prototype ORC system are used for model calibration and validation. Comparison of predicted and measured values under steady-state conditions is pursued first, followed by the analysis of selected transient events. Validation reveals the model’s ability to track real-world temperature and pressure dynamics of the ORC system. Therefore, this modeling framework is suitable for future system design studies, optimization of ORC power generation, and as a basis for development of control-oriented ORC models.