Energy Consumption Modelling and Parameter Estimation for Autonomous Electric Shuttles

This paper presents a novel probabilistic data-driven approach to trip-level energy consumption estimation of battery electric vehicles (BEVs). As there are very few electric vehicle (EV) charging stations, EV trip energy consumption estimation can make EV routing and charging planning easier for drivers. In this research article, we propose a new driver behavior-centric EV energy consumption estimation model using probabilistic neural networks with model uncertainty. By incorporating model uncertainty into neural networks, we have created an ensemble of neural networks using Monte Carlo approximation. Our method comprehensively considers various vehicle dynamics, driver behavior, and environmental factors to estimate EV energy consumption for a given trip. We propose relative positive acceleration (RPA), average acceleration, and average deceleration as driver behavior factors in EV energy consumption estimation, and this paper shows that the use of these driver behavior features improves the accuracy of the EV energy consumption model significantly. Instead of predicting a single-point estimate for EV trip energy consumption, this proposed method predicts a probability distribution for the EV trip energy consumption. The experimental results of our approach show that our proposed probabilistic neural network with weight uncertainty achieves a mean absolute percentage error of 9.3% and outperforms other existing EV energy consumption models in terms of accuracy.

Power-based electric vehicle energy consumption model: Model development and validation

<div>Electric vehicles (EVs) are critical to the transition to a low-carbon transportation system. The successful adoption of EVs heavily depends on energy consumption models that can accurately and reliably estimate electricity consumption. This article reviews the state of the art of EV energy consumption models, aiming to provide guidance for the future development of EV applications. We summarize influential variables of EV energy consumption in four categories: vehicle component, vehicle dynamics, traffic, and environment-related factors. We classify and discuss EV energy consumption models in terms of modeling scale (microscopic vs. macroscopic) and methodology (data driven vs. rule based). Our review shows trends of increasing macroscopic models that can be used to estimate trip-level EV energy consumption and increasing data-driven models that utilize machine learning technologies to estimate EV energy consumption based on a large volume of real-world data. We identify research gaps for EV energy consumption models, including the development of energy estimation models for modes other than personal vehicles (e.g., electric buses, trucks, and nonroad vehicles), energy estimation models that are suitable for applications related to vehicle-to-grid integration, and multiscale energy estimation models as a holistic modeling approach.</div>

This work investigated the effect of a Nu Vinci type Continuous Variable Transmission (CVT) based Energy Efficient Cruise Control (EECC) on an Electric Vehicle's (EV's) energy consumption. Unlike petrol and diesel vehicles, almost all EVs have a fixed gear ratio. Although it reduces the capital cost, it may not offer optimal energy efficiency and may increase the recharge costs. While accelerating and cruising, it may cause an EV to spend more energy than needed. While braking, it may cause underutilisation of the regenerative braking potential. In this work, an EECC was designed to operate the EV's power train close to its peak efficiency region by controlling the CVT ratio and Electric Machine (EM) torque. The EECC exploits data from the lead vehicle and employs a constrained linear Model Predictive Control (MPC) with a novel problem formulation that reduces the computational complexity. The proposed EECC was tested in a MATLAB simulation environment for different drive cycles. The results show that compared to a baseline EV with a fixed gear ratio and an Adaptive Cruise Control, the proposed system can reduce an EV's energy consumption in urban drive cycles by 16.6%.

Referring to GB/T 18386-2017, the impact of NEDC, WLTC and CLTC on the energy consumption of pure electric vehicle is simulated and analysed. Combined with the characteristic parameters of the driving cycles, from the perspective of energy flow loss of each system of the vehicle, the influences of driving cycles on the energy consumption of each part of the vehicle and the working efficiency of the power components are analyzed in detail, and the internal mechanism of the influence rule of driving cycles on energy consumption is revealed. Based on the CLTC, the effects of load mass, ambient temperature and air conditioning on the energy consumption of the electric vehicle are explored. The research results show that the electric vehicle has the longest driving range and the lowest energy consumption under CLTC; as the loading quality increases, the driving range decreases, and the energy consumption per 100 kilometers increases; test temperature and the air conditioner have a great influence on the driving range, especially in the low temperature environment.

Estimating the HVAC energy consumption of plug-in electric vehicles

Estimation of energy consumption of electric vehicles using Deep Convolutional Neural Network to reduce driver’s range anxiety

This paper introduces a new model for electric vehicle mobility and energy consumption in urban traffic networks. The model couples the vehicle mobility described by a set of ODEs over a graph capturing the Origin-destination motion for urban networks,and the energy consumption associate to this mobility patterns. This model is illustrated in a simple pedagogic example showing its capabilities, such as keeping track of the vehicle state of charge, current energy and available storage.

With the promotion and application of electric vehicles, the effects of low-temperature working conditions on the range of electric vehicles have attracted great attention in academic research. However, the effect and mechanism of tire rolling resistance on the energy consumption of electric vehicles under low-temperature working conditions have not been comprehensively investigated. In this research, 235/50R19 radial tires were used as the research object, and HyperMesh and Abaqus software were used for preprocessing and postprocessing, respectively. To validate the finite element model and its rolling resistance predictions, steady-state rolling simulations were performed under two sets of conditions. First, the tire was simulated on an asphalt road at − 20 °C at ten different speeds: 30, 40, 50, 60, 70, 80, 90, 100, 110, and 120 km/h. Second, simulations were conducted at a constant speed of 80 km/h under four ambient temperatures: 0, − 10, − 20, and − 30 °C. The effect of tire rolling resistance on the energy consumption of electric vehicles was determined by calculating the energy loss. The results revealed that at an ambient temperature of − 20 °C, an increase in the driving speed increased the proportion of tire rolling resistance energy consumption. At a driving speed of 80 km/h, as the temperature continued to decrease, the proportion of tire rolling resistance energy consumption increased. For every 10 km/h reduction in the driving speed, the average decrease in the driving range increased by 1.5%, and for every 10 °C decrease in the ambient temperature, the average range loss increased by 2.3%.

Vehicles offer people many conveniences, but at the same time cause considerable problems, for example for the environment. Today, the electrification of the automobile is considered the best way to solve at least a large part of the environmental problems caused by motor vehicles. Apart from the fact that an electric powertrain produces no local emissions, it also differs from a conventional fossil-fuel powertrain in its propulsion dynamics. It is therefore also to be expected that electric vehicles will influence the flow of traffic due to their different dynamic properties. This thesis focuses on using microscopic traffic simulation to provide a first estimation of how the electrification of vehicles could change the traffic flow and to what extent different traffic situations would influence the fuel and energy consumption of vehicles. In most microscopic traffic flow models, drivers and vehicles are modeled together by a combined model in one simulation object. By adjusting some parameter values, different combinations of driver and vehicle can be roughly distinguished. The values are determined on the basis of observational data of the traffic flow. However, the currently still small share of electric vehicles in road traffic and the fact that they look little different from combustion engine driven and hybrid vehicles make it difficult to collect data from electric vehicles in traffic flow, unless the data is collected in fleet tests. In this thesis, drivers and vehicles were modeled separately. The focus of this thesis is on the assessment of the influence of vehicle propulsion on traffic flow. Therefore, the model developed in this thesis was based on the physical capabilities of both the fossil-fuel powered and the electric powered vehicles. The driver model, which was also required, was created on the basis of data of human drivers, which were determined in an accompanying investigation of a driving simulator. Driver and vehicle model were combined again in the microscopic traffic simulation to a car-following model. For observing the impact of electric vehicles on traffic flow and estimating energy consumption of vehicles, a traffic scenario of a part of the City of Duisburg was established. The effect of electric vehicles on traffic flow, as well as the effect of traffic density on fuel and energy consumption was studied based on this scenario. The simulation results indicate that electric vehicles could raise the average speed of traffic flow to a certain degree, when traffic load and driving styles of the drivers remain the same. The increase of the share of electric vehicles has a positive effect on traffic efficiency. In situations with higher traffic density, energy consumption of electric vehicles is far superior to that of vehicles with internal combustion engine powertrains.

An Experimental Comparison of Physics-based and Machine-Learning-based Electric Vehicle Energy Consumption Estimation Methods

The energy consumption of passenger vehicles is affected by the physical properties of the environment. The ambient temperature in particular has a significant impact on the operating energy consumption. To quantify the impact of a changed climate on vehicles with different drivetrain systems, we set up a model that calculates the change in energy demand with respect to multiple global warming levels. In particular, the effect of rising temperatures on the energy consumption of battery electric vehicles and vehicles with internal combustion engines was investigated. Our results indicate that climate change will likely lead to a rise in energy consumption of vehicles with an internal combustion engine. This is mostly due to the increase in cabin climatization needs caused by the higher ambient temperatures. At a global warming level (GWL) of 4.0 °C, the calculated annual energy consumption on average is 2.1% higher than without taking the climate-change-related changes in temperature into account. Battery electric vehicles, on the other hand, are expected to have a lower overall energy consumption (up to −2.4% at 4 °C GWL) in cold and moderate climate zones. They benefit from the lower heating needs during winter caused by global warming.

Investigation into the energy consumption in electric vehicles (EVs) plays a pivotal role in determining their autonomy and assessing the electric system performance across diverse operational scenarios. This study focuses on the concept of energy regeneration, encompassing the recovery and storage of kinetic mechanical energy during braking or descent in EVs. Employing control systems in power electronics becomes necessary to establish a seamless workflow across operational quadrants to ensure efficient energy regeneration in an electric machine functioning as both a motor and a generator. To seamlessly integrate new technologies into practical applications, it is essential to conduct thorough evaluations in laboratories prior to deployment. This paper introduces an experimental platform specifically designed to analyze energy consumption and storage in EVs by emulating their powertrains in a controlled laboratory environment. The platform comprises key components for emulating the powertrain of a single-motor electric vehicle with single-axle traction, including a power converter configured in two quadrants, an energy storage system, a primary rotating electric machine, and a mechanically coupled point load torque (another motor). This paper provides a detailed guide on implementing such a laboratory and for facilitating the testing of diverse motor technologies and controllers under varied operational conditions. This comprehensive approach allows for the assessment of electromechanical system efficiency, focusing on both energy recovery and comprehensive control of electric power converters. Validation tests conducted under urban conditions and on steep terrains demonstrate the effectiveness of the platform in analyzing the energy efficiency of both the induction machine and the power controller.

Research and verification of energy consumption and E-range of heavy-duty plug-in hybrid electric vehicle

A review of machine learning approaches for electric vehicle energy consumption modelling in urban transportation