Road Gradient Research Articles

Field-Based Prediction Models for Stop Penalty in Traffic Signal Timing Optimization

Transportation agencies optimize signals to improve safety, mobility, and the environment. One commonly used objective function to optimize signals is the Performance Index (PI), a linear combination of delays and stops that can be balanced to minimize fuel consumption (FC). The critical component of the PI is the stop penalty “K”, which expresses an FC stop equivalency estimated in seconds of pure delay. This study applies vehicular trajectory and FC data collected in the field, for a large fleet of modern vehicles, to compute the K-factor. The tested vehicles were classified into seven homogenous groups by using the k-prototype algorithm. Furthermore, multigene genetic programming (MGGP) is utilized to develop prediction models for the K-factor. The proposed K-factor models are expressed as functions of various parameters that impact its value, including vehicle type, cruising speed, road gradient, driving behavior, idling FC, and the deceleration duration. A parametric analysis is carried out to check the developed models’ quality in capturing the individual impact of the included parameters on the K-factor. The developed models showed an excellent performance in estimating the K-factor under multiple conditions. Future research shall evaluate the findings by using field-based K-values in optimizing signals to reduce FC.

A Deep Learning Method for Monitoring Vehicle Energy Consumption with GPS Data

A monitoring method for energy consumption of vehicles is proposed in the study. It is necessary to have parameters estimating fuel economy with GPS data obtained while driving in the proposed method. The parameters are trained by fuel consumption data measured with a data logger for the reference cars. The data logger is equipped with a GPS sensor and OBD connection capability. The GPS sensor measures vehicle speed, acceleration rate and road gradient. The OBD connector gathers the fuel consumption signaled from OBD port built in the car. The parameters are trained by a 5-layer deep-learning construction with input data (speed, acceleration, gradient) and labels (fuel consumption data) in the typical classification approach. The number of labels is about 6–8 and the number of neurons for hidden layers increases in proportionate to the label numbers. There are about 160–200 parameters. The parameters are calibrated to consider the wide range of fuel efficiency and deterioration degree in age for various test cars. The calibration factor is made from the certified fuel economy and model year taken from the car registration form. The error range of the estimated fuel economy from the measured value is about −6% to +7% for the eight test cars. It is accurate enough to capture the vehicle dynamics for using the input and output data in point-to-point classification style for training steps. Further, it is simple enough to hit fuel economy of the other test cars because fuel economy is a kind of averaged value of fuel consumption for the time period or driven distance for monitoring steps. You can predict or monitor energy consumption for any vehicle with the GPS-measured speed/acceleration/gradient data by the pre-trained parameters and calibration factors of the reference vehicles according to fuel types such as gasoline, diesel and electric. The proposed method requires just a GPS sensor that is cheap and common, and the calculating procedure is so simple that you can monitor energy consumption of various vehicles in real-time with ease. However, it does not consider weight, weather and auxiliary changes and these effects will be addressed in the future works with a monitoring service system under preparation.

Advisor-Based Modelling of a Passenger-Type Fuel Cell Vehicle and the Effect of Movement Resistances on Battery Performance

The most important factors affecting the performance of electric driven vehicles are battery technology and movement resistance. Although many researches have been made on vehicle performance of battery technologies, the number of researches on the effect of movement resistances (rolling resistance, air resistance, road gradient resistance, transmission resistance) and accessory loss on vehicle performance is quite limited. In this study, modelling of a passenger-type fuel cell vehicle has been made on ADVISOR (ADVISOR-Advanced Vehicle Simulator) simulation programme, and effect of movement resistances and accessory losses on battery performance has been examined. At the end of the simulation, it has been determined that SOC (State of Change) value of the tire with low rolling resistance according to NEDC (New European Driving Cycle) driving cycle is higher than the tire with high rolling resistance by 2.2%; and that decrease of C_x A by two times has resulted in a decrease in SOC value by 1.3%. When vehicle-battery performance of transmission selection has been examined, it has been observed that higher friction loss occurred in 5-speed gearbox due to the differences between the gear rations of the gearbox, and average SOC value decreased by 1.4% due to the fact that accessory load was 1000 W instead of 700 W. During simulations made with relation to the incline of the road, usage of tire with low rolling resistance on road incline (ascend) of 5% resulted in a higher value of SOC by 2%. And on road incline (descend) of 5%, usage of tire with low rolling resistance resulted in a higher value of SOC by 2.2%.

Impacts of Extreme Ambient Temperatures and Road Gradient on Energy Consumption and CO2 Emissions of a Euro 6d-Temp Gasoline Vehicle

The EU aims to substantially reduce its greenhouse gas emissions in the following decades and achieve climate neutrality by 2050. Better CO2 estimates, particularly in urban conditions, are necessary for assessing the effectiveness of various regional policy strategies. In this study, we measured the CO2 emissions of a Euro 6d-temp gasoline direct injection (GDI) vehicle with a three-way catalyst (TWC) and a gasoline particulate filter (GPF) at ambient temperatures from −30 °C up to 50 °C with the air-conditioning on. The tests took place both on the road and in the laboratory, over cycles simulating congested urban traffic, dynamic driving, and uphill driving towing a trailer at 85% of the maximum payloads of both the car and the trailer. The CO2 values varied over a wide range depending on the temperature and driving conditions. Vehicle simulation was used to quantify the effect of ambient temperature, vehicle weight and road grade on the CO2 emissions. The results showed that vehicle energy demand was significantly increased under the test conditions. In urban trips, compared to the baseline at 23 °C, the CO2 emissions were 9–20% higher at −10 °C, 30–44% higher at −30 °C, and 37–43% higher at 50 °C. Uphill driving with a trailer had 2–3 times higher CO2 emissions. In motorway trips at 50 °C, CO2 emissions increased by 13–19%. The results of this study can help in better quantification of CO2 and fuel consumption under extreme conditions. Additional analysis on the occurrence of such conditions in real-world operation is advisable.

Research on Torque Distribution of Four-Wheel Independent Drive Off-Road Vehicle Based on PRLS Road Slope Estimation

In view of the high difficulty in coupling of various electric vehicle parameters, intractable parameter estimation, and unreasonable distribution of vehicle driving torque, the four-wheel hub motor is applied to drive electric vehicles, which can instantly obtain the torque and speed of the hub motor and achieve precise control of the torque of each wheel. According to the vehicle longitudinal dynamics model, a progressive RLS (PRLS) algorithm for real-time estimation of vehicle mass and road gradient is proposed. Meanwhile, by means of taking the longitudinal acceleration of the vehicle and the road gradient obtained from the estimation algorithm as the parameter of the torque distribution at the front and rear axles, a dynamic compensation and distribution control strategy of the front and rear axle torques is designed. Moreover, based on hardware-in-the-loop real-time simulation and real-vehicle tests, the effectiveness of the proposed estimation algorithm and the rationality of the real-time distribution control strategy of driving torque are verified.

Building the road network for city-scale active transport simulation models

City-scale simulation modelling of active modes of transportation (i.e., walking and cycling) is becoming increasingly popular in recent years. The heterogeneous and complex behaviour of these transportation modes, however, indicates the need for a shift from the traditional car and public transport centred modelling approaches towards incorporating the requirements for walking and cycling behaviour, while maintaining the run-time efficiency of the models. In this paper, we introduce and test our algorithm to create road network representations, designed and optimised to be used in city-scale active transportation modelling. The algorithm relies on open and universal data. In addition to the major roads and attributes typically used in transport modelling (e.g., speed limit, number of lanes, permitted travel modes), the algorithm also captures minor roads usually favoured by pedestrians and cyclists, along with road attributes such as bicycle-specific infrastructure, traffic signals, road gradient and road surface type. Furthermore, it simplifies the complex geometries of the network and merges parallel roads, if applicable, to make it suitable for large-scale simulations.To examine the utility and performance of the algorithm, we used it to create a network representation for Greater Melbourne, Australia, and compared the output with a network created using an existing simulation toolkit along with another network from an existing city-scale transport model from the Victorian government. Through simulation experiments with these networks, we illustrated that for routed trips on our network for walking and cycling, it is of comparable accuracy to the common network conversion tools in terms of travel distance of the shortest paths while being more than two times faster when used for simulating different sample sizes. Therefore, our algorithm offers a flexible and adjustable solution for users to create road networks for city-scale active transport modelling while balancing between their desired simulation accuracy and run-time.

Impact of Various Operating Conditions on Simulated Emissions-Based Stop Penalty at Signalized Intersections

Sustainability has become one of the most important goals when optimizing traffic signals. This goal is achieved through utilizing various objective functions to reduce sustainability metrics (e.g., fuel consumption and emissions). However, most available objective functions do not distinguish between the reduction mechanism of various types of emissions. Further, such functions do not consider the compound impact of multiple operational conditions (e.g., road gradient) influencing emissions on the optimized signal plans. This study derives a new Environmental Performance Index representing a surrogate measure for emission estimates that can be used as an objective function in signal timings optimization to reduce emissions under various operational conditions. The Environmental Performance Index is a linear combination of delays and stops. The key factor of the Environmental Performance Index is the emissions-based stop penalty, which represents an emission stop equivalency measured in seconds of delay. This study also uses traffic simulation and emission models to investigate the compound impact of several operational conditions on the stop penalty. Results show that the stop penalty varies significantly with all the investigated conditions and that the stop penalty is unique for different types of emissions. These findings may have significant implications on the current practice of sustainable signal timing optimization.

On the road to a sustainable transport mobility in isolated power systems: The role of light-duty powertrain electrification

Plug-in vehicles are seen as a promising path to reduce the road transport greenhouse gas (GHG) emissions. However, the singularities of small isolated power systems in terms of the high dependence of fossil fuel-based generation and the relative short distances travelled do require a particular analysis to evaluate the potential emission reduction. This paper describes a simulation study for estimating the energy consumption and the corresponding Well-to-Wheels GHG emissions of different light-duty powertrain architectures subjected to real-world driving conditions, taking the island of Tenerife (Canary Islands) as a test case. The simulations have been carried out with a high-level vehicle powertrain system analysis tool, capable to estimate second-by-second vehicle energy consumption. Road gradient, GHG grid intensity and battery capacity impact on the GHG emission reduction were analysed in detail.Based on the results, the current high carbon content of the grid and the additional weight of the large battery packs limit the potential benefit of Battery Electric Vehicles (BEVs) with respect to hybrid architectures (HEV). The simulations also reveal that Plug-in Hybrid Electric Vehicles (PHEVs) currently offer a great potential to reduce the GHG emissions. Unlike other geographical areas, the high Utility Factors derived from the short distances travelled ensure the proper use of this technology.

An Intelligent and Autonomous Sight Distance Evaluation Framework for Sustainable Transportation

Railways are facing a serious problem of road vehicle–train collisions at unmanned railway level crossings. The purpose of the study is the development of a safe stopping sight distance and sight distance from road to rail track model with appropriate computation and analysis. The scope of the study lies in avoiding road vehicle–train collisions at unmanned railway level crossings. An intelligent and autonomous framework is being developed using supervised machine learning regression algorithms. Further, a sight distance from road to rail track model is being developed for road vehicles of 0.5 to 10 m length using the observed geometric characteristics of the route. The model prediction accuracy obtained better results in the development of a stopping sight distance model in comparison to other intelligent algorithms. The developed model suggested an increment of approximately 23% in the current safe stopping sight distance on all unmanned railway level crossings. Further, the feature analysis indicates the ‘approach road gradient’ to be the major contributing parameter for safe stopping sight distance determination. The accident prediction study finally indicates that, as the safe stopping sight distance is increased by following the developed model, it is predicted to decrease road vehicle–train collisions.

Machine learning prediction models for battery-electric bus energy consumption in transit

Road gradient (g), initial state of charge (SoCi), drag coefficient (Cd), road condition (RC), HVAC, driver aggressiveness (Dagg.), passenger loading (PL), stop density (SD), average speed (Va), Multiple Regression Analysis (MLR), Radial Basis Function (RBF), Decision Tree Model (DT), Gradient Boosting Decision Tree (GBDT), Support Vector Machine (SVM), Multilayer Perception Neural Network (MLP-NN), and Radial Basis Neural Network (RBNN) • We developed and validated a simulation model to estimate the E C of BEBs. • A factorial experimental design generated BEB operation scenarios. • Seven data-driven models are developed to predict BEB energy consumption. • The models accommodate vehicular, operational, topological, and external parameters. • Multiple Regression and support vector machine models are deemed appropriate to predict BEB’s E C . The energy consumption (E C ) of battery-electric buses (BEB) varies significantly due to the intertwined relationships of vehicular, operational, topological, and external parameters. This variation is posing several challenges to predict BEB’s energy consumption. Several studies are calling for the development of data-driven models to address this challenge. This study develops and compares seven data-driven modelling techniques that cover both machine learning and statistical models. The models are based on a full-factorial experimental design ( n = 907,199 ) of a validated Simulink energy simulation model. The models are then used to predict E C using a testing dataset ( n = 169,344 ). The results show some minor discrepancies between the developed models. All models explained more than 90% of the energy consumption variance. Further, the results indicate that road gradient and the battery state of charge are the most influential factors on E C , while driver behaviour and drag coefficient have the lowest impact.

A novel two-lane lattice hydrodynamic model on a gradient road considering heterogeneous traffic flow

In the actual traffic, there are not only cars, but also buses, trucks and other vehicles. These vehicles with different maximum speeds or security headway or both are interspersed irregularly to form a heterogeneous traffic flow. In addition, most of the maximum speed of modern cars is hardly affected by gradients due to the fact that the car engine and brakes are rarely operated at their max while the maximum speed of trucks is affected. Considering that the performance of various types of vehicles is multifarious and the vehicles sometimes drive on the road with slopes, a novel two-lane lattice hydrodynamic model on a gradient road considering heterogeneous traffic flow is proposed in this paper. In order to verify the rationality of the model, the linear stability analysis is carried out first, that is, the linear stability conditions are derived from the linear stability theory and the stability curve is drawn accordingly. The results of the above analysis prove that the three factors studied in this paper, namely, time lane change, slope and mixing of different types of vehicles, all have a significant influence on the stability of traffic flow. The modified Korteweg–de Vries (mKdV) equation is deduced by the nonlinear analysis method, which can describe the propagation characteristics of the traffic density waves near the critical point. Last but not least, the numerical simulation for new model is conducted and the numerical simulation results obtained are in good agreement with theoretical ones. In summary, increasing the lane changing rate or the slope on the uphill can improve the traffic flow stability. What is more, increasing the slope can lower the traffic flow stability on the downhill. Finally, in the heterogeneous traffic flow of different types of vehicles, the vehicles with larger security headway will make traffic flow difficult to stabilize, as do the vehicles with larger maximum speed.

JUSTIFICATION OF TRANSPORT WORKS IN THE CONDITIONS OF A BLOCK QUARRY

The increase in oil prices, consumables and tools is necessitate the re–equipment of mining enterprises in Ukraine. The share of automobile transport reaches 35–50 % in the dimension blocks production costs.
 The purpose of the work is to determine transportation costs (TC) of 1 m3 of blocks depending on the volume and transportation distance within the quarry; improving the transportation plan by solving transportation problems of linear programming.
 Fuel consumption and dimension block TC for three models of front–end loaders (FEL) were newly calculated on the basis of a comprehensive study. A comparison of the economic feasibility of FEL and dump truck (DT) use in certain mining conditions have been made. The rationalization plan (RP) of dimension blocks transportation within the quarry was newly proposed.
 The research was performed at the dimension gabbro quarry, blocks transportation in which is performed using FEL CAT 988F (1) and DT KrAZ–256B. FELs CAT 988H (2) and CAT 986H (3) were adopted for comparison.
 The average annual fuel consumption is determined: among the FELs considered in the research, the (3) consumes the least, and the (1) currently used at the enterprise consumes 22.7 % and 9.7 % more fuel compared to the (3) and the (2), respectively, in similar conditions. Trade blocks transportation using FEL is economically more expedient in comparison with the DT at transportation distance up to 300–400 m.
 It is determined that the trade blocks TC using the (2) and the (3) is 5.3 % and 12.6 % lower, respectively, compared to the (1).
 This paper also considers the transportation problem for FELs, in which blocks from several banks must be distributed between several storage areas, provided to minimize TC. According to the proposed RP of transportation, the savings will be up to 13 % compared to the initial conditions, which in monetary terms is 41538–48639 UAH/year depending on the FEL model. The replacement of the (1) for the (3) will reduce block TC by 12.6 % (47000 UAH/year) in the current conditions. The TC using the (3) will be 285303.5 UAH according to the proposed RP, which is 23.7 % (88476 UAH/year) less than the (1) application before the rationalization.
 
 Keywords: dimension stone; front–end loader; commercial blocks transportation costs; transportation problem; cargo traffic rationalization; fuel consumption; haul road gradient.

Mass estimation of ground vehicles based on longitudinal dynamics using IMU and CAN-bus data

This paper considers the problem of real-time estimation of ground vehicle mass, based on longitudinal dynamics. Compared to previous contributions, this work relies on basic data from the vehicle controller area network bus. This allows for application to a very broad range of vehicles, by not using vehicle specific parameters such as current gear indicator and engine torque. The current gear ratio is predicted based on the engine and vehicle speeds. The engine torque is modelled by the mass air flow rate of the engine. The road gradient is estimated with an inertial measurement unit, with fusion of gyroscope and accelerometer using a complementary filter. To handle prolonged dynamic conditions affecting the performance of the filter, compensation of external accelerations is made by use of the vehicle speed from the controller area network bus. The mass is estimated with a recursive least squares filter with forgetting factor. The model is validated experimentally with data obtained from test drives with two different petrol powered passenger cars, equipped with a manual transmission and dual clutch automatic transmission respectively. The test drives consist of various driving conditions with different vehicle loads. The resulting mass estimates are generally good for both vehicles with variations within ±5% of the actual masses.

Effects of driving behavior on real-world emissions of particulate matter, gaseous pollutants and particle-bound PAHs for diesel trucks

This study employed a portable emissions measurement system to investigate the effects of vehicle attributes, driving behavior, and road grade on real-world emissions of particulate matter (PM), regulated gaseous pollutants, and particle-bound polycyclic aromatic hydrocarbons (PAHs) for old-model diesel trucks (model year 1995–2006, 6.7–35.0 metric ton) with little to no tailpipe emission control. The rated power of engines was a major determinant of the distance-specific emission factors of PM, particle-bound PAHs, and most gaseous pollutants. However, the engine size was unrelated to the total hydrocarbon emission factor and the benzo[a]pyrene equivalent (BaPeq) emission factor of PAHs. Aggressive (AG) and normal (NR) driving behaviors were quantitatively defined with a relative positive acceleration. The emission factors of PM, CO2, and THC were significantly different (p < 0.05) between the AG and NR driving modes. AG driving caused an average increase in emissions of PM, CO2, NOx, and particle-bound PAHs by 122%, 56%, 15%, and 128%, respectively, compared to the respective emissions under the NR mode. The BaPeq emission factor of PAHs in the AG mode was more than 10 times that in the NR mode. The road gradient (ranging from −9.3% to 9.0% over the test route) had significant impacts on the emissions of PM, CO2, and NOx. PM, CO2, and NOx emission factors increased by 109%, 168%, and 160%, respectively, in the >6% grade bin and decreased by 95%, 91%, and 90%, respectively, in the equivalent negative-grade bin, implying that the decrease in emissions on negative road slopes may not compensate for the increase in emissions on the equivalent positive road slopes despite the road slope being compensated. The findings of this study will be valuable for developing air quality management strategies and furthering scientific knowledge on the complex interplay of different variables that affect real-world emissions of on-road vehicles.

Prediction of instantaneous real-world emissions from diesel light-duty vehicles based on an integrated artificial neural network and vehicle dynamics model

This paper presents a road vehicle emission model that integrates an artificial neural network (ANN) model with a vehicle dynamics model to predict the instantaneous carbon dioxide (CO2), nitrogen oxides (NOx) and total hydrocarbon (THC) emissions of diesel light-duty vehicles. Real-world measurement data were used to train a multi-layer feed-forward ANN model. The optimal combination of the various experimental variables was selected as the ANN input through a parametric study considering both practicality and accuracy. For CO2 prediction, two variables (engine speed and engine torque) are enough to develop an accurate ANN model. In order to achieve satisfactory accuracy for CO and NOx prediction, more variables were used for ANN training. The trained ANN model was used to predict road vehicle emissions by integrating the vehicle dynamics model, which was used as a supplementary tool to produce ANN input data. The integrated model is practical because it requires relatively simple data for input such as vehicle specifications, velocity, and road gradient. In the accuracy validation, the proposed model showed satisfactory prediction accuracy for road vehicle emissions.

Unpaved road erosion after heavy storms in mountain areas of northern China

More frequent extreme rainfall events associated with global climate change cause greater challenges for soil conservation. Severe erosion occurs on many unpaved roads since these structures create important water flow paths during heavy storms. The present research aimed to investigate the intensity and influencing factors of unpaved road erosion under varied land use and management conditions (sloping cropland, terraced cropland, forest & grass). The erosion occurred in the watersheds contributing runoff water to roads after the greatest rainfall event recorded in the mountain area of northern China caused by Typhon Lekima. The research was conducted in an agricultural-forest-dominated watershed based on field investigation and UAV-based image analysis. A road erosion level classification standard was given according to the occurrence of rills, ephemeral gullies, and gullies. Significant erosion happened on 67% of the unpaved roads; 42% of them suffered moderate to severe erosion in which ephemeral gullies or gullies developed. The average erosion amount from these roads was 2280.75 t ha-1 and was significantly influenced by the watershed land use type and management. The dominant factor governing unpaved road erosion associated with terraced cropland was vegetation coverage on roads. Drainage area was the most important factor for road erosion in sloping cropland and forest & grass land, and road gradient was also a critical factor. Terraces, and forest & grass in drainage areas significantly reduced unpaved road erosion by 85% and, 47%, respectively, compared to sloping cropland. More integrated measures should be used to prevent unpaved road erosion. The results of this research can be applied to road protection against erosion in heavy storms.

An Empirical Study for Green Transportation Scheme of Municipal Solid Waste Based on Complex Data Model Analysis

Solid waste management and air pollution are two pressing issues in the functioning of large cities. This paper studies the optimization problem of the green transportation route of municipal solid waste and establishes a mathematical planning model based on real-time traffic conditions of the city and consideration of a time window and multiple transfer stations with the goal of minimizing energy consumption. In the optimal green transportation process in this paper, comprehensive consideration of vehicle speed, vehicle load, road gradient, and driving distance in different road sections based on real-time traffic conditions is incorporated, which has a better fuel-saving potential than the shortest path. A green transportation program can alleviate the air pollution problem in big cities and promote energy conservation and emission reduction in solid waste transportation.

Estimation of initiation thresholds and soil loss from gully erosion on unpaved roads on China's Loess Plateau

AbstractGullies form easily on unpaved road surfaces during heavy rainstorms on China's Loess Plateau. The integrated effect of rainfall, topography, vegetation, land use, and other factors determines where and when gully erosion occurs; however, the mechanisms driving gully erosion on unpaved road surfaces need to be further understood. Repeated gully erosion on some roads during the storm season provides a good opportunity to better understand the mechanisms behind gully erosion. This article aimed to quantify the integrated threshold conditions required for gully initiation in terms of topography, event rainfall, and upslope land use, and to propose an event‐based model to predict the position and magnitude of road gully erosion. Rill and gully erosion on unpaved roads were investigated after an extreme rainstorm of 212.2 mm in 2017 and a regular rainstorm of 83.8 mm in 2018. A digital surface model (DSM) derived from unmanned aerial vehicle (UAV) images was used to analyze the road gradient (S) and upslope area (A). The runoff (QRS) of each road segment (RS) was estimated by the runoff curve number method. The results showed the following: (1) The mean and total gully volumes under the regular rainstorm were only 26.3% and 8.1% of those under the extreme rainstorm, respectively; (2) Gully formation on the surveyed roads under both the regular and the extreme rainstorm could be explained by the threshold relationship (S − 0.056)QRS3.363 ≥ 72.444; and (3) A non‐linear relationship between gully erosion in road segments and event runoff (QRS) and road gradient (S) was found, and was subsequently used to predict road gully erosion on an event basis.

Electric vehicle energy consumption prediction using stacked generalization: an ensemble learning approach

ABSTRACT In this paper, we present an ensemble stacked generalization (ESG) approach for better prediction of electric vehicles (EVs) energy consumption. ESG is a weighted combination of multiple base regression models to one meta-regressor, which enhances the model prediction and decreases the model variance over a single regressor model. For the current study, we develop ESG by combining three individual base machine learning algorithms, i.e., Decision Tree (DT), Random Forest (RF), and K-Nearest Neighbor (KNN), to predict the EVs’ energy consumption. Tackling the challenge of predicting EVs’ energy consumption, the data were collected from Aichi Prefecture, Japan, combining the digital elevation map with long-term GPS tracking data. EVs energy consumption in terms of energy efficiency (kWh/km) was estimated using several important variables such as average trip speed (km/h), trip distance, nighttime lighting, air conditioner (A/C), heater usage ratio, and road gradient. Several statistical evaluation metrics were used to evaluate the performance of the proposed methods. The prediction results show that ESG is more robust in predicting EVs’ energy consumption and outperformed other models by yielding more acceptable values for proposed evaluation metrics. The results further demonstrate that the accuracy of predictive models for EVs energy consumption can be reasonably accomplished by adopting stacking techniques. The finding of this study could provide essential guidance to decision-makers and practitioners for planned development and optimal placing of EV charging infrastructures in urban areas.

A novel energy management strategy of hybrid electric vehicle via an improved TD3 deep reinforcement learning

The formulation of high-efficient energy management strategy (EMS) for hybrid electric vehicles (HEVs) becomes the most crucial task owing to the variation of electrified powertrain topology and uncertainty of driving scenarios. In this study, a deep reinforcement learning (DRL) algorithm, namely TD3, is leveraged to derivate intelligent EMS for HEV. A heuristic rule-based local controller (LC) is embedded within the DRL loop to eliminate irrational torque allocation with considering the characteristics of powertrain components. In order to resolve the influence of environmental disturbance, a hybrid experience replay (HER) method is proposed based on a mixed experience buffer (MEB) consisting of offline computed optimal experience and online learned experience. The results indicate that improved TD3 based EMS obtained the best fuel optimality, fastest convergence speed and highest robustness in comparison to typical value-based and policy-based DRL EMSs under various driving cycles. LC leads to a boosting effect on the convergence speed of TD3-based EMS wherein a “warm” start of exploring is exhibited. Meanwhile, by incorporating HER coupled with MEB, the impact of environmental disturbance including load mass and road gradient, as an increase of input observations, can be negligible to the performance of TD3-based EMS.