ArticlePDF Available

Intelligent Energy Management Control for Extended Range Electric Vehicles Based on Dynamic Programming and Neural Network

Authors:

Abstract and Figures

The extended range electric vehicle (EREV) can store much clean energy from the electric grid when it arrives at the charging station with lower battery energy. Consuming minimum gasoline during the trip is a common goal for most energy management controllers. To achieve these objectives, an intelligent energy management controller for EREV based on dynamic programming and neural networks (IEMC_NN) is proposed. The power demand split ratio between the extender and battery are optimized by DP, and the control objectives are presented as a cost function. The online controller is trained by neural networks. Three trained controllers, constructing the controller library in IEMC_NN, are obtained from training three typical lengths of the driving cycle. To determine an appropriate NN controller for different driving distance purposes, the selection module in IEMC_NN is developed based on the remaining battery energy and the driving distance to the charging station. Three simulation conditions are adopted to validate the performance of IEMC_NN. They are target driving distance information, known and unknown, changing the destination during the trip. Simulation results using these simulation conditions show that the IEMC_NN had better fuel economy than the charging deplete/charging sustain (CD/CS) algorithm. More significantly, with known driving distance information, the battery SOC controlled by IEMC_NN can just reach the lower bound as the EREV arrives at the charging station, which was also feasible when the driver changed the destination during the trip.
Content may be subject to copyright.
A preview of the PDF is not available
... The optimal strategy is obtained only for that particular cycle. Therefore, given the complex and variable driving conditions in the real world, such energy management strategies cannot be applied in practice and are often used as references for evaluating and optimizing other energy management strategies [11,12]. With the development of energy management, algorithms such as model predictive control (MPC) obtain locally optimal solutions by continuous roll-forward optimization within the prediction-sight distance, which are neither short-sighted nor sensitive compared to instantaneous optimization algorithms. ...
... The information used for updating the cell state c t , f t , and o t are determined by the gating vectors in Formulas (7)- (10). The cell state and output are updated by Formulas (11) and (12). The cell state is reset or restored by f t and the new state c t is obtained by adding partial information through the input gate i t , while the hidden state h t is controlled and updated by the output gate o t . ...
... The information used for updating the cell state ct, ft, and ot are determined by the gating vectors in Formulas (7)-(10). The cell state and output are updated by Formulas (11) and (12). The cell state is reset or restored by ft and the new state ct is obtained by adding partial information through the input gate it, while the hidden state ht is controlled and updated by the output gate ot. ...
Article
Full-text available
To improve the performance of predictive energy management strategies for hybrid passenger vehicles, this paper proposes an Encoder–Decoder (ED)-based velocity prediction modelling system coupled with driving pattern recognition. Firstly, the driving pattern recognition (DPR) model is established by a K-means clustering algorithm and validated on test data; the driving patterns can be identified as urban, suburban, and highway. Then, by introducing the encoder–decoder structure, a DPR-ED model is designed, which enables the simultaneous input of multiple temporal features to further improve the prediction accuracy and stability. The results show that the root mean square error (RMSE) of the DPR-ED model on the validation set is 1.028 m/s for the long-time sequence prediction, which is 6.6% better than that of the multilayer perceptron (MLP) model. When the two models are applied to the test dataset, the proportion with a low error of 0.1~0.3 m/s is improved by 4% and the large-error proportion is filtered by the DPR-ED model. The DPR-ED model performs 5.2% better than the MLP model with respect to the average prediction accuracy. Meanwhile, the variance is decreased by 15.6%. This novel framework enables the processing of long-time sequences with multiple input dimensions, which improves the prediction accuracy under complicated driving patterns and enhances the generalization-related performance and robustness of the model.
... They reduce the real-time computational cost and achieve energy-saving performance close to global optimization approaches but are highly dependent on environmental information. AI-based approaches are promising energy management solutions for MPS-EVs, which include machine learning and deep learning-based knowledge migration methods [13,14,15] and RL-based active control methods [16,17]. The RL demonstrates impressive capabilities for robust regression analysis and strategy development, and the RL-based EMS has shown substantial potential for real-time optimal control in the energy management problem of MPS-EV [18]. ...
... the motor speed at any given moment. The battery state of charge constraint and penalty function is expressed as: (15) where denotes the at moment ; and are the penalty functions for overcharging and over discharging, which are often set according to the needs of the research. Therefore, the objective function for MPS-EV energy management considering the battery charge state can be defined as: ...
Preprint
Reinforcement learning-based (RL-based) energy management strategy (EMS) is considered a promising solution for the energy management of electric vehicles with multiple power sources. It has been shown to outperform conventional methods in energy management problems regarding energy-saving and real-time performance. However, previous studies have not systematically examined the essential elements of RL-based EMS. This paper presents an empirical analysis of RL-based EMS in a Plug-in Hybrid Electric Vehicle (PHEV) and Fuel Cell Electric Vehicle (FCEV). The empirical analysis is developed in four aspects: algorithm, perception and decision granularity, hyperparameters, and reward function. The results show that the Off-policy algorithm effectively develops a more fuel-efficient solution within the complete driving cycle compared with other algorithms. Improving the perception and decision granularity does not produce a more desirable energy-saving solution but better balances battery power and fuel consumption. The equivalent energy optimization objective based on the instantaneous state of charge (SOC) variation is parameter sensitive and can help RL-EMSs to achieve more efficient energy-cost strategies.
... In Xi's research [35], NN was utilized to predict the extender output power for the extended range electric vehicle (EREV). The proposed NN was well trained against the DP optimization results over three artificial driving cycles constructed by repeating US06 two, three, and six times, respectively. ...
... On the contrary, the test vehicle operates in a charge-sustaining mode after arriving at 120-km mark, as the battery SOC has already reached the lower limit. Thus, Xi at el. [35] proposed introducing the electricity consumption per unit distance calculation module, P Cor , to correct the predicted extender power output to ensure the terminal SOC of 0.3. ...
Article
Full-text available
For plug-in hybrid electric vehicles, the equivalent consumption minimum strategy is typically regarded as a battery state of charge reference tracking method. Thus, the corresponding control performance is strongly dependent on the quality of state of charge reference generation. This paper proposes an intelligent equivalent consumption minimum strategy based on dual neural networks and a novel equivalent factor correction, which can adaptively regulate the equivalent factor to achieve the near-optimal fuel economy without the support of the state of charge reference. The Bayesian regularization neural network is constructed to predict the near-optimal equivalent factor online, while the backpropagation neural network is designed to forecast the engine on/off with the aim of improving the quality of equivalent factor prediction. The corresponding neural network training takes advantage of the global optimality of dynamic programming. Besides, the novel equivalent factor correction can guarantee that the electrical energy is gradually consumed along the trip and the terminal battery state of charge satisfies the preset constraints. A series of virtual simulations under a total of nine driving cycles demonstrates that the proposed method can deliver a competitive fuel economy comparing to the optimal solution derived from the dynamic programming, as well as regulating the battery state of charge to reach the desired terminal value at the end of the trip.
... In the interest of simulation accuracy, dynamic programming (DP) is widely recognized as a promising algorithm to find the optimal fuel consumption of APU. In [22], DP algorithm is constructed as a benchmark to compare a Neural network control strategy, considering the vehicle remaining SOC and the distance to charging stations. However, its training driving cycle is NEDC which is no longer suitable for electric vehicles. ...
... Therefore, traffic pattern has a direct influence on the energy consumption and the control strategy operation, where scholars generated several driving cycles and classified them to train the strategy parameters [15,25]. From the recent studies, numerous optimization algorithms relying on forgone driving cycle are proved dependable and effective [22,26]. New and suitable driving cycles should be employed for the accuracy of the energy management study. ...
Article
Urban logistics holds the lifeline of global business and consumption demand. The amount of transport vehicles rises gradually but also leads to graver emission problem. Therefore, electrified vehicles are considered as better options to substitute traditional logistics vehicles. Extended range electric logistics van (ERELV) provides a solution by providing satisfactory driving-range and lower production cost than battery electric logistics van (BELV). This work aims to present a thorough energy consumption and Total cost of ownership (TCO) analysis for an extended range electric van to present its energy potential. Both the ERELV and BELV mathematical model are constructed and compared, and their long-term battery degradation comparison is studied for the first time. Dynamic programming is adopted in the energy management strategy for energy consumption optimization, and the global optimization result reveals the optimal energy consumption of the ERELV. Comparative results demonstrate that the ERELV has a relatively longer drive distance, slower battery aging trend and cheaper TCO (6.6%) when comparing to the BELV. The adoption of ERELV will promisingly reduce operating cost as the alternative transportation solution for urban logistics.
... Still, as long as devices and equipment are related to the term smart, their in timber based on Bluetooth low energy (BLE) and a web-based monitoring system. Other extender applications can be found in [28][29][30][31][32][33][34][35][36][37][38][39][40][41]. ...
... Saban et al. [27] developed a sensor node-based network for remote moisture measurement in timber based on Bluetooth low energy (BLE) and a web-based monitoring system. Other extender applications can be found in [28][29][30][31][32][33][34][35][36][37][38][39][40][41]. ...
Article
Full-text available
Today, the world’s electricity consumption is growing rapidly, and therefore energy demands are also increasing. In the past few decades, various measures have been taken to improve equipment and system design to increase production and transmission efficiency and reduce power consumption. This article proposes a novel Internet of Things (IoT)-based temperature control power extender with two working modes of cooling and heating to solve power shortages. The power is turned on or off accurately and in a timely manner through a temperature-sensing element, thereby avoiding unnecessary power consumption to achieve the goal of energy-saving. This can directly power on or off the power extender through the Internet. It can also use a 2.4G Wi-Fi wireless transmission to transmit, for example, real-time temperature information, the switch status and the master–slave mode. Related data can be controlled, collected and uploaded to the cloud. Each proposed power extender’s temperature setting in a large-scale field can be set uniformly, and no staff is wasted to set the temperature separately. Taking a general industrial electric fan as an example, if it is changed to drive with this temperature control extension cable, and assuming that the industrial electric fan is activated for 900 s per hour, its power-saving rate is 74.75%
... The optimal range strategies can be significantly improved using machine learning and mathematical optimization methods [16]. For instance, promising results were achieved for neural networks utilizing two parameters: battery SOC and distance to the gas stations [43]. A much more complex solution to the same issue was proposed in the paper [44], where stochastic model predictive control is employed to identify the driver's behavior. ...
Article
Full-text available
Extended-Range Electric Vehicles (EREVs) are intended to improve the range of battery electric vehicles and thus eliminate drivers’ concerns about running out of energy before reaching the desired destination. This paper gives an insight into EREV’s performance operating according to the proposed control strategy over various driving cycles, including the Worldwide Harmonized Light-duty Test Cycle Class 3b (WLTC 3b), Federal Test Procedure (FTP-75), and China Light-Duty Vehicle Test Cycle (CLTC-P). Simulation runs were performed in Matlab-Simulink® for different cases of drive range, electricity mix, and vehicle mass. The control strategy goal was to aim at a specified value of battery state of charge at the targeted range value. The obtained test results included: pure electric drive range, acceleration times, EREV range tests, control strategy range errors, Range Extender (REX) utilization metric and distribution of its engagement instances, fuel consumption, total equivalent CO2 emission, powertrain efficiency, and specific energy consumption. The control strategy operated on average with a range error of −1.04% and a range mean square error of 2.13%. Fuel consumption (in range extension mode) varied between 1.37 dm3/100 km (FTP-75) and 6.85 dm3/100 km (WLTC 3b Extra-High 3). CO2eq emission was 95.3–244.2 g/km for Poland, 31.0–160.5 g/km for EU-27, and 1.2–147.6 g/km for Sweden. This paper is a valuable source of information for scientists and engineers seeking to learn the advantages and shortcomings of EREV drives with a proposed control strategy, based on various sets of results.
... The results showed that the optimized operation points of the engine could reach a high efficiency of 40.2% [73]. In Ref. [78], a neural network controller was designed to train data from two inputs: the battery SOC and distance to the gas stations. Stefano et al. [72] applied stochastic model predictive control to identify the driver behaviour, and the engine-battery power ratio changes according to the driving style and traffic condition. ...
Article
To address the urgent environmental challenges of transportation related air pollution and energy shortage, hybrid electric vehicles (HEV) and battery electric vehicles (BEV) with potential for higher energy efficiency are gaining in popularity and gradually replacing those conventional vehicles. Extended range electric vehicle (EREV) is a subset of these new energy vehicles aiming to gain benefits of both HEVs and BEVs and provide a solution to reducing tailpipe emissions whilst providing satisfactory driving-range compared with tradition internal combustion engine (ICE) vehicle counterparts. This paper systematically introduces the current developments of EREV in terms of the powertrain structure, energy management and vehicle performance, while the two key focus areas of auxiliary power unit (APU) charging control and (hybrid) energy storage system ((H)ESS) power management. A case analysis on a typical EREV is presented to study the application potential. Comprehensive comparisons of popular APU control strategies and different energy storage dispositions are shared, several existing issues are also discussed in this paper. Furthermore, a general manufacturing cost analysis for different types of vehicles is presented, and the outlook of EREV for future research is presented.
... Results obtained from optimization algorithm-based EMSs are usually used as the benchmark to improve or evaluate other EMSs. [16][17][18] In addition, optimization algorithms such as the genetic algorithm and particle swarm optimization are also often used to optimize the parameters of EMSs to achieve better control effects. [19][20] In some researches, the model predictive control (MPC) is applied in the EMSs of MPSDSs. ...
Article
Full-text available
This paper develops an online adaptive energy management strategy (EMS) for the promising hierarchical coupled electric powertrain (HCEP) to exert its energy‐saving potential while considering the adaptability to driving conditions and the suppression of mode switching frequency. First, the complex energy management issue of the HCEP is simplified by introducing a simple power allocation method. And, the simplified energy management issue is solved by the Dynamic Programming to obtain the offline optimal working mode sequences of the HCEP. Second, the online working mode decision rules of the HCEP are established according to the obtained working mode sequences. And, the auxiliary rules in the decision rules are further optimized for different types of driving conditions. Then, the principal component analysis and generalized regression neural network are used to construct the driving condition recognizer (DCR) with high prediction accuracy. And, based on the constructed DCR, working mode decision rules, and introduced power allocation method, an online adaptive EMS is developed for the HCEP. Finally, the rationality of the introduced power allocation method and the effectiveness of the developed online adaptive EMS are verified. This paper develops an online adaptive energy management strategy (EMS) for the promising hierarchical coupled electric powertrain (HCEP) applied in vehicles. This online adaptive EMS can not only ensure the energy‐saving effect of the HCEP, but also can effectively avoid frequent working mode switching, as well as has adaptive ability to different driving conditions.
Article
In this study, a coordinated control problem of power sources and actuators of a series-parallel hybrid electric vehicle on mode switching condition is addressed. There exist numerous nonlinear factors in the hybrid driveline system which induce system disturbance and will deteriorate the control performance on mode switching condition. However, coupled system nonlinearity is not taken into consideration for controller design on mode switching condition, which can lead to severe driveline oscillation and thus vehicle longitudinal jerk. For this reason, a disturbance observer and compensation method is introduced to resolve the control problem on mode switching condition. First, an analytic model of driveline torsional vibration considering disturbance brought by system nonlinearity is established and simplified for controller design. Based on the simplified model, a disturbance observer-sliding mode control strategy is proposed to suppress mode switching vibration. Numerical simulation is carried out to validate control performance of the DOB-SMC strategy. Simulation results demonstrate that DOB-SMC strategy has better ability to cancel the unwanted effects of disturbance brought by system nonlinearity compared with sliding mode control method, which can be regarded as an effective approach to attenuate vehicle longitudinal jerk and thus improve ride comfort on mode switching condition.
Article
Energy storage system of extended-range electric vehicle faces great challenges in working efficiency and energy utilization to meet the requirement of various working conditions of vehicle, efficient ultra-capacitor/battery hybrid power source is an effective way to tackle these challenges. This paper takes an ultra-capacitor/battery hybrid power source based on series-parallel switchover technology of ultra-capacitors as the research object, launches a fundamental study on its operation principle, control method and simulation/experimental verification. A significative series-parallel switchover technology of ultra-capacitors is adopted to achieve variable-voltage output of the hybrid power source, which overcomes the technical limitation of the existing hybrid power source with only a fixed working voltage output, enables the voltage-conversion ratio of the power converter to be controlled in a highly efficient range. The bi-directional four working modes of the hybrid power source that can fit the various working conditions of vehicle are analysed. Following this, an incremental digital PID controller suitable for this energy storage system is presented. After that, a comprehensive comparison and analysis of the simulation and small-power experiment based on the simulated working conditions of vehicle are performed. The results have verified the feasibility and effectiveness of the proposed hybrid power source and its control method.
Article
Full-text available
When developing a real-time energy management strategy for a plug-in hybrid electric vehicle, it is still a challenge for the Equivalent Consumption Minimum Strategy to achieve near-optimal energy consumption, because the optimal equivalence factor is not readily available without the trip information. With the help of realistic speeding profiles sampled from a plug-in hybrid electric bus running on a fixed commuting line, this paper proposes a convenient and effective approach of determining the equivalence factor for an adaptive Equivalent Consumption Minimum Strategy. Firstly, with the adaptive law based on the feedback of battery SOC, the equivalence factor is described as a combination of the major component and tuning component. In particular, the major part defined as a constant is applied to the inherent consistency of regular speeding profiles, while the second part including a proportional and integral term can slightly tune the equivalence factor to satisfy the disparity of daily running cycles. Moreover, Pontryagin’s Minimum Principle is employed and solved by using the shooting method to capture the co-state dynamics, in which the Secant method is introduced to adjust the initial co-state value. And then the initial co-state value in last shooting is taken as the optimal stable constant of equivalence factor. Finally, altogether ten successive driving profiles are selected with different initial SOC levels to evaluate the proposed method, and the results demonstrate the excellent fuel economy compared with the dynamic programming and PMP method.
Article
Full-text available
Electric vehicles (EV) are getting more commonplace in the transportation sector in recent times. As the present trend suggests, this mode of transport is likely to replace the internal combustion engine (ICE) vehicles in near future. Each of the main EV components has a number of technologies that are currently in use or can become prominent in the future. EVs can cause significant impacts on the environment, power system, and other related sectors. The present power system can face huge instabilities with enough EV penetration; but with proper management and coordination, EVs can be turned into a major contributor to the successful implementation of smart grid. There are possibilities of immense environmental benefits as well, as the EVs can extensively reduce the greenhouse gas emission from the transportation sector. However, there are some major obstacles for EVs to overcome before replacing the ICE vehicles totally. This paper is focused on reviewing all the useful data available on EV configurations, energy sources, motors, charging techniques, optimization techniques, impacts, trends, and possible directions of future developments. Its objective is to provide an overall picture of the current EV technology and ways of future development to assist in future researches in this sector.
Article
To fully take advantage of external charging conditions and reduce fuel consumption for extended-range electric vehicles, a charging management-based intelligent control strategy is proposed. The intelligent control strategy is applied to different driving patterns based on the various characteristics of urban roads. When the vehicle is driving on arterial roads, a constant power control strategy is applied. When the driver decides to go to a charging station, the extender-off time can be determined based on the current state of the vehicle and the distance to the charging station. When the vehicle is driving on an expressway, a power follower control strategy is applied. The range-extender engine is controlled to work over a wide variety of regions to obtain optimum fuel economy. The simulation results indicate that as the vehicle arrives at the charging station, the proposed charging management-based intelligent control strategy has made the state of charge reach the lowest permissible level after the driver made the decision to charge at the charging station. Therefore, the driver can charge the vehicle with as much clean electric energy as possible from the charging station.
Article
The driving modes of a parallel-series hybrid electric vehicle (HEV) were analyzed. The system efficiency models for the HEV at the charging and discharging mode were proposed based upon the engine efficiency map, motor/generator efficiency map and battery efficiency map. The optimization of energy management strategy for the hybrid system was studied, whose optimization target is the maximum of discharging efficiency at the discharging mode, and the maximum of charging efficiency, the maximum of charging efficiency and charging power at the charging mode with the different state of charge of battery, and the optimal control torque and speed for engine, motor and generator were obtained under the different driving conditions. The simulation of fuel economy shows that the decreasing ratio of fuel consumption of the hybrid system reaches 36.95%.
Article
With energy supplies intense increasingly and environmental protection being concerned in particularly, electric vehicle gained great development in recent years with the characteristics of low exhaust emission, low noise and independent of petroleum. For electric vehicle, the onboard battery capacity is limited. Therefore, the technology of energy utilization improvement has become the research focus. An improved A* algorithm based energy optimal path planning method for electric vehicle was proposed and verified. Energy consumption cost function was built with consideration of the energy loss and recuperation along the path. New heuristic estimated energy cost function was designed to improve A* algorithm. The heuristic estimated energy cost function was proved to be admissible and consistent, which can ensure the optimality of the improved A* algorithm. According to the range anxiety problem, a feasible energy optimal path planning method was proposed with consideration of the state of charge battery, the positions and altitudes of the charging stations and destination. Simulation experiments showed that the proposed method can find the energy optimal path between the origin and the destination. When the battery energy was low, the proposed path planning method can find the feasible energy optimal path with a charging station to reduce range anxiety, which verified the rationality and feasibility of the proposed method. © 2015, Chinese Society of Agricultural Machinery. All right reserved.
Article
The features and structure of range extended electric vehicles (REEV) are presented and its operating modes and control strategies are discussed. The external optimization program of non-dominated sorting genetic algorithm (NSGA-II) is developed and an optimization on the control parameters of REEV is conducted by using simulation software ADVISOR, with minimizing the energy loss of system in a driving cycle as optimization objective. The results show that by using the optimization algorithm proposed, a set of control parameters is found, effectively reducing the energy loss of system.
Article
The basic configuration and operation modes of the series extended-range electric vehicle were analyzed and a novel method for engine choice and fuel economy evaluation was proposed. A co-simulation platform was set up using Cruise and Matlab/Simulink to evaluate the rule-based engine on-off and optimal operation line control strategies. The simulation results with typical driving cycles illustrate that, first, the engine on-off control strategy makes the engine and generator operate at their best efficiency points, and its fuel economy is better than that of the optimal operation line control strategy; secondly, when the optimal operation line control strategy is adopted in the extended-range mode, the engine downsizing will be beneficial and the extended-range electric vehicle(E-REV) will achieve a better fuel economy and control effect.