Search

This paper proposes a fusion model based on the autoregressive moving average (ARMA) model and Elman neural network (NN) to achieve accurate prediction for the state of health (SOH) of lithium-ion batteries. First, the voltage and capacity degradation variation of the battery are acquired through the battery lifecycle data, and the health factor related to the battery aging is selected according to the variation of the voltage profile. Second, the empirical mode decomposition (EMD) is employed to process the capacity degradation data and eliminate the phenomenon of tiny capacity recovery, and multiple data sequences, as well as the related residue, are extracted, then the grey relational analysis (GRA) between sub-sequences and health factor are discussed. Furthermore, the ARMA model and Elman NN model are respectively built by training the subsequent time series data and residue data. Finally, all the individual predictions are combined to generate the estimated SOH sequences. The experimental validation is performed to manifest that the addressed fusion method performs the SOH prediction with satisfactory accuracy, compared with the single ARMA method and Elman NN model.

In this paper, a novel energy management strategy for plug-in hybrid electric vehicles (PHEVs) is proposed, considering engine on/off control and state-of-charge (SOC) trajectory planning. First, the powertrain structure is analyzed in detail, and then the fuel rate is fitted versus the battery power by a series of quadratic equations. Pontryagin's minimum principle is employed to determine the optimal engine on/off switching and battery power commands, and the estimation distribution algorithm is applied to search for the optimal equivalent factor. Numerical simulations and hardware-in-the-loop experiments validate that near-optimal fuel performance can be achieved based on the proposed algorithm with reasonable computational cost. In addition, the SOC reference trajectory optimized by the proposed algorithm is validated to be effective for application with online energy management strategies.

State of charge (SOC) is the key parameter of electric vehicle power battery. This paper proposes an improved SOC estimation method for lithium-ion battery used in a wide temperature range. First, a equivalent circuit is used to establish the mathematical model of power battery, the genetic algorithm (GA) is used to identify the model parameters at different ambient temperatures, simultaneously. Then, adaptive extended Kalman filter (AEKF) is utilized to estimate battery SOC at different ambient temperatures. The accuracy and robustness of the proposed algorithm are verified under different operating conditions and various initial SOC levels. It is found that the proposed SOC estimation method has the wide temperature adaptability, and could converge to the reference value gradually within the steady-state error 2%.

This paper proposes a state of health (SOH) estimation method with integration of grey relational analysis (GRA) with Elman neural network (NN). First, the experimental data of lithium-ion battery life attenuation are analyzed and the health factors (HFs) are extracted. Then, the correlation degree between HFs and SOH are analyzed by the GRA. Finally, the extracted HFs are considered as the model input, and the SOH as taken as a model target output for SOH prediction. The prediction results show that the proposed method has high prediction accuracy that it can be applied to the online SOH estimation.

In this paper, a state of health (SOH) estimation method using long short term memory (LSTM) networks is applied to predict battery life for electric vehicles (EVs). During the discharging process, the battery shows external features that characterize its attenuation degree and current performance. The discharging time under a constant current, the number of charging and discharging cycles, and the charging capacity are employed to build the prediction model with LSTM networks. The internal modeling parameters are trained by public battery datasets, in which discharging process are introduced for battery SOH prediction. Experimental results indicate that the LSTM networks can accurately predict battery SOH, and estimate battery degradation and internal parameter variations.

In this paper, a fixed size least square - supported vector machine (LS-SVM) algorithm is proposed to estimate the state of health (SOH) of the lithium-ion battery. First, the battery constant current discharging voltage profiles are analyzed and partial discharge curve that can be employed to describe the attenuation of the battery's SOH are extracted, and the suitable voltage range is analyzed. The discharging time is regarded as the input variable of the model. Then the fixed size LS-SVM algorithm is employed, and the corresponding support vector is selected according to Renyi entropy. The model parameters are optimized based on Bayes framework. Experimental results validate that the new method proposed in this paper can effectively track the aging of batteries. Compared with the traditional supported vector machine (SVM), the fixed size LS- SVM requires only 10 support vectors to achieve the desirable SOH estimation.

In this paper, a fixed size least square - supported vector machine (LS-SVM) algorithm is proposed to estimate the state of health (SOH) of the lithium-ion battery. First, the battery constant current discharging voltage profiles are analyzed, partial discharge curve that can be employed to describe the attenuation of the battery's SOH are extracted, and the suitable voltage range is analyzed. The discharging time with different beginning voltage point is regarded as the input variable of the model. Then the fixed size LS-SVM algorithm is employed, and the corresponding support vector is selected according to the arbitrary entropy. The model parameters are optimized based on Bayesian framework. Experimental results validate that the proposed method can effectively track the aging of batteries. Compared with the traditional supported vector machine (SVM), the fixed size LS-SVM requires only 10 support vectors to achieve the desirable SOH estimation.

Precise estimation of state of health (SOH) for lithium-ion batteries is critical to improve the state of charge (SOC) estimation precision, guarantee safety and improve the usage lifetime. In this paper, considering actual usage of lithiumion battery in electrical vehicles (EVs), a novel SOH estimation method is proposed based on the dataset of partial charging voltage under the constant-current (CC) mode and current value under the constant voltage (CV) mode. A random forest (RF) model is built to perform the battery SOH estimation, whose number of regression trees and number of input variables per node are optimized by the out of bag (OOB) estimation. In addition, the support vector machine (SVM) and least squares support vector machine (LSSVM) are employed for comparison of the SOH estimation accuracy. The results show that RF can achieve more accurate estimation of SOH, compared with those based on the SVM and LSSVM.

A dual sliding mode based controller (DSMC) is proposed to suppress the payload swing and adjust the crane position precisely for the overhead crane system. The friction coefficient between the crane and the cable is considered for the nonlinear system. The DSMC is employed for the adjustment of the crane position and the payload swing, integral and differential terms are added into the design of the sliding surface. In order to eliminate the chattering caused by conventional switching function, the variable exponent reaching law is implemented. Simulation and experiment show that DSMC exhibits satisfied robustness for the anti-swing control system.

In this paper, a novel velocity predicted method based on Wavelet Transform (WT) and Radial Basis Function Neural Network (RBF-NN) is introduced to achieve the predictive energy management for plug- in hybrid electric vehicles (PHEVs). Three cases including RBF-NN, 1 RBF-NN with WT and 2 RBF-NN with WT, are taken into account for comparison. The corresponding model predictive contorl (MPC) algorithm is employed to testify the improvement of fuel economy based on the predicted velocity profile. Simulation results show that the combined algorithm of 1 RBF-NN and WT can attain satisfactory performance, which increases the prediction accuracy by 7.31% compared with that by other two cases. In addition, the computational time of the proposed predictive method is 2.8649 ms, justifying that it can be efficiently employed in the practical application.

This paper introduces an online, efficient and practical rule-based energy strategy to manage the energy distribution of a hybrid fuel cell/ battery vehicle. By detailing the powertrain model, the fuel cell current can be optimized based on dynamic programming (DP) method to decrease the total cost. Additionally, the braking distribution factor is taken into account to find out an instantaneous optimal braking table based on genetic algorithm (GA). The comprehensive rules are extracted from the optimized results of twelve typical driving cycles. Numerical simulations prove that the proposed method can effectively save hydrogen consumption and prolong the battery life.

Compared with the conventional vehicles, pure electric vehicles (PEVs) with the advantages of no exhaust and high energy efficiency have attracted enormous attention from government and corporation. In this paper, the effect of super capacitor is discussed for absorbing the braking energy recovery directly. First, through reasonably matching parameters of battery pack and super capacitor, the vehicle model of PEV is established based on the AVL cruise simulation software. Then, the parallel regenerative braking control strategy is proposed, and the related numerical simulations are conducted. The simulation results indicate that, the PEV, which equipped with super capacitor is capable of further improving the efficiency of driving range and braking energy recovery compared with that without super capacitor.

To work against the annoying waggle of a two-dimensional overhead crane system, this paper puts forward a novel model predictive control (MPC) strategy to satisfy various control constraints, particularly the maximum swing angle threshold. In order to simplify the model and satisfy the requirements of embedding application conveniently, the trolley acceleration is employed to be control input. In addition, several critical constraints and soft conditions are added into the MPC, aiming to insure the system stability and improve the overall performance. Simulation results justify the feasibility and correctness of the proposed method.

In this paper, a novel optimization method is proposed to minimize the fuel consumption for plug-in hybrid electric vehicles (PHEVs). First, the quadratic relationship between fuel rate and battery power is constructed. Then, based on the above relationship, the optimal battery power sequence can be gained by quadratic programming (QP). In addition, to find the optimal engine on-off command, genetic algorithm is incorporated with the QP for the search of engine on-off command. Finally, the simulation result shows that, compared with the CD/CS method, the proposed strategy provides a better fuel economy with lower computation burden.

In this paper, a novel optimal charging algorithm for lithium-ion batteries is proposed to optimize constant current charging strategy with the target of minimizing temperature increment. First, a battery temperature model is established along with the identified parameters. Then, battery charging problem is formulated and solved by genetic algorithm (GA). Various charging ambient temperatures are adopted to comprehensive testify the proposed algorithm. Experimental results show that the proposed algorithm can attain satisfactory performance with the slightly shortened charging time and can reduce the temperature increment by 42.6%-93.7% compared with the constant current charging strategy.

Energy loss during charging process for lithium­ion battery has become a main bottleneck for large-scale deployment of batteries in electric vehicles (EVs). This paper proposed a new energy loss minimization charging algorithm with satisfied accuracy and low complexity. To determine the charging current profile, an equivalent circuit model (ECM) is established to accurately characterize the dynamic and static performance of the battery. The battery parameters are identified based on genetic algorithm (GA), and Pontryagin's minimum principle (PMP) is employed to find the optimal charge current profile. The experiments prove that the proposed charge method can reduce the charge energy loss obviously, compared with constant current (CC) charge scheme.

This paper focuses on state of charge (SOC) estimation for the battery packs of electric vehicles (EVs). By modeling a battery based on the equivalent circuit model (ECM), the adaptive extended Kalman filter (AEKF) method can be applied to estimate the battery cell SOC. By adaptively setting different weighed coefficients, a battery pack SOC estimation algorithm is established based on the single cell estimation. The proposed method can not only precisely estimate the battery pack SOC, but also effectively prevent the battery pack from overcharge and over-discharge, thus providing safe operation. Experiment results verify the feasibility of the proposed algorithm.

In order to properly manage lithium-ion batteries of electric vehicles (EVs), it is essential to build the battery model and estimate the state of charge (SOC). In this paper, the fractional order forms of Thevenin and partnership for a new generation of vehicles (PNGV) models are built, of which the model parameters including the fractional orders and the corresponding resistance and capacitance values are simultaneously identified based on genetic algorithm (GA). The relationships between different model parameters and SOC are established and analyzed. The calculation precisions of the fractional order model (FOM) and integral order model (IOM) are validated and compared under hybrid test cycles. Finally, extended Kalman filter (EKF) is employed to estimate the SOC based on different models. The results prove that the FOMs can simulate the output voltage more accurately and the fractional order EKF (FOEKF) can estimate the SOC more precisely under dynamic conditions.

The temperature rise of lithium-ion batteries during the charging process is a significant factor that can influence battery capacity degradation and produce potential safety hazards. In this paper, an optimal charging strategy for LiFePO4 batteries is proposed to minimize the charging temperature rise. First, a battery charging temperature rise model is employed to simulate the temperature variation, and a first-order equivalent circuit model is established to identify the model parameters. Then, the genetic algorithm is applied to search the optimal charging current profiles considering different initial state of charge values and different control intervals. Experimental results prove that the proposed charging strategy not only features a lower temperature rise but also shows less charging time compared to the constant current charging method.