State of charge estimation for electric vehicle power battery using advanced machine learning algorithm under diversified drive cycles

Zahid, Taimoor; Xu, Kun; Liu, Weimin; Li, Chenming; Li, Hongzhe

doi:10.1016/j.energy.2018.08.071

Cited by 168 publications

(39 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because the variance of SOC determines the dispersion of the particles, a greater variance will require larger number of particles to reach a steady estimation. For the consideration of computing e ort, a coe cient is introduced to narrow the covariant in Equation (20). e value of , which we set it as 0.6, is a tradeo of accuracy and computing speed.…”

Section: Correction By Sir Particlementioning

confidence: 99%

Online State of Charge Estimation of Lithium-Ion Cells Using Particle Filter-Based Hybrid Filtering Approach

2020

View full text Add to dashboard Cite

Filtering based state of charge (SOC) estimation with an equivalent circuit model is commonly extended to Lithium-ion (Li-ion) batteries for electric vehicle (EV) or similar energy storage applications. During the last several decades, different implementations of online parameter identification such as Kalman filters have been presented in literature. However, if the system is a moving EV during rapid acceleration or regenerative braking or when using heating or air conditioning, most of the existing works suffer from poor prediction of state and state estimation error covariance, leading to the problem of accuracy degeneracy of the algorithm. On this account, this paper presents a particle filter-based hybrid filtering method particularly for SOC estimation of Li-ion cells in EVs. A sampling importance resampling particle filter is used in combination with a standard Kalman filter and an unscented Kalman filter as a proposal distribution for the particle filter to be made much faster and more accurate. Test results show that the error on the state estimate is less than 0.8% despite additive current measurement noise with 0.05 A deviation.

show abstract

Section: Correction By Sir Particlementioning

confidence: 99%

Online State of Charge Estimation of Lithium-Ion Cells Using Particle Filter-Based Hybrid Filtering Approach

2020

View full text Add to dashboard Cite

show abstract

“…Finally, it should be noted that the application of forecasting method is gaining wide-spread acceptance in power and energy industry. Examples of lateral application include occupancy prediction of office buildings [143], State of charge estimation for electric vehicle [144] the estimation of energy consumption in buildings using solar data [145], and forecasting the of district heating consumption [146], security assessment of power systems [147], and restoring microgrids after fault occurrence [148].…”

Section: Predicting Power Demandmentioning

confidence: 99%

Machine-learning methods for integrated renewable power generation: A comparative study of artificial neural networks, support vector regression, and Gaussian Process Regression

Sharifzadeh

Sikinioti-Lock

Shah

2019

Renewable and Sustainable Energy Reviews

260

View full text Add to dashboard Cite

Renewable energy from wind and solar resources can contribute significantly to the decarbonisation of the conventionally fossil-driven electricity grid. However, their seamless integration with the grid poses significant challenges due to their intermittent generation patterns, which is intensified by the existing uncertainties and fluctuations from the demand side. A resolution is increasing energy storage and standby power generation which results in economic losses. Alternatively, enhancing the predictability of wind and solar energy as well as demand enables replacing such expensive hardware with advanced control and optimization systems. The present research contribution establishes consistent sets of data and develops data-driven models through machine-learning techniques. The aim is to quantify the uncertainties in the electricity grid and examine the predictability of their behaviour. The predictive methods that were selected included conventional artificial neural networks (ANN), support vector regression (SVR) and Gaussian process regression (GPR). For each method, a sensitivity analysis was conducted with the aim of tuning its parameters as optimally as possible. The next step was to train and validate each method with various datasets (wind, solar, demand). Finally, a predictability analysis was performed in order to ascertain how the models would respond when the prediction time horizon increases. All models were found capable of predicting wind and solar power, but only the neural networks were successful for the electricity demand. Considering the dynamics of the electricity grid, it was observed that the prediction process for renewable power and wind was fast and accurate enough to effectively replace the alternative electricity storage and standby capacity.

show abstract

“…10 Typical datadriven methods include neural networks, [11][12][13] support vector regression 14 and fuzzy system. 15 The data-driven methods are capable of self-learning from data. Complicated knowledge of electrochemical dynamics is not necessary for the data-driven methods.…”

Section: Introductionmentioning

confidence: 99%

“…The third one is the data‐driven methods 10 . Typical data‐driven methods include neural networks, 11‐13 support vector regression 14 and fuzzy system 15 . The data‐driven methods are capable of self‐learning from data.…”

Section: Introductionmentioning

confidence: 99%

State of charge estimation for lithium‐ion battery based on an intelligent adaptive unscented Kalman filter

Sun

Zhang

et al. 2020

Int J Energy Res

View full text Add to dashboard Cite

Adaptive unscented Kalman filter (AUKF) has been widely used for state of charge (SOC) estimation of lithium-ion battery. The noise covariance of the conventional AUKF method is updated based on the innovation covariance matrix (ICM), which is estimated using the error innovation sequence (EIS). However, the distribution of EIS changes due to the time-varying noise, load current dynamics and modelling error, which will lead to inaccurate ICM estimation. Therefore, an intelligent adaptive unscented Kalman filter (IAUKF) method is proposed to detect the distribution change of EIS. Then, the ICM is estimated based on the EIS after the distribution change. Results show that the IAUKF method can improve SOC estimation accuracy significantly. Compared with that of the AUKF method, the root mean squared error and the mean absolute error of SOC based on the IAUKF method decrease by 43.70% and 72.37% under random walk discharge condition, respectively. In addition, the computation time of the IAUKF method slightly increases by 6.27% compared with that of AUKF method. Finally, the effect of initial parameters on the SOC estimation accuracy was analysed. The results indicate that proper algorithm tuning, such as initial window length of EIS for ICM update and the threshold value, can further improve the SOC accuracy based on the proposed IAUKF method. The proposed IAUKF method also shows high robustness against initial measurement noise covariance.

show abstract

State of charge estimation for electric vehicle power battery using advanced machine learning algorithm under diversified drive cycles

Cited by 168 publications

References 37 publications

Online State of Charge Estimation of Lithium-Ion Cells Using Particle Filter-Based Hybrid Filtering Approach

Online State of Charge Estimation of Lithium-Ion Cells Using Particle Filter-Based Hybrid Filtering Approach

Machine-learning methods for integrated renewable power generation: A comparative study of artificial neural networks, support vector regression, and Gaussian Process Regression

State of charge estimation for lithium‐ion battery based on an intelligent adaptive unscented Kalman filter

Contact Info

Product

Resources

About