A review of deep learning for renewable energy forecasting

With the rapid development of new energy electric vehicles and smart grids, the demand for batteries is increasing. The battery management system (BMS) plays a crucial role in the battery-powered energy storage system. This paper presents a systematic review of the most commonly used battery modeling and state estimation approaches for BMSs. The models include the physics-based electrochemical models, the integral and fractional order equivalent circuit models, and data-driven models. The state estimation approaches are analyzed from the perspectives of remaining capacity and energy estimation, power capability prediction, lifespan and health prognoses, and other crucial indexes in BMS. This present paper, through the analysis of literature, includes almost all states in the BMS. The estimation approaches of state-of-charge (SOC), state-of-energy (SOE), state-of-power (SOP), state-of-function (SOF), state-of-health (SOH), remaining useful life (RUL), remaining discharge time (RDT), state-of-balance (SOB), and state-of-temperature (SOT) are reviewed and discussed in a systematical way. Moreover, the challenges and outlooks of the research on future battery management are disclosed, in the hope of providing some inspirations to the development and design of the next-generation BMSs.

A comprehensive review of battery modeling and state estimation approaches for advanced battery management systems

Lithium-ion battery is an appropriate choice for electric vehicle (EV) due to its promising features of high voltage, high energy density, low self-discharge and long lifecycles. The successful operation of EV is highly dependent on the operation of battery management system (BMS). State of charge (SOC) is one of the vital paraments of BMS which signifies the amount of charge left in a battery. A good estimation of SOC leads to long battery life and prevention of catastrophe from battery failure. Besides, an accurate and robust SOC estimation has great significance towards an efficient EV operation. However, SOC estimation is a complex process due to its dependency on various factors such as battery age, ambient temperature, and many unknown factors. This review presents the recent SOC estimation methods highlighting the model-based and data-driven approaches. Model-based methods attempt to model the battery behavior incorporating various factors into complex mathematical equations in order to accurately estimate the SOC while the data-driven methods adopt an approach of learning the battery's behavior by running complex algorithms with a large amount of measured battery data. The classifications of model-based and data-driven based SOC estimation are explained in terms of estimation model/algorithm, benefits, drawbacks, and estimation error. In addition, the review highlights many factors and challenges and delivers potential recommendations for the development of SOC estimation methods in EV applications. All the highlighted insights of this review will hopefully lead to increased efforts toward the enhancement of SOC estimation method of lithium-ion battery for the future high-tech EV applications.

http://dspace.uniten.edu.my/jspui/bitstream/123456789/13358/1/State%20of%20Charge%20Estimation%20for%20Lithium-Ion%20Batteries%20Using%20Model-Based%20and%20Data-Driven%20Methods%20A%20Review.pdf

State of Charge Estimation for Lithium-Ion Batteries Using Model-Based and Data-Driven Methods: A Review

The growing interest and recent breakthroughs in artificial intelligence and machine learning (ML) have actively contributed to an increase in research and development of new methods to estimate the states of electrified vehicle batteries. Data-driven approaches, such as ML, are becoming more popular for estimating the state of charge (SOC) and state of health (SOH) due to greater availability of battery data and improved computing power capabilities. This paper provides a survey of battery state estimation methods based on ML approaches such as feedforward neural networks (FNNs), recurrent neural networks (RNNs), support vector machines (SVM), radial basis functions (RBF), and Hamming networks. Comparisons between methods are shown in terms of data quality, inputs and outputs, test conditions, battery types, and stated accuracy to give readers a bigger picture view of the ML landscape for SOC and SOH estimation. Additionally, to provide insight into how to best approach with the comparison of different neural network structures, an FNN and long short-term memory (LSTM) RNN are trained fifty times each for 3000 epochs. The error is somewhat different for each training repetition due to the random initial values of the trainable parameters, demonstrating that it is important to train networks multiple times to achieve the best result. Furthermore, it is recommended that when performing a comparison among estimation techniques such as those presented in this review paper, the compared networks should have a similar number of learnable parameters and be trained and tested with identical data. Otherwise, it is difficult to make a general conclusion regarding the quality of a given estimation technique.

/pdf/machine-learning-applied-to-electrified-vehicle-battery-4n39y9roqx.pdf

Qiang Miao

Papers

State-of-charge estimation of lithium-ion batteries based on gated recurrent neural network

State-of-charge estimation of lithium-ion batteries using LSTM and UKF