What Is SoE In Battery, State Of Energy, Methods, Purpose

September 26, 2020January 7, 2023 CS Electrical And Electronics

Hello guys, welcome back to our blog. Here in this article, we will discuss what is SoE in battery (state of energy), methods to estimate the SoE or state of energy in the battery, and why SoE estimation is required.

If you have any electrical, electronics, and computer science doubts, then ask questions. You can also catch me on Instagram – CS Electrical & Electronics.

Also, read:

What Is SoE In Battery (State Of Energy)

In current years, lithium-ion batteries have acquired more and better awareness due to their excellent execution in high specific power, high voltage, high specific energy, and extended cycle life with the fast growth of electric vehicles.

Nonetheless, electric vehicles still have weaknesses in comparison with conventional fuel vehicles. The battery energy viscosity of EVs or electric vehicles is still down, resulting in a less driving range, which has evolved into one of the indicators that customers are nervous about but have lower pleasure. Imprecise driving range computation of electric vehicles will drive users’ mileage anxiety, which involves users’ willingness to buy and limits the expansion of the electric vehicle market.

Why SoE (state of energy estimation is needed?

Proper electric vehicle driving range accounting relies on factual SOE computation of the battery pack. Thus, studying an exact battery pack SOE estimating strategy which enhances the precision of evaluating the driving range is a necessary method to enhance the performance and competitiveness of electric vehicles.

Distinct from the computation of the remaining mileage of conventional fuel vehicles which is entirely conditional on the remaining fuel, the battery pack of the electric vehicle is a complicated electrochemical system.

State of energy (SoE) estimation methods

The SOE or state of energy has many controlling factors and is near related to the coming driving requirements of electric vehicles. Thus, it is essential to accurately evaluate the battery pack state of energy or SOE.

In the ex 10 years, Lithium-ion battery state of energy (SOE) computation has attracted the awareness and analysis of domestic and foreign students. About the purpose of battery SOE, there are various views.

One is to instantly view the remaining discharge energy (RDE) of the battery as a state of energy (SOE), and the other is to utilize the ratio of RDE in the current state and the highest RDE in the completely charged state as a state of energy (SOE).

The importance and meaning of the two descriptions are identical. In some analyses, the RDE of the battery is instantly attached to the state of charge (SOC) and is directly estimated by SOC, including establishing the table relationship or calculation formula between RDE and state of charge (SOC) as shown in the below equation, where Uf (vtg) ix is the battery voltage and Qa is the rated capacity.

The direct estimation approach adopts a fixed voltage value to compute RDE, which cannot represent the following dynamic procedure of the battery’s coming voltage.

Several investigators specify the state of energy or SOE as the ratio of RDE and the maximum RDE, as demonstrated in the below equation, where P(τ ) describes battery output power and EN is the maximum RDE of the battery.

This technique utilizes a filter and observer to evaluate the state of energy or SOE based on the equivalent circuit model with SOE Vs OCV curves.

The extended Kalman filter, the EKF-PF, the adaptive unscented Kalman filter, the central difference Kalman filter, the adaptive H infinity filter-based estimators, and the Particle filter, are used to calculate the SOE or state of energy of the battery pack.

Those techniques require specifying the SOE-OCV table connection of the battery, and the table relationship requires to experiment in advance through precise working conditions, which cannot meet the needs of the battery utilized in different complex environments and working situations. Thus, those methods require further validation under real vehicle applications.

Neural network-based approaches have been used for the analysis of the battery state of energy or SOE, which often needs documented data for the training model to receive real results. The typical issue of a neural network is that it needs a considerable quantity of documented data, and the activity data and processes affect the computation results significantly.

Distinct from the above techniques, the battery pack RDE can be accurately calculated based on the iterative prediction of the battery pack state in the coming profile. Some employed the extended Kalman filter (EKF) and recursive least-squares algorithm (RLS) to determine the battery pack parameters and calculate the SOC respectively and suggested a novel RDE prediction framework by collecting the battery discharge capacity and terminal voltage reactions in the hereafter.

Yet, this RDE prediction technique was only validated under specific known temperature and current inputs, which significantly restricts its applicable application. Some utilized a straightforward equivalent circuit model to predict (remaining discharge energy) RDE with the prospective output power and the temperature of the battery indicated by the weighted moving average method (WMA).

This mean-based prediction approach is easy, but the cut-off discharge time cannot be evaluated which is very essential for remaining discharge energy (RDE) estimation. Here the uncertainty of the coming conditions and utilized the Markov model to predict the future pack load of the battery.

Yet, the state transfer probability is challenging to acquire, and the proposed approach generates a large amount of computation. Moreover, the state of energy (SOE) estimation algorithm presented in the above analyses is especially demonstrated at the cell level, and it is challenging to be involved under actual vehicle requirements.

In demand to overwhelm the overhead shortcomings, a novel battery pack state of energy (SOE) meaning under the state of the full life cycle is presented and evaluated based on a prediction approach, which assumes the inconsistency of the battery pack.

The SOC and parameters of a single cell are firstly get established on the RLS and EKF. It is required to accurately predict the terminal voltage, temperature, current, and cut-off discharge time of the battery pack thereafter which cannot be disconnected from the actual battery thermoelectric coupling model under the condition that the prospective output power has been predicted.

Iterative computations are conducted simultaneously to get the state of energy or SOE of the battery pack.

This was about “What Is SoE In Battery (State Of Energy)“. I hope this article may help you all a lot. Thank you for reading.

Also, read: