Application of Machine Learning in Battery: State of Charge Estimation Using Feed Forward Neural Network for Sodium-Ion Battery

Estimating the accurate State of Charge (SOC) of a battery is important to avoid the over/undercharging and protect the battery pack from low cycle life. Current methods of SOC estimation use complex equations in the Extended Kalman Filter (EKF) and the equivalent circuit model. In this paper, we used a Feed Forward Neural Network (FNN) to estimate the SOC value accurately where battery parameters such as current, voltage, and charge are mapped directly to the SOC value at the output. A FNN could self-learn the weights with each training data point and update the model parameters such as weights and bias using a combination of two gradient descents (Adam). This model comprises the Dropout technique, which can have many neural network architectures by dropping the neuron/mode at each epoch/training cycle using the same weights and biases. Our FNN model was trained with data comprising different current rates and tested for different cycling data, for example, 5th, 10th, 20th, and 50th cycles and at a different cutoff voltage (4.5 V). The battery used for estimating the SOC value was a Na-ion based battery, which is highly non-linear, and it was fabricated in a house using Na0.67Fe0.5Mn0.5O2 (NFM) as a cathode and Na metal as a reference electrode. The FNN successfully estimated the SOC value for the highly non-linear nature of the Na-ion battery at different current rates (0.05 C, 0.1 C, 0.5 C, 1 C, 2 C), for different cycling data, and at higher cut-off voltage of –4.5 V Na+, reaching the R2 value of ~0.97–~0.99, ~0.99, and ~0.98, respectively.

A hybrid method for lithium-ion batteries state-of-charge estimation based on gated recurrent unit neural network and an adaptive unscented Kalman filter

Among the battery state of charge estimation methods, the Kalman-based filter algorithms are sensitive to the battery model while the neural network-based algorithms are decided by hyperparameters. In this paper, a hybrid approach composed of a gated recurrent unit neural network and an adaptive unscented Kalman filter method is proposed. A gated recurrent unit neural network is first used to acquire the nonlinear relationship between the battery state of charge and battery measurement signals, and then an adaptive unscented Kalman filter is utilized to filter out the output noise of the neural network to further improve estimation accuracy. The hybrid method avoids the establishment of accurate battery models and the search for optimal hyperparameters. The data of dynamical street test and US06 test are used as training dataset and validation dataset, respectively, while the data collected from the tests under federal urban driving schedules and Beijing driving cycle conditions are taken as testing dataset. As compared with some hybrid methods proposed in other literature, the hybrid method has the best estimation accuracy and generalization for various driving cycles at different ambient temperatures. The root mean square error and the mean absolute error all are less than 1.5%, and the maximum absolute error are less than 2%. In addition, it also exhibits powerful robustness against the abnormal values of the battery signals and can converge to the true value in just 5 seconds.

The SOC Based Dynamic Charging Coordination of EVs in the PV-Penetrated Distribution Network Using Real-World Data

A successful distribution network can continue to operate despite the uncertainties at the charging station, with appropriate equipment retrofits and upgrades. However, these new investments in the grid can become complex in terms of time and space. In this paper, we propose a dynamic charge coordination (DCC) method based on the battery state of charge (SOC) of electric vehicles (EVs) in line with this purpose. The collective uncoordinated charging profiles of EVs charged at maximum power were investigated based on statistical data for distances of EVs and a real dataset for charging characteristics in the existing grid infrastructure. The proposed strategy was investigated using the modified Roy Billinton Test System (RBTS) performed by DIgSILENT Powerfactory simulation software for a total 50 EVs in 30 different models. Then, the load balancing situations were analyzed with the integration of the photovoltaic (PV) generation and battery energy storage system (BESS) into the bus bars where the EVs were fed into the grid. According to the simulation results, the proposed method dramatically reduces the effects on the grid compared to the uncoordinated charging method. Furthermore, the integration of PV and BESS system, load balancing for EVs was successfully achieved with the proposed approach.

Simulation of multi-power composite electric tractor based on power fluctuation ratio

All Combined with the characteristics of frequent abrupt load in the field operation of tractors, and aiming at the problems of single motor energy input and large fluctuation of battery state in the energy system of pure electric tractors, a power supply structure with multiple battery packs was proposed. A dynamic model considering real-time power and load fluctuation and a multi-power cooperative input model based on fuzzy control threshold logic rule based on power fluctuation ratio are established. Matlab and Simulink are used to simulate the model and compare it with the traditional single power model. The results show that when the speed is constant ploughing, the output power of the lithium battery of the multi-power cooperative input model is effectively compensated compared with that of the single-power model under sudden load. The average fluctuation ratio of rising power decreases from 5.8% / s to 2.7% / s, which realizes “peak clipping” and “slow peak” when the current fluctuates greatly. and then made the estimation of battery state of charge (SOC) more accurate, and prolonged the battery life.