Remaining useful life (RUL) prediction of lithium-ion batteries can reduce the risk of battery failure by predicting the end of life. In this paper, we propose novel RUL prediction techniques based on long short-term memory (LSTM). To estimate RUL even in the presence of capacity regeneration phenomenon, we consider multiple measurable data from battery management system such as voltage, current and temperature charging profiles whose patterns vary as aging. Unlike the traditional LSTM prediction that matches input layer with output layer as one-to-one structure, we leverage many-to-one structure to be flexible for various input types and to substantially reduce the number of parameters for better generalization. Using the NASA lithium-ion battery datasets, we verify the accuracy of the proposed LSTM-based RUL prediction. The experimental results show that the proposed single-channel LSTM model improves the mean absolute percentage error (MAPE) by 39.2% compared to the baseline LSTM model. Furthermore, the proposed multi-channel LSTM model significantly improves the MAPE, e.g., by 63.7% compared to the baseline; the proposed model achieves 0.47-1.88% of MAPE while the state-of-the-art baseline LSTM shows 0.6-6.45% of MAPE.

In the smart grid, one of the most important research areas is load forecasting; it spans from traditional time series analyses to recent machine learning approaches and mostly focuses on forecasting aggregated electricity consumption. However, the importance of demand side energy management, including individual load forecasting, is becoming critical. In this paper, we propose deep neural network (DNN)-based load forecasting models and apply them to a demand side empirical load database. DNNs are trained in two different ways: a pre-training restricted Boltzmann machine and using the rectified linear unit without pre-training. DNN forecasting models are trained by individual customer’s electricity consumption data and regional meteorological elements. To verify the performance of DNNs, forecasting results are compared with a shallow neural network (SNN), a double seasonal Holt–Winters (DSHW) model and the autoregressive integrated moving average (ARIMA). The mean absolute percentage error (MAPE) and relative root mean square error (RRMSE) are used for verification. Our results show that DNNs exhibit accurate and robust predictions compared to other forecasting models, e.g., MAPE and RRMSE are reduced by up to 17% and 22% compared to SNN and 9% and 29% compared to DSHW.

Prognostics and health management is a promising methodology to cope with the risks of failure in advance and has been implemented in many well-known applications including battery systems. Since the estimation of battery capacity is critical for safe operation and decision making, battery capacity should be estimated precisely. In this regard, we leverage measurable data such as voltage, current and temperature profiles from battery management system whose patterns vary in cycles as aging. Based on these data, the relationship between capacity and charging profiles is learned by neural networks. Specifically, to estimate state of health accurately we apply feedforward neural network, convolutional neural network and long short-term memory. Our results show that the proposed multi-channel technique based on voltage, current and temperature profiles outperforms the conventional method that uses only voltage profile by up to 25%–58% in terms of mean absolute percentage error.

Recent advancement of distributed renewable generation has motivated microgrids to trade energy directly with one another, as well as with the utility, in order to minimize their operational costs. Energy trading among microgrids, however, confronts challenges such as reaching a fair trading price, maximizing participants' profit, and satisfying power network constraints. In this paper, we formulate the direct energy trading among multiple microgrids as a generalized Nash bargaining (GNB) problem that involves the distribution network's operational constraints (e.g., power balance equations, voltage limits). We prove that solving the GNB problem maximizes the social welfare and also fairly distributes the revenue among the microgrids based on their market power. To address the nonconvexity of the GNB problem, we propose a two-phase approach. The first phase involves solving the optimal power flow problem in a distributed fashion using the alternative direction method of multipliers to determine the amount of energy trading. The second phase determines the market clearing price and mutual payments of the microgrids. Simulation results on an IEEE 33-bus system with four microgrids show that the proposed framework substantially reduces total network cost by 37.2%. Our results suggest direct trading need be enforced by regulators to maximize the social welfare.

As the number of smart meters increases, compression of metering data becomes essential for data transmission, storing and processing perspectives. Specifically, feature extraction can be used for the compression of metering data and further be utilized for smart grid applications such as customer clustering. So far, there are many studies for compression and clustering based on daily load profiles. However, in order to account for long-term characteristics of electricity consumption, utilizing yearly load profiles (YLPs) is vital for customer load clustering and analysis. In this paper, we propose a deep learning based YLP compression that jointly captures daily and seasonal variations. By leveraging convolutional autoencoder (CAE), YLPs in 8,640-dimensional space are compressed to 100-dimensional vectors. We apply the proposed CAE framework to YLPs of 1,405 residential customers and verify that the proposed CAE outperforms other dimensionality reduction methods in terms of reconstruction errors, e.g., by 19-40%, or the compression ratio is increased by 130% or higher than other methods for the same reconstruction error. In addition, clustering analysis is performed on the encoded YLPs. Our results confirm that year-round characteristics are well captured during the clustering process and also clearly visualized with load images.

Electric load data are essential for data-driven approaches (including deep learning) in smart grid, and advanced smart meter technologies provide fine-grained data with reliable communications. Despite the recent development of smart metering devices, however, missing data still arise due to unexpected device power off, communication failure, measuring error, or other unknown reasons. In this paper, we investigate a deep learning framework for missing imputation of smart meter data by leveraging a denoising autoencoder (DAE). Then, we compare the performance of the proposed DAE with traditional methods as well as other recently developed generative models, e.g., variational autoencoder and Wasserstein autoencoder. The proposed DAE based imputation shows significantly better results compared to other methods in terms of root mean square error (RMSE) by up to 28.9% for point-wise error, and by up to 56% for daily-accumulated error.