Abstract: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 the 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 the 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. INDEX TERMS Lithium-ion battery, neural network, remaining useful life, capacity estimation, state of health.
The advent of advanced metering infrastructure (AMI) generates a large volume of data related with energy service. This paper exploits data mining approach for customer baseline load (CBL) estimation in demand response (DR) management. CBL plays a significant role in measurement and verification process, which quantifies the amount of demand reduction and authenticates the performance. The proposed data-driven baseline modeling is based on the unsupervised learning technique. Specifically we leverage both the self organizing map (SOM) and K-means clustering for accurate estimation. This two-level approach efficiently reduces the large data set into representative weight vectors in SOM, and then these weight vectors are clustered by K-means clustering to find the load pattern that would be similar to the potential load pattern of the DR event day. To verify the proposed method, we conduct nationwide scale experiments where three major cities' residential consumption is monitored by smart meters. Our evaluation compares the proposed solution with the various types of day matching techniques, showing that our approach outperforms the existing methods by up to a 68.5% lower error rate.
Abstract:In this paper, we propose novel techniques to reduce total cost and peak load of factories from a customer point of view. We control energy storage system (ESS) to minimize the total electricity bill under the Korea commercial and industrial (KCI) tariff, which both considers peak load and time of use (ToU). Under the KCI tariff, the average peak load, which is the maximum among all average power consumptions measured every 15 min for the past 12 months, determines the monthly base cost, and thus peak load control is extremely critical. We aim to leverage ESS for both peak load reduction based on load prediction as well as energy arbitrage exploiting ToU. However, load prediction inevitably has uncertainty, which makes ESS operation challenging with KCI tariff. To tackle it, we apply robust optimization to minimize risk in a real environment. Our approach significantly reduces the peak load by 49.9% and the total cost by 10.8% compared to the case that does not consider load uncertainty. In doing this we also consider battery degradation cost and validate the practical use of the proposed techniques.
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. INDEX TERMS Deep learning, smart grid, missing imputation, smart meters, denoising autoencoder, generative model, daily load profile (DLP).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.