skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Residential micro-hub load model using neural network

 [1];  [2];  [2]
  1. ORNL
  2. University of Waterloo, Canada
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE Laboratory Directed Research and Development (LDRD) Program
OSTI Identifier:
DOE Contract Number:
Resource Type:
Resource Relation:
Conference: IEEE North America Power Symposium, Charlotte, NC, USA, 20151004, 20151006
Country of Publication:
United States

Citation Formats

Sharma, Isha, Canizares, Claudio, and Bhattacharya, Kankar. Residential micro-hub load model using neural network. United States: N. p., 2015. Web. doi:10.1109/NAPS.2015.7335091.
Sharma, Isha, Canizares, Claudio, & Bhattacharya, Kankar. Residential micro-hub load model using neural network. United States. doi:10.1109/NAPS.2015.7335091.
Sharma, Isha, Canizares, Claudio, and Bhattacharya, Kankar. 2015. "Residential micro-hub load model using neural network". United States. doi:10.1109/NAPS.2015.7335091.
title = {Residential micro-hub load model using neural network},
author = {Sharma, Isha and Canizares, Claudio and Bhattacharya, Kankar},
abstractNote = {},
doi = {10.1109/NAPS.2015.7335091},
journal = {},
number = ,
volume = ,
place = {United States},
year = 2015,
month = 1

Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Decentralized load management is an essential part of the power system operation. Forecasting load demand at the substation level is generally more difficult and less accurate compared to forecasting total system load demand. In this paper, Multi-Layered Feed Forward (MLFF) neural network is used to predict the bus-load demand at the substation level. The MLFF network is trained using Back-Propagation (BP) algorithm with adaptive learning technique. The algorithm is tested for two systems having different load patterns.
  • In this paper an attempt is made to forecast load using fuzzy neural network (FNN) for an integrated power system. Here, the proposed system uses a two stage FNN for a short term peak and average load forecasting (STPALF). The first stage FNN deals with the load forecasting and the second stage algorithm can be worked independently for network security. This technique is used to forecast load accurately on week days as well as holidays, weekends and some special occasions considering historical data of load and weather information and also take necessary control action for network security.
  • Operators of on-load nuclear reactors may remotely obtain access to the core from both ends, and the reactors can be continuously fueled without shutting them down. Such an operation offers a fuel management advantage, but a safeguards challenge, because it provides a greater opportunity for the diversion of nuclear material. Automated analysis using pattern recognition and neural network software can help interpret data, call attention to potential anomalies, and improve safeguards effectiveness. Automated software analysis, based on pattern recognition and neural networks, was applied to data collected from a radiation core discharge monitor system located adjacent to on-load reactor core.more » Unattended radiation sensors continuously collect data to monitor on-line refueling operations in the reactor. The huge volume of data collected from a number of radiation channels makes it difficult for a safeguards inspector to review it all, check for consistency among the measurement channels, and find anomalies. Pattern recognition and neural network software can analyze large volumes of data from continuous, unattended measurements, thereby improving and automating the detection of anomalies. The authors developed a prototype pattern recognition program that determines the reactor power level and identifies the times when fuel bundles are pushed through the core during on-line refueling. Neural network models were also developed to predict fuel bundle burnup to calculate the region on the on-load reactor face from which fuel bundles were discharged based on the radiation signals. In the preliminary data set, which was limited and consisted of four distinct burnup regions, the neural network model correctly predicted the burnup region with an accuracy of 92%.« less
  • The development of feed-forward artificial neural network based models to predict gas consumption on a daily basis is the subject of this paper. The discussion concerns an iterative process based on network sensitivities and intuition regarding selection of proper input factors. The method is applied in forecasting gas consumption for two regions in the State of Wisconsin, namely, a portion of metropolitan Milwaukee and a region near Fond du Lac. The investigation includes a study of the effects of various network sizes and training algorithms given a limited availability of relevant historical data. The effects of using multiple sources ofmore » weather data are also investigated for regions without a centrally located weather recording station. The training results indicate that feed-forward artificial neural network based models reduce the residual predicted consumption root mean squared errors by more than half when compared to models based on linear regression using identical input factors. The inclusion of weather data from multiple sources leads to further error reduction. The models were then used for load forecasting starting in Summer 1994. The static models performed well from the Summer 1993 until the severe cold weather of mid-January 1994. This poor performance was scrutinized; investigations point to probable causes that include severe weather patterns non-existent in the training data and the excitation of consumer behavioral modes either not evident in the training data or not well represented by time-of-the-year input factor proxies. The search for a solution lead to dynamic models that were implemented by updating the neural networks with new data as it became available. These models performed significantly better but still poorly during and after the severe weather of January 1994.« less
  • Predicting the thermal load for the next 24 hours is essential for optimal control of heating, ventilating, and air conditioning (HVAC) systems that use thermal cool storage. It can be useful in minimizing costs and energy in nonstorage systems. A cooperative research project between a US. university and a Japanese corporation investigated four generally used prediction methods to examine the basic models with variations and to compare the accuracy of each model. A cooling and heating seasonal data set with known next-day weather was used to evaluate the accuracy of each prediction method. The results indicate that an artificial neuralmore » network (ANN) model produces the most accurate thermal load predictions. After the initial comparisons with a computer-generated data set, the ANN model was applied to two measured building loads from another research project. These sets included typical measurement noise related to continuous field monitoring. The predictions of the next-day cooling load using the ANN prediction model were close to the actual data, even when the next-day weather was forecast. This confirms that the ANN model has sufficient accuracy and is the correct method for practical utilization in HVAC system control, thermal storage optimal control, and load/demand management.« less