Spatial-temporal load forecasting using AMI data

Xu, Jin; Yue, Meng; Katramatos, Dimitri; Yoo, Shinjae

doi:10.1109/smartgridcomm.2016.7778829

Cited by 10 publications

(10 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, we compared with two traditional autoregressive (AR) models, knmV-AR [32] and NCST-LF [33], which also explored the load spatiotemporal correlation, in order to suggest the advantages of the proposed scheme in the utilization of spatiotemporal data. For the two AR models, we replaced their meter measured data with zones' data and used the columns of the created HLM as the historical load vector.…”

Section: Comparison Methodsmentioning

confidence: 99%

“…Due to the close geographic distance and the same time, there will be a strong spatiotemporal correlation between the HLM elements [32], which is a prerequisite for the matrix to have (approximately) low rank and will be demonstrated in the next subsection.…”

Section: Historical Load Tensormentioning

confidence: 97%

“…According to the existing research [32][33][34], the daily load curves for several consecutive days in a given zone have consistent trends due to similar external factors such as weather, human behavior, and infrastructure. Therefore, daily load ahead of the forecasted hour for several consecutive days must contain a large amount of information for the inference.…”

Section: Model Input Datamentioning

confidence: 97%

See 2 more Smart Citations

Spatiotemporal Feature Learning Based Hour-Ahead Load Forecasting for Energy Internet

Zhang

Wang

2020

Electronics

View full text Add to dashboard Cite

In this paper, we analyze the characteristics of the load forecasting task in the Energy Internet context and the deficiencies of existing methods and then propose a data driven approach for one-hour-ahead load forecasting based on the deep learning paradigm. The proposed scheme involves three aspects. First, we formulate a historical load matrix (HLM) with spatiotemporal correlation combined with the EI scenario and then create a three-dimensional historical load tensor (HLT) that contains the HLMs for multiple consecutive time points before the forecasted hour. Second, we preprocess the HLT leveraging a novel low rank decomposition algorithm and different load gradients, aiming to provide a forecasting model with richer input data. Third, we develop a deep forecasting framework (called the 3D CNN-GRU) featuring a feature learning module followed by a regression module, in which the 3D convolutional neural network (3D CNN) is used to extract the desired feature sequences with time attributes, while the gated recurrent unit (GRU) is responsible for mapping the sequences to the forecast values. By feeding the corresponding load label into the 3D CNN-GRU, our proposed scheme can carry out forecasting tasks for any zone covered by the HLM. The results of self-evaluation and a comparison with several state-of-the-art methods demonstrate the superiority of the proposed scheme.

show abstract

Section: Comparison Methodsmentioning

confidence: 99%

Section: Historical Load Tensormentioning

confidence: 97%

Section: Model Input Datamentioning

confidence: 97%

See 1 more Smart Citation

Spatiotemporal Feature Learning Based Hour-Ahead Load Forecasting for Energy Internet

Zhang

Wang

2020

Electronics

View full text Add to dashboard Cite

show abstract

“…The random variation is usually based on fluctuations observed in historical data for a selected period [15]. Depending on the data selected and used to train the model, two groups of approaches can be identified [16][17][18][19][20][21][22][23][24][25]: 3. Hybrid models: They represent any combination of two or more of the methods described above [17,25].…”

Section: Prediction Modelsmentioning

confidence: 99%

Modeling System Based on Machine Learning Approaches for Predictive Maintenance Applications

Martins

Rodrigues

Henriques

2020

KEG

View full text Add to dashboard Cite

Industry 4.0 must respond to some challenges such as the flexibility and robustness of unexpected conditions, as well as the degree of system autonomy, something that is still lacking. The evolution of Industry 4.0 aims at converting purely mechanical machines into machines with self-learning capacity in order to improve overall performance and contribute to the optimization of maintenance. An important contribution of Industry 4.0 in the industrial sector is predictive maintenance and prescriptive maintenance. This article should be analysed as a methodology proposal to implement an automatic forecasting model in a test bench for the recognition of a machine’s failure and contribute to the development of algorithms for preventive and descriptive maintenance. Keywords: Industry 4.0, Artificial intelligence, Machine learning, Predictive maintenance, Prescriptive maintenance

show abstract

“…Spatial-temporal analyses are also used to forecast energy demand; for instance, authors in [8] provided a framework that facilitates the exploitation of lowdimensional structures to govern the interactions between the surrounding residential SM users. In [9], a k-nearest Vector Autoregressive framework with exogenous input for models with spatial-temporal variation of electricity consumption in individual household load forecasting was proposed. Another example is the work presented in [10], where authors studied how an electrical load is distributed in a city using area-independent agents and the relationships among neighbouring areas.…”

Section: Introductionmentioning

confidence: 99%

Definition of Residential Power Load Profiles Clusters Using Machine Learning and Spatial Analysis

2021

View full text Add to dashboard Cite

This study presents a novel approach for discovering actionable knowledge and exploring data-based models from data recorded by household smart meters. The proposed framework is supported by a machine learning architecture based on the application of data mining methods and spatial analysis to extract temporal and spatial restricted clusters of characteristic monthly electricity load profiles. In addition, it uses these clusters to perform short-term load forecasting (1 week) using recurrent neural networks. The approach analyses a database with measurements of 1000 smart meters gathered during 4 years in Guayaquil, Ecuador. Results of the proposed methodology led us to obtain a precise and efficient stratification of typical consumption patterns and to extract neighbour information to improve the performance of residential energy consumption forecasting.

show abstract

Spatial-temporal load forecasting using AMI data

Cited by 10 publications

References 17 publications

Spatiotemporal Feature Learning Based Hour-Ahead Load Forecasting for Energy Internet

Spatiotemporal Feature Learning Based Hour-Ahead Load Forecasting for Energy Internet

Modeling System Based on Machine Learning Approaches for Predictive Maintenance Applications

Definition of Residential Power Load Profiles Clusters Using Machine Learning and Spatial Analysis

Contact Info

Product

Resources

About