A Data-Driven Approach for Generating Synthetic Load Patterns and Usage Habits

Kababji, Samer El; Srikantha, Pirathayini

doi:10.1109/tsg.2020.3007984

Cited by 35 publications

(13 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, considering the high complexity of the above mentioned model-based methods, the data-driven methods become a favorable replacement as they do not require prior knowledge. [11] proposes a GAN-based scheme to generate synthetic labeled load patterns and usage habits, which requires no model assumptions. Furthermore, [12] introduces a model-free method for scenario generation of smart grid, and GAN is used to capture the spatial and temporal correlations of renewable power plants.…”

Section: Related Workmentioning

confidence: 99%

Variational Autoencoder Generative Adversarial Network for Synthetic Data Generation in Smart Home

Razghandi¹,

Zhang²,

Erol‐Kantarci³

et al. 2022

Preprint

View full text Add to dashboard Cite

Data is the fuel of data science and machine learning techniques for smart grid applications, similar to many other fields. However, the availability of data can be an issue due to privacy concerns, data size, data quality, and so on. To this end, in this paper, we propose a Variational AutoEncoder Generative Adversarial Network (VAE-GAN) as a smart grid data generative model which is capable of learning various types of data distributions and generating plausible samples from the same distribution without performing any prior analysis on the data before the training phase. We compared the Kullback-Leibler (KL) divergence, maximum mean discrepancy (MMD), and Wasserstein distance between the synthetic data (electrical load and PV production) distribution generated by the proposed model, vanilla GAN network, and the real data distribution, to evaluate the performance of our model. Furthermore, we used five key statistical parameters to describe the smart grid data distribution and compared them between synthetic data generated by both models and real data. Experiments indicate that the proposed synthetic data generative model outperforms the vanilla GAN network. The distribution of VAE-GAN synthetic data is the most comparable to that of real data.

show abstract

Section: Related Workmentioning

confidence: 99%

Variational Autoencoder Generative Adversarial Network for Synthetic Data Generation in Smart Home

Razghandi¹,

Zhang²,

Erol‐Kantarci³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Robust non-parametric regression is used in Mateos and Giannakis ( 2012 ) with application to electrical load curves. More recent solutions are indicated in Tang et al ( 2014 ) with the introduction of the portrait data, and in El Kababji and Srikantha ( 2020 ), among which the Generative Adversarial Networks together with a kernel density estimator are run on individual appliances. The various steps of data pre-processing can be combined a comprehensive approach that includes time synchronisation, noise cleansing, missing data imputation and performance assessment (Martinez-Luengo et al, 2019 ).…”

Section: Data Qualitymentioning

confidence: 99%

Data Consistency for Data-Driven Smart Energy Assessment

Chicco

2021

Front. Big Data

View full text Add to dashboard Cite

In the smart grid era, the number of data available for different applications has increased considerably. However, data could not perfectly represent the phenomenon or process under analysis, so their usability requires a preliminary validation carried out by experts of the specific domain. The process of data gathering and transmission over the communication channels has to be verified to ensure that data are provided in a useful format, and that no external effect has impacted on the correct data to be received. Consistency of the data coming from different sources (in terms of timings and data resolution) has to be ensured and managed appropriately. Suitable procedures are needed for transforming data into knowledge in an effective way. This contribution addresses the previous aspects by highlighting a number of potential issues and the solutions in place in different power and energy system, including the generation, grid and user sides. Recent references, as well as selected historical references, are listed to support the illustration of the conceptual aspects.

show abstract

“…If the dimensionality reduction algorithm is adopted, if the user is only interested in certain user characteristics, the displayed information is also incomplete. Multisource data fusion is the basis of computational intelligence and is the embodiment of quantification, automation, and fusion thinking [20]. e process of conducting decision research decomposes the task into different subtasks, constructs a multisource data fusion model for the data sources of the subtasks, and is oriented to the task context, and the process of data collection, processing, and analysis should be oriented to the context in which the subtasks are located, covering all the data sources required for the task.…”

Section: Visual Communication Design Model Designmentioning

confidence: 99%

A Study of Big-Data-Driven Data Visualization and Visual Communication Design Patterns

Zhu

2021

Scientific Programming

View full text Add to dashboard Cite

This paper provides an in-depth study and analysis of big-data-driven data visualization and visual communication design models. The characteristics of new media and the definition of traditional media are analyzed; the importance of the new media environment is derived through comparison; and the successful cases of new media integration today are analyzed. In this process, we will optimize the traditional science and technology intelligence service model, optimize the various components that make up the science and technology intelligence wisdom service, achieve model optimization and reflect the four characteristics of science and technology intelligence wisdom service, and reconstruct the science and technology intelligence wisdom service using the literature research method. The design based on imagery schema theory is manifest, inclusive, and somewhat innovative and at the same time has a high degree of consistency and internal logical relationship with the visual representation of multidomain heterogeneous data at the cognitive level and displays purpose. This internal logical relationship is systematically organized and deeply analyzed, and the methodology from subpattern extraction and visual interaction design to the deep integration of visual representation is proposed in combination with specific application scenarios and cases.

show abstract

A Data-Driven Approach for Generating Synthetic Load Patterns and Usage Habits

Cited by 35 publications

References 20 publications

Variational Autoencoder Generative Adversarial Network for Synthetic Data Generation in Smart Home

Variational Autoencoder Generative Adversarial Network for Synthetic Data Generation in Smart Home

Data Consistency for Data-Driven Smart Energy Assessment

A Study of Big-Data-Driven Data Visualization and Visual Communication Design Patterns

Contact Info

Product

Resources

About