See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/335990489

An Overview of Deep Learning in Smart Grids

Article · September 2019

CITATION READS

1 2,682

2 authors:

Bhanu PRAKASH Kolla Rudra Kalyan Nayak

Koneru Lakshmaiah Education Foundation VIT Bhopal University, Madhya Pradesh
185 PUBLICATIONS 2,046 CITATIONS 79 PUBLICATIONS 491 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Rudra Kalyan Nayak on 19 October 2020.

The user has requested enhancement of the downloaded file.

 An Overview of Deep Learning in Smart Grids
Dr. Rudra Kalyan Nayak ([email protected])
Koneru Lakshmaiah Education Foundation (Deemed to be University)
Dr. Kolla Bhanu Prakash ([email protected])
Koneru Lakshmaiah Education Foundation (Deemed to be University)
Dr. Ramamani Tripathy ([email protected])
United School of Business Management, Odisha, India

Introduction

In recent years, with the growth of the computational methods particularly in case of data management and analysis several
machine learning (ML) approaches has been implemented in various industries. According to the current situation most of
the researchers have concentrated their studies on deep learning (DL) because deep learning has been treated as one of the
emerging areas for feature extraction and handling huge amount of data where machine learning methods fail. Altogether,
artificial intelligence encloses with numerous subfields, counting as machine learning, deep learning, computer vision,
neural network and natural language processing etc. Deep learning utilizes massive neural networks with a lot of layers of
processing units, for advances in computing power and enhanced training techniques to learn versatile patterns from huge
quantity of data. Common applications include image and speech recognition. Deep learning is implemented through neural
network. The motivation behind neural network is the biological neuron. Generally, deep learning is a sub-branch of
machine learning and machine learning is also another sub-branch of artificial intelligence as given in the figure1.

Figure 1: Depiction of AI, ML, DL

The biggest advantages of deep learning are:

• Deep learning gives better performance on dissimilar troubles that significantly outperforms other solutions in
diverse domains. This take in speech recognition, language, computer vision, playing games etc.
• Deep learning overcomes the limitation of machine learning methods in case of feature extraction that means it takes
less time.
• For adopting new type of problems in upcoming days deep learning architecture performs well.
• Robustness to natural variations in the data is automatically learned.
o
Importance of deep learning over machine learning

• Machine learning is not enough capable for managing high dimensional data wherever input & output
is relatively large.
• Deep learning has sufficient power for managing the high dimensional data as well as to focus on the exact features by
themselves. This procedure is known as feature mining.
• Deep learning can do better than traditional methods. For example, in image classification, deep learning algorithms
gave 41percent more correct result than machine learning algorithm and gave 27 percent more accurate in facial
recognition and lastly 25 percent in case of voice recognition.

Deep learning works in smart grids

For electricity data analysis industries were adopted novel methods like machine learning, fuzzy logic, data mining,
artificial neural network (ANN),support vector machine (SVM) and genetic algorithm etc. to get better outcomes for
estimating the exact electricity demand and also used these methods to forecast both energy production and consumption.
Among all methods deep learning plays a vital role in smart grids applications.

IEEE India Info. Vol. 14 No. 3 Jul – Sep 2019 Page 122
Smart Grids (SG)

According to digital technology “grid" is recognized as the electric grid and it is bi-directional communication line of
network between consumers and utility. Now a days electricity disturbance is common problems which cause a series of
failures such as in banking sector, communications, traffic, and security etc. For this purpose smart grid technology is
implemented based on varieties of artificial intelligence approach. This scheme gives permission for supervising,
investigate, organize and communication within the supply chain to assist progress effectiveness, shrink energy utilization
and cost, and make best use of the transparency and reliability of the energy supply chain. The smart grid system
overcomes the drawbacks of traditional electrical grids where smart net meters concept has been implemented in earlier
days. The concept of smart grid approach not only concentrates on utilities and technologies but also focuses on the
consumer choices regarding their consumption of electricity energy.

The layout of smart grids is as follows in figure 2.

Figure 2: Powers ahead with smart grids

Smart grid is totally a new concept for all researchers in comparison with traditional electricity grids. The smart grid is the
renovation of the electricity flow system. So this theme takes a significant role in the research society. Smart grid is the
mixture of information and all digital communication approaches by means of power grid systems to facilitate two way
communication and power flow that can boost safety, reliability, and effectiveness of the power scheme. In addition we can
say that smart grid is the arrangement of tools, systems and procedure to manufacture power grid intelligent and
computerized. Due to huge amount of data collection in each and every moment from different sources, it is necessary for
all power companies to take responsibilities of all collected data and find out the clear knowledge about electricity
consumption behaviors which is shown in block diagram 3 below. So deep learning algorithm is required for feature
extraction and data analysis. Most of the studies were done on smart meter data for instance load outline, customer
classification, load predicting and anomaly detection. For doing improved understanding of electricity utilization, flexible
demand management and useful energy control smart meter data analysis are essential.

According with the advancement of smart grids the United Stated Department of Energy recommend four types of
technology such as:

• Technologies regarding measurement and intelligent.

• Another approach is automated flow of communication among constituent of the electric grid and their Integrations.
• Computerized controls for division and repairs.
• Enhanced organization management dashboards and decision support software

IEEE India Info. Vol. 14 No. 3 Jul – Sep 2019 Page 123
 Figure3: Block diagram of smart grids

RES, DER and MSW stand for renewable energy source, distributed energy resource and municipal solid waste
respectively in figure 3.

The important benefits related with the smart grid include:

• Electricity flows of transmission are much more proficient.

• When power disturbances of electricity arise smart grids approach gives quicker resolution.
• Smart grid approach provides minimum power and operation cost for consumers.
• Reduced peak demand, which will also help lower electricity rates.
• Improved bonding of customer-owner power generation systems, counting renewable energy systems
• Enhanced the security feature.

Challenges of smart grids

Although smart grid has numerous profits still there are lots of challenges in its achievement. Smart grids occupy a lot of
stakeholders who are in charge of dissimilarity aspect of the energy system and sorry to say, the synchronization and rate of
adoption of the novel tools happens at diverse paces. The vast investments required for together the infrastructure and new
devices are also a hurdle to adoption. Lastly, regulatory obstacle and jurisdictional topic issues present important
challenges that must be overcome.

Conclusion

From literature studies, it is revealed that most of the researchers concentrated their experiment on smart grids using
artificial learning and machine learning methods. Still it is a challenging factor for all. Therefore a novel enhanced model
has been adopted based on deep learning to solve more complex problems of smart grids which have not been yet solved by
machine learning methods. By increasing the number of layers in the well-known artificial neural networks, nowadays we
are referring to deep neural networks as the principal direction in deep learning. This algorithm has immense application
potential in case of smart grids but researchers are still struggling on their studies on deep learning techniques due to many
hidden layers are included in this algorithm. Numerous problems required to be studied in deepness, for example, for a
specific use case which structure of deep learning is preferable and how many layers are appropriate for the particular
problem and also concentrate on security of power system.

References

[1] Gungor, V.C.; Sahin, D.; Kocak, T.; Ergut, S.; Buccella, C.; Cecati, C.; Hancke, G.P. Smart grid
technologies: Communication technologies and standards. IEEE Trans. Ind. Inform. 2011, 7, 529–539.

IEEE India Info. Vol. 14 No. 3 Jul – Sep 2019 Page 124
[2] Tureczek, A.M.; Nielsen, P.S. Structured literature review of electricity consumption classification using
smart meter data. Energies 2017, 10, 584.
[3] Hong, Y.-Y.; Chen, Y.-C. Application of algorithms and artificial–intelligence approach for locating multiple
harmonics in distribution systems. IEE Proc. Gener.Transm.Distrib. 1999, 146, 325–329.
[4] Liu, J.; Li, X.; Liu, D.; Liu, H.; Mao, P. Study on Data Management of Fundamental Model in Control Center
for Smart Grid Operation. IEEE Trans. Smart Grid 2011, 2, 573–579.
[5] Liu, C.-X.; Zeng, Q.A.; Liu, Y. A Dynamic Load Control Scheme for Smart Grid Systems. Energy
Procedia2011, 12, 200–205.
[6] Zhang Ni, A Survey of the Development and Application of Artificial Intelligence Technology[J]. Coal Mine
Machinery, 2009,(02): 4-7.
[7] Bo Li, ZhiyuanGao. Application analysis and prospect of artificial intelligence technology in smart grid[J].
Electric Power, 2017,50(12):136-140.
[8] Bengio Y. Learning Deep Architectures for AI[J]. Foundations & Trends in Machine Learning,2009.
[9] Shihua Wen. Application of Computer Science in Smart Grid[J]. China High Technology Enterprises,
2016(21):47-49.
[10] Chengxiang Zhu, Junhua Yang, Sizhe Chen, ZhihuiLuo. Demand response technology based on artificial
intelligence theory in the context of smart grid[J]. Shaanxi Electric Power, 2015, 43(07): 63-69.
[11] H. Xu, H. Huang, R. S. Khalid, and H. Yu, “Distributed machine learning based smart-grid energy
management with occupant cognition,” 2016 IEEE Int. Conf. Smart Grid Commun.SmartGridComm 2016,
pp. 491–496, 2016.
[12] Zhang, Dongxia, Xiaoqing Han, and Chunyu Deng. "Review on the research and practice of deep learning
and reinforcement learning in smart grids." CSEE Journal of Power and Energy Systems 4.3 (2018): 362-370.

About the authors

Dr. Rudra Kalyan Nayak is presently working as Associate Professor in the Department of Computer
Science & Engineering at Koneru Lakshmaiah Education Foundation (Deemed to be University),
Vaddeswaram, Andhra Pradesh, India. He has got his M.Tech in Information Technology and Ph.D in
Computer Science & Engineering from Siksha ‘O’ Anusandhan (Deemed to be) University, Odisha,
India. He has more than 10 years of teaching, mentoring and research experience. Dr. Nayak has
interests in teaching the subjects Design and Analysis of Algorithms, Artificial Intelligence & Expert
Systems, Theory of Computation, Compiler Design & Language Processor, Computer Graphics &
Multimedia, Perception & Computer Vision. His research interest lies in the field of Financial Engineering and
Bioinformatics using various intelligent data mining and Machine Learning techniques. His intellectual contributions
include publication of 3 patents, 9 publications in peer reviewed international journals and conferences like Elsevier,
Springer, IOS Press. He is professional member of ISTE and ACM.

Dr. Kolla Bhanu Prakash is working as Professor and Research Group Head for A.I and Data Sciences
Research group in CSE DEPARTMENT, KoneruLakshmaiah Education Foundation. Dr.
KollaBhanuPrakash has 12+ years of experience working in the academia, research, teaching,
academic administration. His current research interests include Machine learning, Deep Learning,
Media Mining, Image Processing, Smart Grids and Natural Language Processing. He was the recipient
of Best Speaker award during M.Sc. He has authored over 4 books and 36 research papers in various
national and international journals and conferences. His publications are indexed in Web of Science,
Scopus, DBLP and Google scholar. He is Senior Member – IEEE.

Dr. Ramamani Tripathy is presently working as Assistant Professor of department of Master of

Computer Application at United School of Business Management in Bhubaneswar, Odisha, India. She
has 10+ years of teaching, mentoring research and academic experience at USBM. She has a Ph.D in
Computer Science & Engineering (CS&E) from SOA University, Odisha. Prof. Tripathy’s teaching
interests are Computer Networks, Operating System, Distributed system, Computer Graphics, Data
mining, and Unified Modeling Language. Her research interest lies in the field of Bioinformatics
using various soft computing techniques. Her intellectual contributions include 9 publications in
leading journals and conferences.

Education is the passport to the future, for tomorrow belongs to those who prepare for it today. – Malcolm X

The roots of education are bitter, but the fruit is sweet. – Aristotle

Education is what remains after one has forgotten what one has learned in school. – Albert Einstein

IEEE India Info. Vol. 14 No. 3 Jul – Sep 2019 Page 125

View publication stats