ISSN (Online) 2321 – 2004

ISSN (Print) 2321 – 5526

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN ELECTRICAL, ELECTRONICS, INSTRUMENTATION AND CONTROL ENGINEERING
Vol. 3, Issue 7, July 2015

Artificial Intelligence in Power Station

Reshmi Banerjee
Assistant Professor, Department of Electrical Engineering, Guru Nanak Institute of Technology, Kolkata, India

Abstract: Artificial intelligence is the science of automating intelligent behaviours currently achievable by humans.
Power system has grown tremendously over a few decades. As the size and complexity of the power system consisting
of generators, transmission lines, power transformers, distribution transformers etc. increases the possibility of inviting
faults. The acquisition of data, the processing of those data for use by the operator, and control of remote devices are
the fundamental building blocks of all modern utility control systems. Manual calculations, technical analysis and
conclusions initially adopted the power system design, operation and control. As the power system grew it become
more complex due to the technical advancements, variety and dynamic requirements.

Keywords: Artificial intelligence, Expert system, Artificial neural network, Fuzzy logic, Power station.

I. INTRODUCTION
There are three types of major power plants known for the Advantages of artificial neural networks :
massive electricity generation :  Speed of processing.
i) Thermal power plants, ii) Hydal power plants,  They do not need any appropriate knowledge of the
iii) Nuclear power plants. system model.
One may expect that the mobile sensing will play an  They have the ability to handle situations of incomplete
increasingly important role in the monitoring of power data and information, corrupt data.
system. Artificial intelligence is known to be the  They are fault tolerant.
intelligence exhibited by machines and software, for  Artificial neural networks are fast and robust.
example, robots and computer programs. Disadvantages of artificial neural networks :
An expert system obtains the knowledge of a human  Large dimensionality.
expert in a narrow specified domain into a machine  Results are always generated even if the input data are
implementable form. Expert systems are unable to learn or unreasonable.
adopt to new problems or situations. Expert systems are  They are not scalable i.e. once an artificial neural
also called as knowledge based systems or rule based network is trained to do certain task, it is difficult to
systems. Expert systems are computer programs which extend for other tasks without retraining the neural
have proficiency and competence in a particular field. network.
Artificial neural networks are biologically inspired Fuzzy logic or fuzzy systems are logical systems for
systems which convert a set of inputs into a set of outputs
standardisation and formalisation of approximate
by a network of neurons, where each neuron produces one
reasoning. It is similar to human decision making with an
output as a function of inputs. A fundamental neuron can
be considered as a processor which makes a simple non ability to produce exact and accurate solutions from
linear operation of it’s inputs producing a single output. certain or even approximate information and data. Fuzzy
They are classified by their architecture : number of layers logic is the way like which human brain works, and we
and topology : connectivity pattern, feed forward or can use this technology in machines so that they can
recurrent. perform somewhat like humans.

II. METHODOLOGY
There are mainly three techniques : i)Expert system
techniques, ii)Artificial neural networks, iii)Fuzzy logic
systems.
 Since expert systems are basically computer programs, the
process of writing codes for these programs is simpler than
actually calculating and estimating the value of parameters
used in generation, transmission and distribution.
 Any modifications even after design can be easily done
because they are computer programs.

 As artificial neural networks operate on biological
institutes and perform biological evaluation of real world
problems, the problems in generation, transmission and
distribution of electricity can be fed to the artificial neural
Fig. 1 : Artificial neural system networks so that a suitable solution can be obtained.

Copyright to IJIREEICE DOI 10.17148/IJIREEICE.2015.3717 86

 ISSN (Online) 2321 – 2004
ISSN (Print) 2321 – 5526

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN ELECTRICAL, ELECTRONICS, INSTRUMENTATION AND CONTROL ENGINEERING
Vol. 3, Issue 7, July 2015

 Given the constraints of a practical transmission and  Control of power system like voltage control, stability
distribution system, the exact values of parameters can control, power flow control, load frequency control.
be determined.  Control of power plants like fuel cells power plant
 For example, the value of inductance, capacitance and control, thermal power plant control.
resistance in a transmission line can be numerically  Automation of power system like restoration,
calculated by artificial neural networks taking in various management, fault diagnosis, network security.
factors like environmental factors, unbalancing  Can be used in anything from small circuits to large
conditions, and other possible problems. mainframes.
 Fuzzy logic can be used for designing the physical  Can be used to increase the efficiency of the
components of power systems. components used in power systems.
 As most of the data used in power system analysis are
approximate values and assumptions, fuzzy logic can be
of great use to derive a stable, exact and ambiguity free
output.

IV. CONCLUSION
A reliable, continuous supply of electrical energy is
essential for the functioning of today’s modern complex
and advanced society. Electricity is one of the prime
factors for the growth and determines the value of the
society. So, implementation of artificial intelligence is
very important in power system.

REFERENCES
Fig. 2 : Fuzzification [1] Stuart Russell, Peter Norvig, Artificial Intelligence : A Modern
Approach, Pearson.
III. RESULT [2] Jeff Heaton, Artificial Intelligence for Humans, Volume 1 :
Fundamental Algorithms, Create Space Independent Publishing
 Replacing human workers for dangerous and highly Platform.
[3] Philip C. Jackson Jr., Introduction to Artificial Intelligence, Dover
specialized operations, such as live maintenance of high Publications.
voltage transmission lines, has been a long standing [4] Keith Frankish, William M. Ramsey, The Cambridge Handbook of
effect in the power community. Artificial Intelligence, Cambridge University Press.
 Operation in hazardous environments, such as [5] Kevin Warwick, Artificial Intelligence : The Basics, Routledge.
[6] Patrick Henry Winston, Artificial Intelligence, Pearson.
radioactive locations in nuclear plants, access to tight [7] George F. Luger, Artificial Intelligence : Structures and Strategies for
spaces, such as cable viaducts and cooling pipes, and Complex Problem Solving, Pearson.
precise positioning of measurement equipment. [8] Stephen Lucci, Danny Kopec, Artificial Intelligence in the 21st
 Expert systems use the interface mechanism and Century, Mercury Learning & Information.
[9] David L. Poole, Alan K. Mackworth, Artificial Intelligence :
knowledge to solve problems which cannot be or Foundations of Computational Agents, Cambridge University
difficult to be solved by human skill and intellect. Press.
 Results are permanent and consistent. [10] Mohamad Hassoun, Fundamentals of Artificial Neural Networks, A
Bradford Book.
 Can be easily documented. [11] Sandhya Samarasinghe, Neural Networks for Applied Sciences and
 Results can be easily transferred and reproduced. Engineering, Auerbach Publications.
 The understanding of the working of neurons and the [12] Martin T Hagan, Howard B Demuth, Mark H Beale, Neural Network
Design, Martin Hagan.
pattern of their interconnection can be used to construct [13] Laurene V. Fausett, Fundamentals of Neural Networks :
computers for solving real world problems of Architectures, Algorithms and Applications, Pearson.
classification of patterns and pattern recognition. [14] Simon Haykin, Neural Networks : A Comprehensive Foundation,
 Fuzzification provides superior expressive power, Prentice Hall.
[15] James A. Anderson, An Introduction to Neural Networks, A
higher generality and an improved capability to model Bradford Book.
complex problems at low or moderate solution cost. [16] Timothy J. Ross, Fuzzy Logic with Engineering Applications, Wiley.
 Stability analysis and enhancement. [17] Hung T. Nguyen, Elbert A. Walker, A First Course in Fuzzy Logic,
 Power system control. Chapman and Hall/CRC.
[18] Masao Mukaidono, Fuzzy Logic for Beginners, Wspc.
 Fault diagnosis. [19] P.K.Nag, Power Plant Engineering, Tata Mc Graw Hill Education.
 Load forecasting. [20] Prabha Kundur, Power System Stability and Control, Mc Graw Hill
 Reactive power planning and its control. Education.
[21] Allen J. Wood, Bruce F. Wollenberg, Gerald B. Sheble, Power
 Operation of power system like unit commitment, Generation, Operation and Control, Wiley-Interscience.
hydro-thermal coordination, economic dispatch, [22] Robert Miller, James Malinowski, Power System Operation, Mc
congestion management, maintenance scheduling, state Graw Hill Education.
[23] Mark Fennell, Transmission of Electrical Power Explained Simply,
estimation, load and power flow. Create Space Independent Publishing Platform.
 Planning of power system like generation expansion [24] B.M.Weedy, B.J.Cory, N.Jenkins, Electric Power Systems, Wiley.
planning, power system reliability, transmission
expansion planning, reactive power planning.

Copyright to IJIREEICE DOI 10.17148/IJIREEICE.2015.3717 87