Potentials and Promises of Computational Intelligence for ...brain2grid.com/wp-content/uploads/2008/12/... · CI methods can contribute to electric power systems and smart grids in

1

Abstract-- The electric power grid is a complex adaptive system

under semi-autonomous distributed control. It is spatially and temporally complex, non-convex, nonlinear and non-stationary with a lot of uncertainties. The integration of renewable energy such as wind farms, and plug-in hybrid and electric vehicles further adds complexity and challenges to the various controllers at all levels of the power grid. A lot of efforts have gone into the development of a smart grid to align the interests of the electric utilities, consumers and environmentalists. Advanced computational methods are required for planning and optimization, fast control of power system elements, processing of field data and coordination across the grid. Distributed and coordinated intelligence at all levels and across levels of the electric power grid – generation, transmission and distribution is inevitable if a true smart grid is to be reality. Computational intelligence (CI) is the study of adaptive mechanisms to enable or facilitate intelligent behavior in complex, uncertain and changing environments. These adaptive mechanisms include artificial and bio-inspired intelligence paradigms that exhibit an ability to learn or adapt to new situations, to generalize, abstract, discover and associate. The paradigms of CI mimic nature for solving complex problems. CI is successor of artificial intelligence and is the way of the future computing. This paper presents the potentials and promises of CI to realize an intelligent smart grid.

Index Terms—adaptive control, adaptive critic designs,

computational intelligence, FACTS, plug-in vehicles, renewable energy, wide area monitoring and control.

I. INTRODUCTION

HE North American electric power grid is the world’s largest single machine ever built by man. It is a complex adaptive system under semi-autonomous control. It is

spatially and temporally complex, nonconvex, nonlinear and non-stationary with uncertainties at many levels. Its efficient, reliable and safe operation and control is a challenge today as seen from the 2003 Northeast American Blackout [1]. The complexity grows with growing demand to integrate of renewable sources of energy such as wind and solar farms.

To overcome the challenges brought about complexity, shear size of the network and others, the power system needs intelligence at all levels – horizontally and vertically, be it at component, area, or system level. This means intelligent

This work was supported by the National Science Foundation, USA under

CAREER Grant ECCS # 0348221 and EFRI # 0836017. G. K. Venayagamoorthy is the Director of the Real-Time and Intelligent

Systems Laboratory, Missouri University of Science and Technology, Rolla, MO 65409-0249 USA (e-mail: [email protected]).

information processing is needed everywhere a critical decision is made for performance control, locally or globally. For this to be possible, large number of distributed sensors and actuators are required on the network. Equipped with features to handle this complexity, the traditional grid is elevated to or referred to as a smart grid.

The traditional way of modeling, control and optimization needs to be augmented, or even replaced in some cases, with intelligent techniques capable of rapid adaptation, having dynamic foresight, being fault-tolerant and robust to disturbances and randomness. This paper describes the potentials and promises of the computational intelligence for smart grid operation and control.

II. SMART GRID

The smart grid can be viewed as a digital upgrade of the existing electricity infrastructure to allow for dynamic optimization of current operations as well as incorporate dynamic gateways for alternative sources of energy production. The telecommunication industry today is already using ‘smart’ technologies, some of which can be borrowed for smart operation of the electricity industry. The objectives of a smart grid are to minimize the cost of energy and reduce emissions.

A smart grid or sometimes referred to as the Intelligent Grid/Intelligrid and FutureGrid must have certain basic functions for modernization of the grid (as indicated in the Energy Independence and Security Act of 2007), including: • Have a self-healing capability. • Be fault-tolerant by resisting attacks. • Allow for integration of all energy generation and storage

options including plug-in vehicles. • Allow for dynamic optimization of grid operation and

resources with full cyber-security. • Allow for incorporation of demand-response, demand-

side resources and energy-efficient resources. • Allow electricity clients to actively participate in the grid

operations by providing timely information and control options.

• Improve reliability, power quality, security and efficiency of the electricity infrastructure.

In order to carry out the functions mentioned above,

advanced monitoring, forecasting, decision making, control and optimization algorithms are required. These algorithms must be fast, scalable and dynamic. In this paper,

Potentials and Promises of Computational Intelligence for Smart Grids

Ganesh K. Venayagamoorthy, Senior Member, IEEE

T

978-1-4244-4241-6/09/$25.00 ©2009 IEEE

Authorized licensed use limited to: Missouri University of Science and Technology. Downloaded on January 25, 2010 at 00:56 from IEEE Xplore. Restrictions apply.

2

computational intelligence based algorithms that deliver the above mentioned features for smart grid operations are discussed.

III. COMPUTATIONAL INTELLIGENCE METHODS

Computational intelligence (CI) is the study of adaptive mechanisms to enable or facilitate intelligent behavior in complex, uncertain and changing environments. These adaptive mechanisms include those bio-inspired and artificial intelligence paradigms that exhibit an ability to learn or adapt to new situations, to generalize, abstract, discover and associate [2]. The typical paradigms of CI are illustrated in Fig. 1 [3]. These paradigms can be combined to form hybrids as shown in Fig. 1 resulting in Neuro-Fuzzy systems, Neuro-Swarm systems, Fuzzy-PSO systems, Fuzzy-GA systems, Neuro-Genetic systems, etc.

Neuro-FuzzySystems

Neuro-GeneticSystems

Neuro-SwarmSystems

Fuzzy-PSO Systems

Fuzzy-GASystems

Evolutionary-SwarmSystems

Immunized-SwarmSystems

Immunized-NeuroSystems

SwarmIntelligence

Evolutionarycomputing

Fuzzy SystemsFuzzy Systems

Neural Networks

ImmuneSystems

Neuro-FuzzySystems

Neuro-GeneticSystems

Neuro-SwarmSystems

Fuzzy-PSO Systems

Fuzzy-GASystems

Evolutionary-SwarmSystems

Immunized-SwarmSystems

Immunized-NeuroSystems

SwarmIntelligence

Evolutionarycomputing

Fuzzy SystemsFuzzy Systems

Neural Networks

ImmuneSystems

Fig. 1. Five main CI paradigms and typical hybrids.

Other advanced CI techniques include adaptive critic designs [4], reinforcement learning and approximate dynamic programming [5]. Adaptive critic designs (ACDs) use neural networks based designs for optimization over time using combined concepts of reinforcement learning and approximate dynamic programming [4-5]. ACDs use two neural networks, the critic and action networks, to solve the Hamilton-Jacobi-Bellman equation of optimal control. The critic network approximates the cost-to-go function J of Bellman’s equation of dynamic programming (1) and is referred to as the heuristic dynamic programming (HDP) approach in ACDs,

∑∞

=+=

1

)()(k

k ktUtJ γ , (1)

where γ is a discount factor between 0 and 1, and U(t) is a utility function or a local performance index. The action network provides optimal control to minimize or maximize the cost-to-go function J. Fig. 2 shows the HDP based ACD approach. There are several other members of the ACD family that vary in complexity and power [4].

Y(t)

J(t)ACTION

NETWORK

SystemModel

TDL

TDL

CRITICNETWORK

1

Yref

Electricity Infrastructure

RealReal--Time Intelligent Time Intelligent TechnologiesTechnologies

Y(t)

J(t)ACTION

NETWORK

SystemModel

TDL

TDL

CRITICNETWORK

1

Yref

Electricity Infrastructure Y(t)

J(t)ACTION

NETWORK

SystemModel

TDL

TDL

CRITICNETWORK

1

Yref

Yref

Electricity Infrastructure

RealReal--Time Intelligent Time Intelligent TechnologiesTechnologies

Fig. 2. A typical adaptive critic design framework.

IV. CI METHODS FOR SMART GRIDS

CI and advanced CI techniques have been applied to solving challenging problems today in electric power systems. CI methods can contribute to electric power systems and smart grids in a number of ways as illustrated in Fig. 3. A few of these problems addressed by author and other researchers are discussed below. Many of these solutions are directly applicable to smart grids – modernization of the grid using full digital technology.

Self-Healing

BehavioralModeling

Fast and AccurateDecisionMaking

Computational IntelligenceComputational IntelligenceTechniquesTechniques

Identification/Prediction

ofNonlinearDynamics

Robust/Optimal/Adaptive/Coordinated

Nonlinear Control

Complex and Large ScaleOptimization

Immunity

EmissionReduction

Self-Healing

BehavioralModeling

Fast and AccurateDecisionMaking

Computational IntelligenceComputational IntelligenceTechniquesTechniques

Identification/Prediction

ofNonlinearDynamics

Robust/Optimal/Adaptive/Coordinated

Nonlinear Control

Complex and Large ScaleOptimization

Immunity

EmissionReduction

Fig. 3. Capabilities of CI methods for smart grids.

As known widely, the nonlinear modeling of the time

varying dynamics of an electric power system is still a challenge using classical techniques but computational intelligence paradigms such as neural networks have been shown to successful model the nonlinear dynamics offline or online. Multilayer perceptrons, radial basis functions (RBFs), simultaneous recurrent neural networks (SRNs) and echo state networks have all been reported very effective models for the design of nonlinear, adaptive and/or optimal controllers such as for generator excitation systems, turbines and FACTS devices [6-8]. Most of the identification/modeling techniques are based on supervised learning techniques as illustrated in Fig. 4. Neural networks can be trained online to track the


3

changing dynamics which is a beauty and a power of a CI approach for system identification.

Plant

Neural NetworkModel

OptimizationAlgorithm

Time delaylineTime delayline

∑

Plant Inputs Plant outputs

Parameters adjustment

neural networkoutputs

+

−

errors

Plant

Neural NetworkModel

OptimizationAlgorithm

Time delayline

Time delaylineTime delayline

∑

Plant Inputs Plant outputs

Parameters adjustment

neural networkoutputs

+

−

errors

Fig. 4. Identification of power system dynamics of interest.

The increasing complexity and highly nonlinear nature of electric power systems today requires a fast and an accurate online monitoring system, a wide area monitor (WAM), for effective control of power networks with an adaptive wide area controller (WAC). Similar approaches as illustrated in Fig. 4 are suitable for WAM [9, 10]. WAM is essential for smart grids to the functions mentioned in Section II such as self-healing, fault-tolerance and dynamic optimization.

As several generation sources connect and disconnect to a smart grid, it will become necessary to identify the dynamics of the system at any given point in time and adjust the parameters, and maybe even the structures, of the local and auxiliary controllers. Preliminary results of an external damping controller on a traditional power system have been demonstrated in a real-time on a laboratory platform [11, 12].

Furthermore, it has been shown that critic networks based on advanced neural network architectures, such as simultaneous recurrent neural network, are capable of optimizing the performance multiple local controllers on generators and FACT devices to improve system damping [13]. SRNs are powerful neural networks for prediction based on recurrence and have been shown to overcome the varying time delays in the communication channels for WAC. The real-time implementation of a two-area four machine power system shown in Fig. 5 was carried out at the author’s lab - Real-Time Power and Intelligent Systems Laboratory at Missouri University of Science and Technology. Despite varying time delays, a system damping higher than that provided by local controllers (see Fig. 6) can be guaranteed with WAC over a significant range of delays [14]. This is possible due to multi-step prediction capabilities of SRNs.

Wide-area coordinating control is becoming an important issue and a challenging problem especially when large wind farms are connected to the power grid. An optimal wide-area monitor based on a RBF neural network has been developed in [15] to identify the input-output dynamics of the nonlinear power system. Its parameters are optimized through particle swarm optimization. Based on the WAM, a WAC is then designed by using the dual heuristic programming (DHP) method and RBFs, while considering the effect of signal

transmission delays. The WAC operates at a global level to coordinate the actions of local power system controllers including those on the wind farm. Each local controller communicates with the WAC, receives remote control signals from the WAC to enhance its dynamic performance, and therefore helps improve system-wide dynamic and transient performance.

Damping System Oscillation

Wide Area Controller

Wide Area Monitor

⎟⎠⎞⎜

⎝⎛ Δ+− T10

1τt

^3ωΔ

Load in Area 1 Load in Area 2

1 5 67

89

2 4

10 11 3

Δω3(t)

u1(t)

u3(t-τ2)

G1 G3

G4G2

Δω1(t)

AREA 1 AREA 2

2se τ−

( )tu3

( )1τ−t3Δω1sτe−

( )T10t^1ωΔ Δ+

Damping System Oscillation

Wide Area Controller

Wide Area Monitor

⎟⎠⎞⎜

⎝⎛ Δ+− T10

1τt

^3ωΔ

Load in Area 1 Load in Area 2

1 5 67

89

2 4

10 11 3

Δω3(t)

u1(t)

u3(t-τ2)

G1 G3

G4G2

Δω1(t)

AREA 1 AREA 2

2se τ−

( )tu3

( )1τ−t3Δω1sτe−

( )T10t^1ωΔ Δ+

Fig. 5. Two-area four-machine power system with communication delays in WAC [13].

1 2 3 4 5 6 7 8 9

377.5

378

378.5

379

379.5

380S

peed

of G

1 in

rad

/sPSS

PSS and WACS for 0.02s delay



Time in seconds

1 2 3 4 5 6 7 8 9

377.5

378

378.5

379

379.5

380S

peed

of G

1 in

rad

/sPSS




Time in seconds Fig. 6. Speed of generator G1 during a disturbance with different communication delays.

Control of electric power systems relies on the availability

and quality of sensor measurements. However, measurements are inevitably subjected to faults caused by broken or bad connections, bad communication, sensor failure, or malfunction of some hardware or software. These faults in turn may cause the failure of power system controllers and consequently lead to severe contingencies in the power system. To avoid such contingencies, a sensor evaluation and (missing sensor) restoration scheme (SERS) by using auto-associative neural networks (auto-encoders) and particle swarm optimization. Based on the SERS, a missing-sensor-fault-tolerant control (MSFTC) was developed for controlling a static synchronous series compensator (SSSC) connected to a power network. This MSFTC improved the reliability, maintainability and survivability of the SSSC and the power network. The details of this work are presented in [16]. Such fault-tolerant technologies will be needed in a smart grid to improve its reliability and security.

The future grid will have the capability where plug-in vehicles can charge from it or discharge to it. An intelligent method for scheduling optimal usage of energy stored in these vehicles is necessary. The batteries on these vehicles can either provide power to the grid when parked, known as


4

vehicle- to-grid (V2G) concept or take power from the grid to charge the batteries on the vehicles, known as grid-to- vehicle (G2V). Finding optimal times to charge or discharge in order to maximize profits to vehicle owners while satisfying system and vehicle owners’ constraints was studied. In order to do this, binary particle swarm optimization was applied in [17]. Price curves from the California ISO database are used in this study to have realistic price fluctuations. Different fleets of vehicles are used to approximate varying customer base and demonstrate the scalability of parking lots for V2G. The results for consistency and scalability are included. Fig. 7 shows a typical parking lot for plug-in electric vehicles connected to the utility grid. Table I shows the profit made by the vehicle owners based on a PSO scheduling on a given day. More details on this work are given in [17].

TABLE I

RESULTS WITH THE THREE PARKING LOTS SIZED DIFFERENTLY AND THE PROFITS MADE ON A GIVEN DAY

# of Vehicles

Case Study

Power into Lot (MWh)

Power out of Lot

(MWh)

Net Power Out (MW)

Profit

50 CS1 0.0089 0.1131 0.1042 $11.41 CS2 0.3492 0.3421 -0.0072 $19.09

500 CS1 0.0984 1.2533 1.1549 $128.42 CS2 3.5167 3.8271 0.3104 $234.22

5000 CS1 1.0359 12.1769 11.1401 $1223.49 CS2 31.9632 35.2408 3.2777 $2200.40

Fig. 7. A typical parking lot with plug-in electric vehicles [15].

V2G can reduce dependencies on small expensive units in the existing and future power systems as energy storage that can decrease running costs. It can contribute to efficient management of load fluctuation, peak load; however, it increase spinning reserves and reliability. As number of gridable vehicles in V2G is much higher than small units of existing systems, unit commitment (UC) with V2G is more complex than basic UC for only thermal units. Particle swarm optimization was proposed in [18] to solve the V2G scheduling problem, as PSO can reliably and accurately solve complex constrained optimization problems easily and quickly without any dimension limitation and physical computer memory limit.

Climate dioxide (CO2) is now widely accepted as a real condition that has potentially serious consequences for human society and industries need to factor this into their strategic plans. The power and energy industry represents a major portion of global emission, which is responsible for 40% of the global CO2 production followed by the transportation sector (24%) [19]. Economic growth does not have to be linked to an increase of GHG emissions and can be attained in addition to the usage of renewable energy sources by using energy efficiency technologies for power system generation, transmission, and distribution. The development of intelligent energy-efficient control technologies will both soften negative effects of the climate change on the economy and enhance energy security. A summary of the impacts on CO2 reduction by the applications of intelligent techniques has been reported in [20]. Furthermore, UC with V2G can take into account besides fuel cost minimization, the reduction of emissions by explicitly factoring cost of emissions in the cost function. Table II shows some typical results on a small 10-unit and 20-unit systems with 50,000 and 100,000 gridable vehicles. Intelligent optimization become very useful as the parameters to optimize grows enormously.

Maintenance scheduling of generating units and other elements in a power system is an important task, and plays important role in the operation and planning activities of the electric power utility and the future smart grid. Computational intelligence tools are excellent for determining optimal maintenance schedules [21]. In a smart, resources need to be scheduled including real and reactive power generation, and energy storage.

Other computational intelligence applications for smart grid include forecasting dynamic loads and electric energy. Typically, amount of load is dependent on a number of factors including time of the day, day type (weekday or weekend), temperature, humidity, season and location. Neural networks have been extensively studied and shown to provide accurate load predictions [22, 23]. Neural networks have also shown to be promising tools for wind and solar energy predictions [24, 25]. When alternative sources of energy are connected to the power grid, be it wind farms or plug-in vehicles, dynamic load and electric energy at a given time needs to be predicted in order to carry out an efficient and economical operation of a smart grid.

There are several other CI applications for the smart grid including energy and power flow management algorithms. Adaptive critic designs and PSO-based fuzzy logic have been demonstrated to carry out optimal energy dispatch in a grid independent photovoltaic system [26]. Numerous researchers have reported on the successful use of CI approaches for voltage and reactive power control [27]. Reconfiguration of distribution systems have carried out with PSO and genetic algorithms. CI techniques have the potential to carry out dynamic reconfiguration and restoration of terrestrial and industrial power systems [28] after small and large disturbances. All these problems and solutions are applicable to smart grids.


5

TABLE II RESULTS WITH THE PSO TECHNIQUE FOR UNIT COMMITMENT WITH V2G

V. CONCLUSIONS

The electric power grid is rapidly growing and demanding new technologies for efficient, reliable and secure operation and control as the demand for electricity increases. The complexity of a smart power grid is much more than that of the traditional power grid as time-varying sources of energy and new dynamic loads are integrated into it. Advanced intelligent techniques are required to handle the smart grid operation in an efficient and economical manner. The potentials of computational intelligence paradigms have been demonstrated for various challenges in the traditional power system. Such techniques are promising solutions to deliver the expectations of a smart grid.

REFERENCES [1] S. M. Amin, B. F. Wollenberg, “Toward a Smart Grid”, IEEE Power and

Energy Magazine, Vol. 3, No. 5, Sept.-Oct. 2005, pp. 34-41. [2] A. Engelbrecht, Computational Intelligence: An Introduction, John Wiley

& Sons, Ltd, England, 2007, ISBN 978-0-470-03561-0. [3] G. K. Venayagamoorthy, “A Successful Interdisciplinary Course on

Computational Intelligence”, IEEE Computational Intelligence Magazine – A special issue on Education, Vol. 4, No. 1, February 2009, pp. 14-23.

[4] G. K. Venayagamoorthy, R. G. Harley, D. C. Wunsch, "Comparison of Heuristic Dynamic Programming and Dual Heuristic Programming Adaptive Critics for Neurocontrol of a Turbogenerator", IEEE Transactions on Neural Networks, vol. 13, no. 3, May 2002, Page(s): 764 -773.

[5] P. J. Werbos, “Approximate Dynamic Programming for Real Time Control and Neural Modelling”, in White DA and Sofge DA (Eds.), Handbook of Intelligent Control, Van Nostrand Reinhold, New York, 1992, ISBN 0-442-30857-4, pp. 493 – 525.

[6] G. K. Venayagamoorthy, R. G. Harley, “A Continually Online Trained Neurocontroller for Excitation and Turbine Control of a Turbogenerator”, IEEE Transactions on Energy Conversion, vol. 16, no.3, September 2001, Page(s): 261-269.

[7] J. W. Park, G. K. Venayagamoorthy, R. G. Harley, “MLP/RBF Neural Networks Based On-Line Global Model Identification of Synchronous Generator”, IEEE Transactions on Industrial Electronics, vol. 52, no. 6, Dec. 2005, pp. 1685-1695.

[8] G. K. Venayagamoorthy, R. P. Kalyani, “Two Separate Continually Online Trained Neurocontrollers for a Unified Power Flow Controller”, IEEE Transactions on Industry Applications, vol. 41, no. 4, July 2005, pp. 906-916.

[9] X. Li, G. K. Venayagamoorthy, “A Neural Network Based Wide Area Monitor for a Power System”, IEEE PES General Meeting, San Francisco, CA, USA, June 12 – 16, 2005.

[10] G. K. Venayagamoorthy, “Online Design of an Echo State Network Based Wide Area Monitor for a Multi-machine Power System”, Neural Networks, vol. 20, issue 3, April 2007, pp. 404-413.

[11] G. K. Venayagamoorthy GK, S. R. Jetti, “A Dual Function Neuron Based External Controller for a Static Var Compensator”, IEEE Transactions on Power Delivery, Vol. 23, Issue 2, April 2008, pp. 997 – 1006.

[12] S. Ray, G. K. Venayagamoorthy, “Wide-Area Signal-Based Optimal Neurocontroller for a UPFC”, IEEE Transactions on Power delivery, Vol. 23, Issue 3, July 2008, pp. 1597 – 1605.

[13] G. K. Venayagamoorthy, “Dynamic Optimization of a Multimachine Power System with a FACTS device Using Identification and Control ObjectNets”, 39th IEEE IAS Annual Meeting on Industry Applications, Seattle, WA, USA, October 2004, pp. 2643-2650.

[14] S. Ray, G. K. Venayagamoorthy, “Real-Time Implementation of a Measurement based Adaptive Wide Area Control System Considering Communication Delays”, IET Proceedings on Generation, Transmission and Distribution, Vol. 2, Issue 1, Jan. 2008, pp. 62 - 70.

[15] W. Qiao, G. K. Venayagamoorthy, R. G. Harley, “Optimal Wide-Area Monitoring and Non-Linear Adaptive Coordinating Control of a Power System with Wind Farm Integration and Multiple FACTS Devices”, Neural Networks, Vol. 21, Issues 2-3, March/April 2008, pp. 466-475.

[16] W. Qiao, G. K. Venayagamoorthy, R. G. Harley RG, “Missing-Sensor-Fault-Tolerant Control for SSSC FACTS Device with Real-Time Implementation”, IEEE Transactions on Power Delivery, to appear.

[17] C. Hutson, G. K. Venayagamoorthy, K. Corzine, “Intelligent Scheduling of Hybrid and Electric Vehicle Storage Capacity in a Parking Lot for Profit Maximization in Grid Power Transactions”, IEEE Energy 2030, Atlanta, GA, USA, Nov. 17-18, 2008.

[18] A. Saber, G. K. Venayagamoorthy, “Optimization of Vehicle-to-Grid Scheduling in Constrained Parking Lots”, IEEE Power General Society General Meeting, Calgary, CA, July 26 - 30, 2009.

[19] S. Labatt and R. R. White, “Carbon Finance: The Financial Implications of Climate Change,” John Wiley & Son, Inc., Hoboken, New Jersey, 2007.

[20] G. K. Venayagamoorthy, G. Braband, “Carbon Reduction Potential with Intelligent Control of Power Systems”, 2008 IFAC World Congress, Seoul, South Korea, July 6-11, 2008.

[21] Yare Y, Venayagamoorthy GK, Aliyu UO, “Optimal Generator Maintenance Scheduling Using a Modified Discrete PSO”, IET Proceedings on Generation, Transmission and Distribution, Vol. 2, No. 6, Nov. 2008, pp. 834-846.

[22] Z. A. Bashir, M. E. El-Hawary, “Applying Wavelets to Short-Term Load Forecasting Using PSO-Based Neural Networks”, IEEE Transactions on Power Systems, Vol. 24, No. 1, Feb. 2009, pp. 20-27.

[23] D. C. Park, M. A. El-Sharkawi, R. J. Marks, L. E. Atlas, M. J. Damborg, “Electric Load Forecasting Using an Artificial Neural Network”, IEEE Transactions on Power Systems, Vol. 6, No. 2, May 1991, pp. 442-449.

[24] T. G. Barbounis, J. B. Theocharis, M. C. Alexiadis, P. S. Dokopoulos, “Long-term Wind Speed and Power Forecasting Using Local Recurrent Neural Network Models”, IEEE Transactions on Energy Conversion, Vol. 21, No. 1, March 2006, pp. 273-284.

[25] R. Welch, S. Ruffing, G. K. Venayagamoorthy, “Comparison of Feedforward and Feedback Neural Network Architectures for Short Term Wind Speed Prediction”, IEEE-INNS International Joint Conference on Neural Networks, Atlanta, GA, USA, June 14-19, 2009.

[26] R. Welch, G. K. Venayagamoorthy, “A Fuzzy-PSO Based Controller for a Grid Independent Photovoltaic System”, IEEE Symposium on


6

Computational Intelligence and Data Mining, Honolulu, USA, April 1-5, 2007, pp. 227-233.

[27] L. Grant, G. K. Venayagamoorthy, G. Krost, G. Bakare, “Swarm Intelligence and Evolutionary Approaches for Reactive Power and Voltage Control”, IEEE Swarm Intelligence Symposium, St. Louis, MO, USA, September 21-23, 2008.

[28] P. Mitra, G. K. Venayagamoorthy, “Real-Time Implementation of an Intelligent Algorithm for Electric Ship Power Reconfiguration”, IEEE Electric Ship Technologies Symposium, Baltimore, Maryland, USA, April 20-22, 2009.

BIOGRAPHY

Ganesh Kumar Venayagamoorthy (S’91, M’97, SM’02) received the Ph.D. degree in electrical engineering from the University of KwaZulu Natal, Durban, South Africa, in 2002. Currently, he is an Associate Professor of Electrical and Computer Engineering, and the Director of the Real-Time Power and Intelligent Systems (RTPIS) Laboratory at Missouri University of Science and Technology (Missouri S&T). He was a Visiting Researcher with ABB

Corporate Research, Sweden, in 2007. His research interests are in the development and applications of advanced computational algorithms for real-world applications, including power systems stability and control, power electronics, alternative sources of energy, sensor networks, signal processing, and evolvable hardware. He has published 2 edited books, 5 book chapters, 65 refereed journals papers, and over 230 refereed international conference proceeding papers. He has attracted in excess US $6.75 Million in competitive research funding from external funding agencies in the last six years.

Dr. Venayagamoorthy is a recipient of several awards including a 2007 US Office of Naval Research Young Investigator Program Award, a 2004 NSF CAREER Award, the 2008 IEEE St. Louis Section Outstanding Educator Award, the 2006 IEEE Power Engineering Society Walter Fee Outstanding Young Engineer Award, the 2005 IEEE Industry Applications Society (IAS) Outstanding Young Member Award, the 2003 International Neural Network Society Young Investigator Award, and a 2008, 2007 and 2005 Missouri S&T Faculty Excellence Award.

Dr. Venayagamoorthy has involved in the leadership and organization of many conferences including the General Chair of the 2008 IEEE Swarm Intelligence Symposium (St. Louis, USA) and the Program Chair of 2009 International Joint Conference on Neural Networks (Atlanta, USA). He is currently the Chair of the IEEE Power Engineering Society (PES) Working Group on Intelligent Control Systems, the Chair of IEEE Computational Intelligence Society (CIS) Task Force on Power Systems Applications, the Vice-Chair of the IEEE PES Intelligent Systems Subcommittee, Vice-Chair of the IEEE PES Students Subcommittee and the Chapter Chairs of IEEE CIS and IEEE Industry Applications Society St. Louis Chapters.

He is a Fellow of the Institute of Engineering and Technology (IET), UK and the South African Institute of Electrical Engineers.


/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 150 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.00333 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages false /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 600 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.00167 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org) /PDFXTrapped /False

/CreateJDFFile false /Description >>> setdistillerparams> setpagedevice

Documents

Potentials and Promises of Computational Intelligence for ...brain2grid.com/wp-content/uploads/2008/12/... · CI methods can contribute to electric power systems and smart grids in