Intelligent Systems in Power Systems and Energy Markets
Carlos Ramos1, Chen-Ching Liu2
2
1
Polytechnic of Porto, Portugal
University College Dublin, Ireland
csr@isep.ipp.pt; liu@ucd.ie
In a world increasingly conscientious about the environmental aspects, Power and Energy Systems are
experiencing huge transformations in many ways. Electric energy produced from power plants is
transmitted and distributed to end users through the power grid. Power industry performs the engineering
design, installation, operation and maintenance tasks to provide high quality and secure energy supply
taking into account the system’s abilities to withstand uncertain events such as weather related outages.
Sustainable development has also been a challenge for power systems. In recent years, there is a
significant increase in installation of Distributed Generations, mainly based on renewable resources such
as wind and solar. The integration of these new generation systems leads to systems that are much more
complex. Indeed the number of generation sources greatly increases as the grid embraces the large
number of smaller and distributed resources. Uncertainties of the wind or solar energy lead to technical
challenges such as forecasting, scheduling, operation, control and risk management.
Electricity markets and new renewable energy sources bring a higher level complexity through the
economic dimension. The power grid is a physical infrastructure with specific constraints. Although
reliability is a key requirement, complexity of the power grids and the potential cascading events
combining natural and human causes can lead to catastrophic failures. Several severe incidents, including
blackouts, occurred (e.g. the 14th August 2003 Blackout in USA; the 4th October 2006 quasi-blackout
affecting 9 European countries; and the 10th November 2009 Brazilian blackout affecting 70 million
people in 18 states, including major southern cities São Paulo and Rio de Janeiro). Power system experts
are fully aware that there is a risk of degradation with the increasing complexity of power systems.
Some of the tasks involved in power engineering and economics are highly complex. Practical power
system engineering and energy market problems require logic reasoning, heuristic search, perception,
and/or the ability to handle uncertainties [1-2]. Artificial Intelligence (AI) or intelligent systems (IS) are
potentially powerful tools to provide new solutions where other computer technologies cannot. This new
generation of technologies needs to combine the technical operation of the power grid with the trading
activities of electricity markets. Intelligent systems are natural decision support tools for planning,
operation, maintenance, market monitoring, and risk management.
Reference [3] provides an analysis of the applications of IS techniques to Power and Energy Systems. The
main problems types where IS are used are the following:
- Alarm Processing, Diagnosis and Restoration - Expert systems and other knowledge-based systems have
been used to address alarm processing, diagnosis and restoration support at the dispatch and control
center level but also at substation and generation plant level. When data and information concerning a
large number of events, covering a large set of foreseen cases, are available, an artificial neural network
can be designed and trained with good results. The power industry performs training of the Control
Center operators using software simulators, some experiences have been gained with Intelligent Tutoring
Systems;
- Forecasting - The level of load demand determines the amount of energy supply from power plants.
Neural Networks have been extensively used for forecasting tasks. They have also been used in price
forecasting for Electricity Markets, namely combined with Fuzzy Logic.
- Security Assessment - Security assessment evaluates a power system’s ability to face a set of
contingencies in static and dynamic situations. Pattern recognition methods and artificial neural networks
have been developed for security assessment.
- Planning and Scheduling – These techniques are necessary in electrical networks and generation
expansion planning and in several operation problems, e.g., unit commitment, optimal dispatch, hydro
thermal coordination, network reconfiguration, and maintenance scheduling. The techniques frequently
used to solve these problems are computational intelligence and bio-inspired techniques, inspired in
nature and animal behavior, including: artificial neural networks, fuzzy systems and genetic algorithms,
simulated annealing, tabu search, swarm intelligence, and ant colony.
- Energy Markets – While power system engineering deals with the technical aspects of the grid, Energy
Markets lead to an economic perspective of the power industry. Multi-Agent System, Machine Learning /
Data Mining, and Game Theory techniques are used in solving problems in Energy Markets.
While the connection between IS and Power Systems and Energy Market researchers is beneficial for
both IS and power system communities, power and energy applications are not very visible at AI or IS
events and journals. On the other hand, the Power and Energy scientific community develops various
applications of Computational Intelligence methods. Table I illustrates the AI techniques, type of
problems and application areas in Power and Energy fields for the papers submitted to this special issue, a
good collection of samples of the state-of-the-art work in the area.
Table I – Analysis of Submitted Papers
AI
Areas
Generic
Number of
Submissions.
IS Techniques
Type
Problem
of
Knowledge-Based
Systems
3
Expert Systems (2),
Intelligent
Tutoring
Systems
Design, Monitoring,
Training
Probabilistic
Reasoning
Incomplete
Information
Fuzzy Systems
1
Probabilistic
Reasoning
Incomplete
Information
Fuzzy Systems
Neuro-Fuzzy
Monitoring
Constraints
Basic Search
1
2
Constraint Satisfaction
Local Search, Tabu
Search
Control
Planning
Evolutionary
Algorithms
9
Genetic Algorithms (9)
Optimization
(6),
Test,
Assignment,
Planning
Swarm Intelligence
8
Particle Swarm
Optimization (8)
Optimization (4),
Planning(2),
Reconfiguration,
Dispatching,
Selection,
Classification,
Estimation,
Forecasting
Game Theory
2
Game Theory (2)
Classification (2)
Pattern Matching
Machine Learning,
Data Mining
1
15
Pattern Recognition
Neural Networks (6),
Reinforcement
Learning (4), k-means
(2), Decision Trees,
Support
Vector
Machines,
NeuroFuzzy
Monitoring
Classification (4),
Forecast (3),
Monitoring (2),
Estimation (2),
Clustering, Control,
Optimization,
Modelling
1
4
Forecasting
(3),
Diagnosis, Control
(2), Optmization
Power
and
Applications
Energy
Cogeneration
Power
Plants,
Transformers
Quality
Control,
Incident Analysis in Power Systems
Control Centers
Viscosity Control in Fossil Fuel
Power Plants
Electricity Market Pricing
Oscillations
in
Interconnected
systems; Incident Analysis in Power
Systems
Control
Centers;
Minimization of Voltage Sag/Swells
Metering Systems; Stability of Power
Systems
Smart Grid
Switch Allocation in Distribution
Networks,
Distributed
Network
Expansion Planning
General Power Systems, Distribution
Systems, Assignment of Computer
Resources to Real-Time Digital
Simulators, Market Power, ShortTerm Scheduling of Distributed
Energy Resources, Benefits of
Distributed Generation Owners,
Minimization of Voltage Sag/Swells
Metering
Systems,
Distributed
Generation Planning, Vulnerability
Analysis in Power Systems
Composite Load Model Parameters,
Short-Term
Load
Forecasting,
Demand Response
Short-Term
Scheduling
of
Distributed
Energy
Resources,
Dispatching Wind Farms Reactive
Power Sources, Distributed Network
Expansion
Planning,
Feeder
Reconfiguration
in
Distribution
Systems, Feature Selection in NonTechnical Losses Detection
Learning Behavior of Electricity
Markets
Players,
Generation
Reliability in Power Markets
Reduction of Power Consumption
Stability Assessment, Transmission
Expansion Planning, Hydrocarbon
Development, Viscosity Control in
Fossil Fuel Power Plants, Generation
Reliability in Power Markets, ShortTerm Load Forecasting, Control of
Wind Generators, Analysis of
Agent-based
Systems
4
Multi-Agent Systems
(2),
Agent-based
Simulation (2)
Modeling,
Forecasting,
Optimization,
Control, Negotiation
Overvoltages in Power System
Restoration,
Non-intrusive
load
Monitoring, Stability of Power
Systems, Learning Behavior of
Electricity
Markets
Players,
Electricity Market Pricing, Bids in
Electricity Markets, Profits in
Electricity Markets, Incident Analysis
in Power System Control Centers
Electricity
Markets,
Bids
in
Electricity Markets, Profits in
Electricity Markets, Smart Grid
As shown in Table 1, Neural Networks, Reinforcement Learning, Genetic Algorithms, Fuzzy Systems,
Particle Swarm Optimization, and Multi-Agent Systems are the common techniques for several problem
areas related to classification, forecasting, planning, optimization, and control. When we compare the 41
papers submitted for this special issue with papers of the topics of the special issues proposed by IEEE
Intelligent Systems in the last 2 years, several new fruitful AI applications to Power Systems and Energy
Markets are not represented in this special issue. Some examples are:
- Agents and Data Mining [4], concerning agents for information mining, particularly interesting for the
Smart Grid and Energy Markets;
- Context-Awareness and Smart Environments [5] or Ambient Intelligence [6], useful for management of
Control Centers and Energy Markets;
- Group Decision Making and Emotional Computing [7], important for tasks such as Power System
Restoration and Energy Pricing;
- Intelligent Monitoring of Complex Environments [8], essential for Cyber-Physical infrastructures of
Power Systems and Smart Grid; cyber security of the power grid is an excellent example [9];
- Intelligent Search and Retrieval of Market Information, based on Semantic Web [10], allowing a layer
of competitive intelligence in Energy Markets and Business [11].
Based on the above observations, we recommend that:
- Power Systems and Energy Markets researchers and organizations explore new AI areas of research,
since these new areas have the potential to create a knowledge-oriented business for the field and a
human-level intelligence in the developed systems [12];
- AI researchers start to focus on Power Systems and Energy Markets applications field, since as the
Energy resources increase the cost and Environment and Sustainability are crucial for the society,
research investments in these fields will increase. Applications to Power and Energy systems are highimpact areas involving complex problems.
In the introductory article of this special issue by the Guest Editors, we analyze the AI techniques, types
of problems, and application areas of all papers submitted to this special issue. We also identify an
opportunity for the cross-fertilization between Power Systems and Energy Markets researchers and new
developments of AI. The papers selected for this special issue will provide the state-of-the-art information
about research being conducted using AI in Power Systems and Energy Markets. The work presented by
Vale, Pinto, Praça, and Morais describes MASCEM, a multi-agent simulator for Electricity Markets,
using reinforcement learning for forecasting bids. The paper by Guan, and Kezunovic applies Fuzzy
Systems and Petri Nets to support power systems control center operators on the economic impact of
incidents. Ferreira, Ramos, Vale, and Santos use Data Mining techniques operated on the data of the
California state Electricity Market for achieving important conclusions about the Network Transmission
Expansion Planning. The work presented by Catterson, Davidson, and McArthur discusses practical
applications of AI technology, namely Intelligent Agents, to the emerging area of Smart Grid. The work
of Almeida and Kagan combines Genetic Algorithms and Fuzzy Systems for optimization of power
systems voltage control.
References
1. I. Dabbaghchi, R. D. Christie, G. W. Rosenwald, and C. C. Liu; AI Application Areas in Power Systems,IEEE
Intelligent Systems, vol. 12, n. 1, pp. 58-66; 1997.
2. C. C. Liu, D. Pierce, H. Song; Intelligent System Applications to Power Systems; IEEE Computer Applications
in Power, vol 10, n. 4; 1997.
3. Z. Vale; Intelligent Power Systems; Wiley Encyclopaedia of Computer Science and Engineering; John Wiley &
Sons, Inc.; 2008.
4. L. Cao, V. Gorodetsky, and P. Mitkas; Agents and Data Mining; IEEE Intelligent Systems; vol. 24, n.3, pp 1415; 2009.
5. H. R. Arabnia, W. C. Fang, and C. Lee; Context-Aware Middleware and Intelligent Agents for Smart
Environments; IEEE Intelligent Systems, vol 25, n. 2, pp. 10-11; 2010.
6. C. Ramos, J. C. Augusto, and D. Shapiro; Ambient Intelligence: the Next Step for AI; IEEE Intelligent
Systems, vol 23, n. 2, pp. 15-18, 2008.
7. G. Marreiros, R. Santos, C. Ramos, and J. Neves; Context-Aware Emotion-Based Model for Group Decision
Making; IEEE Intelligent Systems, vol 25, n. 2, pp. 31-39; 2010.
8. C. Micheloni, P. Remagnino, H. L. Eng, and J. Geng; Intelligent Monitoring of Complex Environments, IEEE
Intelligent Systems, vol. 25, n.3, pp 12-14, 2010.
9 C. W. Ten, C. C. Liu, and M. Govindarasu; Vulnerability Assessment of Cybersecurity for SCADA
Systems,IEEE Transactions on Power Systems, vol. 23, n. 4, , pp. 1836-1846; 2008.
10. V. Temma; Semantic Web Support for Intelligent Search and Retrieval of Business Knowledge; IEEE
Intelligent Systems, vol. 25, n.1, pp 84-88; 2010.
11. H. Chen; Business and Market Intelligence 2.0, Part 2; IEEE Intelligent Systems, vol. 25, n.2, pp 74-82; 2010.
12. J. Beal and P. Winston; The new frontier of Human-level Artificial Intelligence; IEEE Intelligent Systems, vol.
24, n.4, pp 84-88; 2009.