Seminar Report_Artificial Intelligence in Power Station

GOVERNMENT POLYTECHNIC COLLEGE THIRURANGADI

Seminar Report On

ARTIFICIAL INTELLIGENCE IN POWER STATION Submitted by PRAJISHA

DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING

Seminar Report 2011-12

Artificial Intelligence in power station

2011- 2012

Department of EC, Govt. Polytechnic College, Thirurangadi

Page |2

Seminar Report 2011-12

Artificial Intelligence in power station

Page |3

GOVERNMENT POLYTECHNIC COLLEGE THIRURANGADI

DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING

CERTIFICATE This is to certify that this seminar report entitled “ARTIFICIAL INTELLIGENCE IN POWER STATION” is an authentic report of the seminar presented by PRAJISHA in partial fulfillment of the requirements for the award of the diploma in Electronics and Communication Engineering under the Directorate of Technical Education, Government of Kerala during the year 2011-2012. Place:

Chelari

Register No: 89200412 Date:

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

Page |4

Staff in charge Head of Department Anil Kumar. M. G G

Anil Kumar. M.

Head of section in E&C section in E&C G.P.T.C Thirurangadi Thirurangadi

Head of G.P.T.C

INTERNAL EXAMINER EXTERNAL EXAMINER

ACKNOWLEDGEMENT I extend my sincere gratitude towards Mr. Mohamed Shareef.K.P (Principal, Govt. Polytechnic College, Thirurangadi) and Anil Kumar. M.G (Head of Department, EC) for giving us his invaluable knowledge and wonderful technical guidance I express my thanks to all staff members of the electronics and communication engineering department for their kind co-operation and guidance. I also thank all the other faculty members of EC department and my friends for their endless help and support.

- Prajisha

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

Page |5

ABSTRACT Recently, due to concerns about the liberalization of electricity supply, deregulation, and global impact on the environment, securing a reliable power supply has become an important social need worldwide. To ensure this need is fulfilled, detailed investigations and developments are in progress on power distribution systems and the monitoring of apparatus. These are on (1) “digital technology” based on the application of semiconductor high-speed elements, (2) intelligent substations applying IT (information technology), and (3) system configurations aimed at high-speed communication. Incorporated in these are demands for the future intelligent control of substations, protection, monitoring, and communication systems that have advantages in terms of high performance, functional distribution, information-sharing and integrated power distribution management. Today’s conventional apparatus also requires streamlining of functions, improvements in sensor technology, and standardized interfaces. By promoting these developments, the following savings for the whole system can be expected: (1) reduced costs in remote surveillance in the field of apparatus monitoring, operation, and maintenance, (2) reduced maintenance costs based on

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

Page |6

the integrated management of equipment, and (3) reduced costs due to space saving as a result of miniaturizing equipment.

CONTENTS 1.

INTRODUCTION…………………………………………………… ……………… 6

2.

SUBSTATION……………………………………………………… …………………. 7 2.1.

TRANSMISSION SUBSTATION…………………………………………… 8

2.2.

DISTRIBUTION SUBSTATION……………………………………………. 8

3.

INTELLIGENT SUBSTATION………………………………………………. 10 3.1.

CONCEPT

OF

INTELLIGENT

SUBSTATIONS…………………… 10

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

APPARATUS

3.2.

Page |7

MONITORING

SYSTEM……………………………….. 10 POWER

3.3.

SYSTEM

CONTROLS……………………………………………. 12 4.

DEVICES THAT CAN CONTRIBUTE TO

AN

INTELLIGENT

SUBSTATION………………………….… 14 SWITCHGEAR

4.1.

AND

TRANSFORMER……………………………..… 14 PROTECTION

4.2.

AND

CONTROL………………………………………... 15 5.

LATEST

PROTECTION

AND

CONTROL

SYSTEM……………..… 16 UNIFIED

5.1.

PROTECTION

AND

CONTROL

UNIT……………….. 17 5.2.

REMOTE CONTROL FUNCTIONS BY WEB CORRESPONDENCE………………………………………….. 19

5.3.

CONNECTION BETWEEN PROTECTION/ CONTROL EQUIPMENT AND APPARATUS……………………. 21

6.

TASK FORCE SCOPE AND GENERAL GUIDELINES…………… 24 6.1.

TASK

FORCE

REPORT

OUTLINE……………………………………... 26 6.2.

SUMMARY OF PANEL PRESENTATIONS……………………….. 29

7.

CONCLUSIONS…………………………………………………… ………………. 31

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12 8.

Artificial Intelligence in power station

Page |8

REFERENCE……………………………………………………… …………………. 32

1.

INTRODUCTION

The upgradation of our 500-kV trunk transmission system has almost been completed, and the electricity system has been considerably improved. Yet, cost reductions are required to cope with the entry of IPP (Independent Power Producer) and the introduction of power source distributors caused by the deregulation of electric utilities. To achieve this, each electricity supply company is decreasing expenditure by efficiently using equipment, improving operations, and effectively controlling plant-and-equipment investment. In addition, power systems will become more complex, requiring operation in an uncertain and less structured environment. Consequently, secure and economic operation of power systems requires improved and innovative methods of control. The power distribution system also requires reductions in initial investments, such as the unit price of apparatus and miniaturization, and reduced costs for the whole life cycle, including the operation/maintenance costs of the substation system. The construction of a new power distribution system has been considered to meet these requirements. It will adopt “digital technology” and “IT-related technology,” which has made rapid advances in recent years. This system aims at minimizing the total cost, not only reducing of the unit price but also the cost of installation, construction, operation, and maintenance. This article discusses the construction of intelligent substations in the power distribution system, as well as protection/control-unified equipment as examples of the new technology.

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

2.

Artificial Intelligence in power station

Page |9

SUBSTATION

A substation is a part of an electrical generation, transmission, and distribution system. Substations transform voltage from high to low, or the reverse, or perform any of several other important functions. Electric power may flow through several substations between generating plant and consumer, and its voltage may change in several steps. A substation that has a step-up transformer increases the voltage while decreasing the current, while a step-down transformer decreases the voltage while increasing the current for domestic and commercial distribution. The word substation comes from the days before the distribution system became a grid. The first substations were connected to only one power station, where the generators were housed, and were subsidiaries of that power station. Substations generally have switching, protection and control equipment, and transformers. In a large substation, circuit breakers are used to interrupt any short circuits or overload currents that may occur on the network. Smaller distribution stations may use recloser circuit breakers or fuses for protection of distribution circuits. Substations themselves do not usually have generators, although a power plant may have a substation nearby. Other devices such as capacitors and voltage regulators may also be located at a substation. Substations may be on the surface in fenced enclosures, underground, or located in special-purpose buildings. High-rise buildings may have several indoor substations. Indoor substations are usually found in urban areas to reduce the noise from the transformers, for reasons of appearance, or to protect switchgear from extreme climate or pollution conditions.

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 10

Where a substation has a metallic fence, it must be properly grounded (UK: earthed) to protect people from high voltages that may occur during a fault in the network. Earth faults at a substation can cause a ground potential rise. Currents flowing in the Earth's surface during a fault can cause metal objects to have a significantly different voltage than the ground under a person's feet; this touch potential presents a hazard of electrocution.

2.1. TRANSMISSION SUBSTATION A transmission substation connects two or more transmission lines. The simplest case is where all transmission lines have the same voltage. In such cases, the substation contains high-voltage switches that allow lines to be connected or isolated for fault clearance or maintenance. A transmission station may have transformers to convert between two transmission voltages, voltage control/power factor correction devices such as capacitors, reactors or static VAr compensators and equipment such as phase shifting transformers to control power flow between two adjacent power systems. Transmission substations can range from simple to complex. A small "switching station" may be little more than a bus plus some circuit breakers. The largest transmission substations can cover a large area (several acres/hectares) with multiple voltage levels, many circuit breakers and a large amount of protection and control equipment (voltage and current transformers, relays and SCADA systems). Modern substations may be implemented using International Standards such as IEC61850.

2.2. DISTRIBUTION SUBSTATION A distribution substation transfers power from the transmission system to the distribution system of an area. It is uneconomical to directly connect electricity consumers to the main transmission network, unless they use large amounts of power, so the distribution station reduces voltage to a value suitable for local distribution.

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 11

The input for a distribution substation is typically at least two transmission or subtransmission lines. Input voltage may be, for example, 115 kV, or whatever

is common in the area. The output is a number of feeders. Distribution voltages are typically medium voltage, between 2.4 and 33 kV depending on the size of the area served and the practices of the local utility. The feeders run along streets overhead (or underground, in some cases) and power the distribution transformers at or near the customer premises. In addition to transforming voltage, distribution substations also isolate faults in either the transmission or distribution systems. Distribution substations are typically the points of voltage regulation, although on long distribution circuits (of several miles/kilometers), voltage regulation equipment may also be installed along the line. The downtown areas of large cities feature complicated distribution substations, with high-voltage switching, and switching and backup systems on the low-voltage side. More typical distribution substations have a switch, one transformer, and minimal facilities on the low-voltage side.

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

3.

Artificial Intelligence in power station

P a g e | 12

INTELLIGENT SUBSTATION

3.1. CONCEPT OF INTELLIGENT SUBSTATIONS In conventional substations, substation apparatus, such as switchgear and transformer, control, protection and monitoring equipment is independent of every other device, and connection is based on the signals coming through the cable. On the other hand, an intelligent substation shares all information on apparatus, control, protection, measurement and apparatus monitoring equipment through one bus by applying both “digital technology” and “IT-related technology.” Moreover, high efficiency and miniaturization can be achieved because the local cubicle contains unified control/protection and measurement equipment that is one integrated system (see Fig. 1). Since an optical bus shares the information between the apparatus and equipment, the amount of cable is sharply reduced. Moreover, as international standards (IEC 61850 and 61375 etc.) are adopted and the system conforms to the telecommunications standard, equipment specifications can be standardized for different vendors.

3.2. APPARATUS MONITORING SYSTEM All the data from each monitoring and measuring device is transmitted and used for a higher-level monitoring system via an optical bus. The required data is accessed through the Intranet or the Internet at the maintenance site of an electricity supply company or a manufacturer and the apparatus can be monitored from a remote location. The construction, analysis and diagnosis of the database

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 13

including trend management and history management also become possible. As a result, signs of abnormalities can be checked out well in advance, and prompt action can be taken in times of emergency. Maintenance plans can also be drafted to ensure reliability, by inspecting revision description and parts management, efficient maintenance planning and reliability maintenance are also realized simultaneously.

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 14

Fig. 1—Intelligent Substation System Configuration (Image). The whole substation system is combined by optical LAN, and apparatus composition is simplified.

3.3. POWER SYSTEM CONTROLS Power system controls can be broadly classified into two categories: local and area (regional/system-wide). The boundary between these two categories is not precise as area controls are often implemented by optimally adjusting local control parameters and set points. Area controls main characteristic is the need to process information gathered at various points of the network and to model the behavior of large parts of the power system. This type of control is usually not limited to the automatic feedback type but often includes strategies based on empirical knowledge and human intervention. Local control, on the other hand, is typically implemented using conventional automatic control rules, such as, PID control, which are believed to offer adequate performance in most applications. Still, this is not to discount the usefulness of new intelligent methodologies, such as, fuzzy logic controllers, for local controls. For convenience, power system higher level controls are classified here as: •

Generation scheduling and automatic control: includes unit commitment, economic dispatch, and automatic generation control; in the past, well established control methods were used but this situation has been changing to deal with the new scenario created by the power industry restructuring;

•

Voltage control: is mostly of the local type but some systems have already gone to a higher coordinated secondary control to allow a more effective use of reactive power sources and increase stability margins;

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12 •

Artificial Intelligence in power station

P a g e | 15

Preventive security control: has the objective to detect insecure operating points and to suggest corrective actions; the grand challenges in this area are on-line Dynamic and Voltage Security Assessment (DSA and VSA);

•

Emergency control: manages the problem of controlling the system after a large disturbance; it is an event driven type of control and includes special protection schemes;

•

Restorative control: its main function is to re-energize the system after a major disturbance followed by a partial or total blackout. Intelligent system techniques may be of great help in the implementation of

area power system controls. Most of these applications require large quantities of system information, which can be provided by modern telecommunications and computing technology, but require new processing techniques able to extract salient information from these large sets of raw data. Importantly, such large data sets are never error free and often contain various types of uncertainty. Finally, control actions may be based on operating strategies specified in qualitative form, which need to be translated into quantitative decisions. An important aspect to be considered in the implementation of power systems controls is that, in the restructured power system environment, several of these activities will fall under the category of ancillary services. Therefore, besides the technical issues, economic and financial infrastructure should be taken into account in the design and implementation of control schemes. Information regarding the state-of-the-art in the application of intelligent systems to power system problems can be found on the bibliography listed in section VII.

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

4.

Artificial Intelligence in power station

P a g e | 16

DEVICES THAT CAN

CONTRIBUTE TO AN INTELLIGENT SUBSTATION 4.1. SWITCHGEAR AND TRANSFORMER The burden can be drastically decreased because the sensor signal from the PCT is digitized at the sensor output edge and the load on the PCT only reaches that of an A-D (analog-to-digital) converter. Rogowski coils are used as the current sensors and capacitive potential dividers are used as the voltage sensors. These sensors drastically reduce the size of the switchgear (see Fig. 2).

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 17

Fig. 2—Gas Combined Switchgear Miniaturization by Digital Correspondence Sensor. 550-kV GCS (gas combined switchgear) is shown as an example. GCB: gas circuit breaker CT: current transformer PT: potential transformer Present studies on miniaturizing conventional equipment have so far been aimed at standardizing series. Advanced miniaturization will be attempted by digitizing this system, corresponding to its need.

4.2. PROTECTION AND CONTROL Intelligent substations require protection and control equipment to be installed outdoors and this needs to be compact so that the local cubicle is able to contain them. Outdoor installation requires improvements in insulation against heat and airtightness besides parts reliability. Compact protection and control equipment will generate demand for unified fabrication of protection/control and high-density components. The current protection/control system that uses compact equipment is described below. Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

5.

Artificial Intelligence in power station

P a g e | 18

LATEST PROTECTION AND CONTROL SYSTEM

Trends in Protection and Control Systems : Due to the rapid progress in today’s information field, applying digital technology and adding IT function to the protection/control system are possible, to support

stable power supply, and improve maintenance. In

Japanese

protection/control systems, digitization has made advances since the last half of the 1980s. Digital technology has unique advantages, namely minimizing maintenance and improving reliability, and it has speeded up the conversion from individual analog-type to digital-type relays. Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 19

Now, however, digitization is not only required for independent singlefunction equipment, but for the “systematic operation and employment” of the whole substation. Such systems have greatly improved efficiency in employment and maintenance using IT. The key phrases to fulfill these needs are as follows: (1) Slimming of total system as a protection control equipment → Unification of equipment (2)

High efficiency of employment/maintenance support using IT

technology → Extended employment/maintenance by remote control (3) System directly linked to the equipment for protection/control → Distributed installation near the apparatus Thus, there has been a need for constructing a high efficiency system through system-wide miniaturization and integration of IT.

5.1. UNIFIED PROTECTION AND CONTROL UNIT The protection and control units of the substation are designed and allocated with respect to individual functions and uses. Units are made according to their respective protection and control object. A cable or an exclusive-use LAN transfers the information between the protection/control equipment (see Fig. 3).

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 20

Fig. 3—Example of Combination of Optical LAN Application and Protection/Control Equipment, and Problems of Large-scale System. The present substation system consists of exclusive use of LAN (local-area network) for every information unit.

In detail, the information from the protective unit is transmitted to DAU (data acquisition unit) by optical transmission, and is then forwarded to the control room terminal, which has superior control. Such method is generally and commonly used.

This system’s digital equipment, protection, control, and information object equipment have a common basis. Therefore, combining the control/ protection equipment of every circuit unit can slim the total system. This equipment is compact, and configures the protection and the control units in one cubicle. Thus, hardware is reduced and there are considerable savings in power consumption.

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 21

The functions of the operation unit, which is the central component of the equipment, can be improved and shared, reducing the number of sections. As the dimensions of the whole unit are reduced by 50%, both the protection and control units can be configured into a single unit. The characteristics of this single unit are discussed below: (1) Operation unit The protection and control units need to be separated in the operation unit, which is equivalent to the center of the unit. Therefore, the CPU (central processing unit) was separate and the use of a high-performance 32-bit RISC (reduced instruction set computer) processor enabled us to reduce the total number of boards to 70%. (2) Input Transducer The input transducer, providing input current and voltage to the system, was until now, individually mounted away from the operation unit. However adopting a toroidal coil reduced the space by half, but doubled the number of mountings. The input section was improved to the extent that it is only a card mounted in the operation unit. (3) Power unit The power unit supplies power to the operation unit. As the number of CPU boards applied to the operation unit has been reduced and the application circuit for the protection/control unit has been standardized, the power supply capacity is halved as is the mounting space. Consequently, the power unit has become so compact that it can be mounted in the operation unit. (4) Interface By mounting the Ethernet* LAN port in the operation unit, it can now respond to a flexible network configuration. Ethernet LAN is based on the TCP (transmission control protocol)/IP (Internet protocol), which is a general-purpose standard network interface. This is a high performance all-in-one operation unit.

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 22

We slimmed down the system by mounting the protection and control equipment, which until now had been independent, into a single 350-mm width panel (see Fig. 4). The advantages of this equipment are as follows: (a) Perfect isolation between protection and control unit from input to output (b) Large reduction in installation space (Half the conventional space) (c) Direct coupling between protection and control unit by an isolated interface

5.2. REMOTE CONTROL FUNCTIONS BY WEB CORRESPONDENCE The amount of operation and maintenance needs to be reduced and detailed information in real time is required on the digital protection and control unit, during disturbances, or when the operations manager is notified of changes in the status of local equipment to ensure system stability. Also, there have been demands for remote operation, and manned-control-station operation to remote unmanned substations. A conventional digital panel saves and analyzes system information (the current/voltage data) when faults occur, and the CPU has highly automated observation functions. However, our system collects the voltage and current data that is saved inside the panel, in the remote maintenance section, and the results of automatic observation are analyzed and applied immediately. The system has an interface, which directly acquires the data via the network from the protection and control units in the substation. It is normally situated in the processing unit and the various kinds of information and operations supplied from the remote end, enable us to view progress in the network (see Fig. 5).

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 23

Fig. 4: Compact Type Operation Unit and Single Protection/ Control Equipment. The protection and control part are separated by independent structure, and CPU, input transducer and power supply unit are mounted in equipment. A protection unit is shown in the right of this figure. The interface characteristics are as follows: (1) The TCP/IP which is widely used as the standard network interface has been adopted, improving operability enabling easy access to exclusively used networks. By using an ordinary browser, most personal computers can access the network easily. (2) The server is in the panel, and individual and detailed information is disclosed to the operator as web-site information. Also, the information is

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 24

accessible by many clients at the same time via an exclusive-use network, and the data is the same even when faults occur. (3) By using an ordinary browser, connection using a general public circuit is possible without limiting the communication medium or use of the exclusive network. The cost of the communication and network equipment is reduced, and as the system is highly efficient, it further reduces costs.

5.3. CONNECTION BETWEEN PROTECTION/CONTROL EQUIPMENT AND APPARATUS In this system, the substation and the maintenance site are connected with the exclusive-use network in random time. The system can be constructed with shared and same-time data. Furthermore, the control and protection units are dispersed on the outside of the cubicle, with the units and the equipment communicating directly to one another. As a result, a large-scale substation system can be constructed at low a cost (see Fig. 6). The system has the following advantages because the protection and control units are placed near the equipment: (1) Reduced mounting space for the protection and control units

(2) Reduced construction costs by shortening the cable route from the equipment to the units, and the construction period (3) Higher reliability of information because the e information from the equipment is directly transmitted to the network.

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 25

Fig. 5— The Example of Construction of a Network and Simultaneous Employment of Data. By mounting general-purpose network interfaces as standard, it is possible to carry out operation and check the data simultaneously with the equipment of the other site by the ordinary browser. From now on, the system configuration increasing operation efficiency is also expected.

Increased reliability is expected, as a higher class network is multiplexed with one for waiting and the f other for regular use. Here, the regular diagnosis for t each unit is possible, by establishing an exclusive-use a server in both the maintenance site and the substation. Utilizing the system for future diagnoses is possible.

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 26

Fig. 6—Direct Combination of Apparatus, and an Example of Network Application. Intensive management of equipment and protection/control information is attained by arranging cubicle type protection/ control equipment near the apparatus, and carrying out network combination of this equipment.

6.

TASK FORCE SCOPE AND GENERAL GUIDELINES

The established scope for TF 38.02.20 is as follows : Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

•

Artificial Intelligence in power station

P a g e | 27

To review the current implementation of power system controls, including the role of human operators;

•

To review changes in the operation and control requirements of future power systems and identify the limitations of existing methods of control in meeting these requirements;

•

To review advances in intelligent system techniques and identify how they may be applied to meet the expanded control requirements of future power systems by complementing conventional controls and replacing some of the functions performed by human operators;

•

To develop guidelines for implementation of advanced controls using intelligent systems to assist in the secure and economic operation of power systems in the new electric utility environment.

In order to narrow the broad range of topics that may be included under the topics in the TF scope, some general guidelines were established with the intention of helping the TF members on focusing their work on aspects of the problem considered more relevant. These guidelines are: •

Emphasis will be on operator decision aids as opposed to feedback control methods or design. Operators are seen by many to have taken a great amount of new responsibilities in the deregulated environment without appropriate advances in decision aids and so there is a clear need of improvement in operator tools.

Further, there is less consensus surrounding the use of intelligent closed loop control in reliability critical systems, such as the power system, although intelligent controls are generally accepted as useful in consumer product applications.

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

•

Artificial Intelligence in power station

P a g e | 28

All intelligent system methods will be considered, which includes as a minimum symbolic processing (e.g., rule-based systems, logic programming, model-based reasoning), computational approaches (e.g., fuzzy sets, artificial neural nets), evolutionary programming, and genetic algorithms. The lines of separation among the various categories of intelligent approaches have blurred in recent years. In addition, these methodologies are more easily linked in the types of complex problems to which they are applied and thus for completeness in the applications, all methods should be considered.

•

Consideration should be given to fundamental understanding of operator needs and operational requirements, including interpretation of output and presentation of analysis. There has been some concern that operator needs have not been given the proper emphasis in prototype control center applications, which has limited the usefulness of the developed tools. There may be a need to initiate a survey of operator needs and experience.

•

Intelligent systems as an assistant for training operators should be included in the report. This may also be the natural link for including operator experience in developed systems. Training is seen by many as one of the most useful and important applications of intelligent system techniques.

•

Management of uncertainties should be carefully considered, including probabilistic techniques and risk analysis. There appears to be greater need within the control center for understanding of risk and uncertainties that will arise in a competitive environment. This is particularly true as it relates to financial risks. There appears to be some role for intelligent systems in assisting the operators in the management of risk.

•

The scope of the report should be limited to operations with minor consideration of operational planning but specifically does not include

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 29

planning. In the interest of time and focus, the report will serve a better purpose by avoiding the broader issues that develop under planning.

6.1. TASK FORCE REPORT OUTLINE The authors believe that an efficient way for the TF to perform its duties is to agree upon a report outline and to assign writing responsibilities to chapter lead editors and contributors. Later, the work performed by the members responsible for each chapter will be reviewed and re-oriented, if necessary, at the TF meetings. Most of the cooperative writing work is expected to be performed by e-mail. In the following, a first draft of the report outline is presented to be used as a starting point for the definition of its final version.

BACKGROUND Scope of applications : • Limiting to power system operations • Power system planning outside scope

Scope considered for intelligent system methods : • Rule-based methods/Logic programming • Model-based reasoning • Artificial neural nets • Evolutionary programming •

Approximate reasoning (fuzzy sets, certainty factors, etc.)

Sumary of report conclusions

POWER SYSTEM OPERATION Control functions and time frames : • Open and closed loop • Time frames of interest

Decision requirements in operations : Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 30

• Operator responsibilities • Centralized vs. decentralized

Limitations of existing tools : • Response to changing and unforeseen conditions • Usefulness of infrequently used applications •

Complexity of system prevents possibility of all situations being fully analyzed

•

Inability to improve with experience or easily incorporate experience

Trends in operations : • State of the art • Recent computational advancements •

Developments in static and dynamic security methodologies

• Survey (or discussion) of operator needs • Movement towards real-time controls

Challenges arising from the new utility environment : • New operator responsibilities • Increasingly stressed systems •

Possibility of broader fluctuations in system operating conditions arising from power sales and contracts

• Data limitations; concerns of proprietary data •

Regulatory requirements/restrictions on centralized decisions

• Variations in regulations from region to region

APPLICATION OF INTELLIGENT CONTROL METHODS FOR OPERATIONS Review of previous CIGRÉ and other historical efforts :

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 31

• Preventive/security control: Static and Dynamic • Emergency controls • Restoration control • Operational planning • System design aids vs. decision aids

State of the art and applicability : • Review of methods • Rule-base/logic approaches • Artificial neural nets •

Approximate reasoning

• Evolutionary programming and genetic algorithms

Limitation of methods : • Limits of formal performance measures • Concerns with reliability of adaptive/learning methods • Computational concerns for faster controls • Implementation concerns

New application areas identified by report : •

Economic controls – interaction with trading and contracts

• Controls as ancilary services

IMPLEMENTATION Guidelines/requirements : • Hardware • Software • Control center integration

Operator training/user interface Evaluation/testing methodologies : • Logical verification methods Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 32

• Simulation approaches

Software maintenance and updates : • Regular improvements in knowledge base

6.2. SUMMARY OF PANEL PRESENTATIONS This panel presents five papers. The papers concern topics related to security assessment, emergency control and control center operations under deregulation. These provide a sampling of problems appropriate for application of Intelligent Systems techniques. The papers are summarized in the following:

On-line Dynamic Security Assessment : This paper describes experiences in developing an efficient and effective tool for online DSA and points out problems that can be solved with artificial intelligence. Traditionally, the primary difficulty for on-line dynamic security lay in the required computational speed. Recently, technological advances have greatly increased processor power and some former barriers have been removed. Still, there are areas tasks that cannot be managed effectively on computations alone but require engineering judgement, experience and analysis. These tasks may be best addressed by Artificial Intelligence programming techniques.

Using a Neural Network to Predict the Dynamic Frequency Response of a Power System to an UnderDepartment of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 33

Frequency Load Shedding Scenario : This paper proposes a method to quickly and accurately predict the dynamic response of a power system during an underfrequency load-shedding scenario. Emergency actions in a power system, due to loss of generation, typically, calls for underfrequency load shedding measures. Due to the slow and repetitious use of dynamic simulators, the need for a fast and accurate procedure is evident when calculating optimal loadshedding strategies. A neural network (NN) seems to be an ideal solution for a quick and accurate way to replace standard dynamic simulations. The steps taken to produce a viable NN and corresponding results will be discussed.

Intelligent

systems

applications

to

emergency

control : This paper discusses the needs for more effective emergency control systems and the possibilities of using intelligent systems to reach this goal. The paper provides also pointers to already published work on using intelligent systems in this context.

Control Centre Operations and Training under Deregulation A New Zealand Example : This paper presents an overview of the control centre operation environment after deregulation in New Zealand. New Zealand is one of leading countries that has deregulated the power industry. As a result, the transmission network operation has become more market driven. The control centre operation environment has subsequently changed from low cost based to market based optimization. The roles of the control centre personnel have changed. There is more focus on risk analysis, real time operational planning and real-time power system security. A brief review of the changes followed by some valid opinion of the control centre personnel are also discussed in this paper.

7.

CONCLUSIONS

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 34

We described the emerging new technology in the electricity supply system. With the progress in communication technology and expansion of IT-related technology, research and development have also been progressing based on the concept of an intelligent system, not only on units or equipment, but also the constitution of the system itself. It is entirely conceivable that the needs of future clients will become even more diversified in this field. We will have to speed up the development and release of products that have compatibility and are low in cost, in accordance with the demands of these future clients.

8.

REFERENCE

Department of EC, Govt. Polytechnic College, Thirurangadi

Seminar Report 2011-12

Artificial Intelligence in power station

P a g e | 35

1. F. Iwatani et al., “Protective Relaying System Technologies Contributed to Optimum Equipment Configuration in Electric Power Network Systems,” Protective Relaying System Study Group, Asian Conf. on Power System Protection (Oct. 2001) 2. www.ieee.org

Department of EC, Govt. Polytechnic College, Thirurangadi