Faults Diagnosis On A Power System Transmission Line Using Neural Network
CHAPTER ONE
Objectives of the study
The objective of this study is to implement a complete, accurate and efficient scheme for distance protection of power system transmission lines. In order to perform or achieve this goal, the task is subdivided into different neural networks for fault detection, classification and isolation (location) in different zones on the transmission lines.
Also to compare the results obtained with the method of ANN with those obtained using the method of symmetrical components.
CHAPTER TWO
LITERATURE REVIEW
The idea of transmission lines in power system
In the early days of commercial electric power, transmission of electric power at the same voltage as used by lightings and mechanical loads restricted the distance between generating plants and consumer.
In 1882, generation was with direct current, which could not easily be increased in voltage for a long-distance transmission. Different kinds of load (example, lighting, fixed motors, traction/railway system) required different voltages and so use different generators and circuits for their individual supply of power.
Due to this big problem and inefficiency of the system to transmit voltage from one place to another, generators were meant to be closer to their loads. Also this problem affected the industries that, they developed a power system called Distributed Generation, whereby small generators were located near their individual loads in the industries. [1]
In 1886, in Barrington, Massachusetts, a IKV, AC power at 2MVA transmitted for 30km distance was installed at Cerchi, Italy. Nikola Tesla on May 16, 1888 delivered a lecture titled A New System of Alternating Current Motors and Transformers.
Here he described the equipment which allowed efficient generation and use of polyphase alternating currents. The transformer, Tesla polyphase, and Single-phase induction motors were essential for a combined AC distributed system for both lighting and machinery.
The ownership of the right to the Tesla patents was a key advantage to the Westinghouse Company in offering a complete alternating current power system for both lighting and power for different communities.
The innovation of the Tesla Alternating current power system was regarded as one of the most influential electrical innovations in the universe. The Tesla’s Transformer is now used in stepping-up voltage from generators to high-voltage transmission lines and then stepping-down voltage to local distribution circuits or industrial customers.
Generating stations and load using different frequencies could be interconnected using rotary converters and common generating plants are used for every type of load to achieve important economics of selling power, with lower overall capital investment required. Load factor on each plant was increased allowing for higher efficiency, lower cost for the consumer and increase in over all use of electric power. By allowing multiple generating plants to be interconnected over a wide area (Transmission power in a short, medium and long distance areas), electricity production cost was reduced.
The first transmission of the three-phase alternating current using high voltage took place in 1889 during the international exhibition in Frankfurt Germany. Here a 25kV transmission line of approximately 175km long connected to Lauffen on the Neckar and Frankfurt.
Now voltages used for electric power transmission were increased throughout the 20th century.
By 1914, fifty-five transmission systems each were operating at more than 70kV was in service, but as at then, the highest voltage used was 150kV.
In the 20th century, rapid industrialization made the electrical transmission line system a critical part of the infrastructure especially in most industrialized nations. World war1 brought about great destructions of many large electrical generating plants which were built by governments to provide power to big factories that produced military weapons and later used to supply power to the whole cities by means of long distance transmission lines. [2].
CHAPTER THREE
METHODOLOGY
In this chapter, we are going to discuss the application of two methods, the symmetrical component and artificial neural network (ANN) methods.
There are two types of faults that occur in a power system, the balanced and unbalanced faults. We are going to discuss there application one after the other.
BALANCED FAULT
This is a severe type of fault that rarely occurs in the power system. Example of this type of fault is a three phase fault. In this work, balanced fault is diagnosed using Bus impedance matrix and ANN methods, while unbalanced fault is diagnosed using symmetrical component and ANN methods.
To diagnose the balanced three phase fault, equivalent single phase circuit is used.
Equivalent single phase circuit
The information relating to one single phase gives the information relating to the other two phases as well. Therefore, it is sufficient to do the calculations in a single phase circuit and relate them to other phases. The calculation is done in two ways.
One single phase of the three phase circuit is taking.
CHAPTER FOUR
FAULT DIAGNOSIS USING SYMMETRICAL COMPONENTS AND ARTIFICIAL NEURAL NETWORK (ANN)
MODELING OF THE POWER SYSTEM TRANSMISSION LINE
Here, a 330kV transmission line system is (medium line) used to develop and implement the proposed network architecture using the back-propagation algorithm.
In this work, we are using the New-Haven – Nkalagu – Abakiliki transmission line system as a case study. The Figure 4.1 describes the one-line diagram of the above name area of case study. The system parameters are used to train and test the neural network.
CHAPTER FIVE
Conclusion and further works
Conclusion
In this work, different methods which include the symmetrical component and neural network methods have been used in diagnosis of faults on power system transmission line.
The work showed that the ANN method can easily be applied in power system transmission line to obtain the faulty conditions which can be implemented in the protection scheme (distance protection scheme) for the protection of the transmission line system.
However, in the both methods, the neural network was found to be the best. This is because; it is the fastest, simple, self explanatory, and accurate in the determination and analysis of fault in the power system transmission lines.
Suggestion for future work
As a matter of fact, further studies on this work can be done in areas of the use of electronic and digital relay conditions of a protection scheme for fault diagnosis on a particular power system component. Also, the software program generated by the selected ANN can be integrated into the power system protection scheme for automatic control and protection of any power system component. In this case, neural network is used to analyze, monitor and diagnose faults on the hybrid nature power generation, transmission and distribution without frequent visit to the protection and control unit.
There are still other areas of the neural network that needs to be studied with the aim of discovering new information for the improvement and development of best protection scheme for the power system and its components as the world advances in technology.
CONTRIBUTIONS TO KNOWLEDGE
- To the best of my knowledge, this is the first time an intelligent system such as Artificial Neural Network is used to analyze a component of the Nigerian Power system, the transmission line (New Haven – Nkalagu – Abakaliki).
- This work serves as a generalized method and can also be applied in other power system components for fault analysis. for fault diagnosis on any transmission line
- It can be adapted to solve other problems in electrical engineering and other engineering disciplines at large.
REFERENCES
- Wikipedia,” Internet research on Power generation, transmission and distribution using distribution generation (DG).
- Tesla memorial society of New York. (www.Teslasociety.com)
- P.H.C.N. Benin District, “High voltage fault daily log book (dispatch section) 2000.
- Matlab, 2007 version.
- M. M. Tawfik, M. M. Marcus: “Power engineering a new approach for fault location on the transmission line” IEEE Power Engineering review, Nov. 1998. pp 58 – 60.
- A. J. Mazon, I. Zamora, J. F. Minambres, M. A. Zorrozua, J. J. Barandiran and K. Sagasta Bietia. “ A new approach to fault location in two terminal transmission lines using neural network”. IEEE Trans. on power delivery, Nov. 2000,pp 1 – 2.
- A. Y. Abdulaziz, A. M. Ibrahim, M. M. Mansour and H. E. Talaat. “ Modern approach for protection of series compensated transmission lines”. IEEE Trans. on power delivery, May, 2005, pp 1 – 2.
