Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Do you really want to delete this prezi?
Neither you, nor the coeditors you shared it with will be able to recover it again.
Make your likes visible on Facebook?
You can change this under Settings & Account at any time.
Online Partial Discharge monitoring System for oil-filled Power transformer By UHF Sensors
Transcript of Online Partial Discharge monitoring System for oil-filled Power transformer By UHF Sensors
System for oil-filled Power transformer By UHF Sensors INTRODUCTION Power Transformer is on of the most important device in the power system components By :
Sulaiman Al Zakari
Mohammad Al Motairy the durability of the power system depends mainly on the transformer running continuity A scheduled maintenance and checking on the transformer condition is required At First It is a static electrical device that transfer the same amount of power between two electrical circuits by the means of coupling two conducting windings inductively. Transformers can be classified in many ways: Power capacity
Duty of a transformer
Purpose: distribution, rectifier
Three phase winding configuration: autotransformer, delta, wye...etc. In this project We will deal with a step up / down high voltage oil insulated power transformer its power rating lays in MVA region, and it is used in power plants as step up for the transmission lines or
in distribution stations as step down for the loads Secondly What is Power Transformer ? Insulation Types 2 Types of Insulations in power transformer Gas insulation (SF6) : Liquid insulation
(Oil-Impregnated Paper) Impregnated paper Sulfur hexafluoride (SF6) is the most common gas used as insulation in many power system equipments it is compressed to be in the form of a liquid It has many advantages its inflammable, safer in case of internal arc, less noise and lighter insulate the conductor turns (windings) from each other, and to insulate them from the core Mineral oil It is the insulating fluid that has the greatest use in electrical equipment It has many functions, such as: to Provide a dielectric medium (insulation).
to dessipate the heat.
used to diagnose the condition of the trasformer. This function, is main key of this project Thirdly Transformer Fautls 2 types of Faults Winding Faults It is the case where the insulation of the transformer winding breaks down. Which causes a
short circuit between the turns with each other (turn-to-turn fault),
between two phases (phase-to-phase fault) , phase to ground,
between windings and the core of the transformer. the protection from this fault is achieved by using the differential relays Insulation Faults It is the case where the insulation of the Transformer breaks down there are two method to protect the transformer from this fault By using Buchholz relay By the detection of the partial discharge It is a gas- actuated relay installed on the oil-filled power transformers When the fault is developing slowly, a gas accumulates till it reaches a predetermined value.
it closes the contact to send an alarm If serious fault occur, an enormous amount of gas is generated in the main tank. The oil in the main tank rushes towards the conservator via the buchholz relay,
the contact closes to send alarm or disconnect the transformer. this approach will be adopted in this project
We should take a look on the Partial Discharge in separate subject Partial Discharge in Oil Transformers What is Partial Discharge It is defined as a phenomenon that occurs in a wide variety of electric equipment and it develops slowly as time passes by Its known as an electrical discharge that partially bridges the insulation between conductors(electrodes)  when the time basses out,
a complete failure in insulation material occurs which causes a total breakdown. It may occur in
solid, liquid or gaseous insulation media. Oil insulated power transformers are one of the commonly used electric equipment that suffers from this electric phenomenon. Causes and Developing Mechanism of PD in oil the most important cause of the PD signal in Oil insulation is Voids (bubbles) Two types of voids are considered that leads to two different mechanisms of the initiation of PD in oil transformers First,
Is the momentarily created and disappeared suddenly voids
because of the electrical stress enhancement points such as metallic asperities and protrusions Second type
is the physically stable voids existing within the oil-impregnated paper insulation it causes a classical discharge behavior in that the +ve and -ve discharges during the applied sinusoidal voltage cycle. PD singal in oil transformer is effected by many factors: The aging of the oil.
The smoothness of the electrode
Excessive electrical stress when using the transformer above its rated voltage helps PD to develop.
Impurities accumulated in the oil
The increasing of the pressure of the oil, moisture and temperature, also stimulate the PD. Partial Discharge Detection Techniques Among the past decades, many researchers and experts designed and invented detection methods for PD,
these can be summarized as following Conventional method it is by measuring the induced apparent charge by a detection circuit. It is for the testing purposes, and experimental studies. Non-conventional method: A. Non-electrical: B. Electrical: Acoustic detection
Visual or optical detection
Chemical detection. Pulse Current Detection.
Radio Interference Voltage.
Dielectric Loss Analysis.
The Radio Frequency Analysis (RFA). Noise in Partial Discharge. Classification of Noises in PD Signal Sinusoidal noise,
such as communication carrier signal AM/FM The detected signal will have a noises superimposed on it, these noises must be filtered in order to analyze the PD signal and judge on the insulation condition, Pulse type (repetitive or random) noise
This type of noise possibly comes from power electronics, switching operations or
Radio Frequency (RF) emissions from power equipment. White noise,
It is referred to some random signal with flat spectrum density Online Monitoring system Preface & Definition The measurement of partial discharge in the insulation will give an instantaneous and time-resolved assessment of the insulation condition. PD monitoring is thus one of the most important tools available that allows the engineer to assess the condition of transformer insulation. What is PD monitoring system It describes the measurement and recording of partial discharges over a longer period It means that the PD measurement will be carried out during the normal operation of the transformer allows long time trending of partial discharge events, rather than only at a flash point during a periodic test. PD monitoring in power system components generally The application of PD monitoring on different power system components differs in terms of other regarding sensor coupling, detecting method, and so forth In general, the PD monitoring application for any power system can be explained simply as: PD occur
Optical Feature Extraction
Analysis of PD Pattern Classification
Source & Location System Components Hardware: To make it easier,
the components of the system is classified into
Hardware & Software UHF Sensor It is classified under the antenna type coupler sensors.
As the name implies, it operates at the Ultra high frequency range (UHF) practical antenna design can differ depending on the application In our case for the power transformer, the sensor can be installed in many ways: 1. Inside the transformer.
this gives accurate readings,
but not Practical 2. Drain Valve Type (Probe) It is very useful with high sensitivity,
But not as practical as the next one 3. Window Type,
It is mounted externally on the transformer tank It is very useful due to the easiness of its installation. But has lower sensitivity
So, signal amplification is required In our System: A two-wire Archimedean planar spiral antenna is the inside disign of the sensor This broadband antenna can be designed with different band-width
300MHz to 3000 MHz Amplifier Due to the PD signal characteristic,
the signal is detected after a stage of attenuation.
Hence, an amplifier is needed to improve the signal strength. Communication mediums Coaxial cables or Fiber optic
LAN network Data Acquisition Card To sample the measured signals (PD signal in our case)
and converting the resulting samples into digital numeric values that can be manipulated by a computer Supervisory computers and the main server To handle the software part of this system Software The software part handles the two main thing in this Project: 1. Signal Processing,
Visualization of the detected signal
Data Acquisition (sampling and digitizing)
Filtering the noises 2. Data Processing
PD Pattern clissification
Fault Recognition the most important three processing parts that needs our attention are:
Visualization of the detected signal.
PD Pattern classification We will discuss them later To illustrate the operation principle of the system, lets follow this chart: Control Unit
or SCADA the Sensors detects the PD signal transfer it by Coaxial cable The amplifier enhance the signal strength DAQ device digitize the signal
and collect the data The supervisory computer
Filter the noises
Process the collected data.
and visualize the signal.
Performs a signal pattern classification
Provides a user interface
Share the information with the Central Server. Central Server Unit where:
diagnose the condition of the transformer
stores a history logs for the data
Recognize the fault and issues the alarm
provide a connection with the station general network or SCADA Operation Principle Signal Modeling &Processing The Visualization of the signal can be achieved by many methods The most common methods are: A. Phase-Resolved Partial Discharge (PRPD).
B. Time resolved method
C. 3-Phase Amplitude Relation Diagram (3 PARD): Filtering of the noise there are many approaches to filter the noises from the detected signal the selection between them depends on the system environment and the noises types the most common methods are: A. Gating and Windowing.
B. Pulse arrival time difference.
C. Digital filter method.
D. Signal processing method
A. Statistical method
B. PD pulse shape method PD Pattern Classification Why To understand and trace the PD pattern
To provide a data base for the PD signal characteristic and use it to recognize any changing that indicates the presence of the fault the most common methods are:
A. Pulse Sequence Analysis (PSA):
B. Distance classifier (k-NN
C. Neural Network (NN):
D. Support Vector Machine (SVM) Each part of the signal modeling and processing methods and approaches mentioned above is an advance topic.
that requires a specialists in each matter. Conditions for Successful System The system must judge carefully so the power system doesn’t interrupted by false alarm.
The responsible of this is the PD pattern recognition section. 1. Reliability. 2. Immunity The sensor design must be shielded from external noises
and the detected signal must be filtered 3.Maintenance free The hardware parts of the system must be rugged and well designed.
All the connections must be well chcked 4. Weather-proof this should be considered during the designing of the system Conclusion Partial discharge signal in the insulation oil of the power transformer radiates Electro-Magnetic waves in the Ultra-High frequency range UHF Sensors can be used to detect these signals with proper calibration and suitable rang selection It is found that the suitable range for detecting PD signal in oil transforemr is
( 0.3 – 3 GHz). In this project, a complete system of online partial discharge monitoring of high voltage oil-filled power transformer is discussed. It is found that:
Ultra-High frequency detection technique appears to be the most reliable and suitable method, due to its accurate output and its immunity from different interferences in comparing with the other techniques Further filtering is necessary to get the proper signal This projects showed that the processing and the analyzing of the detected signal can be accomplished in many ways, and the PD signal can be recognized with different approaches The selection between these approaches depends mainly on the complexity of the power system scheme desgin that needs to be monitored EE 498 1. Internal type,
which is not practical , but has high sensitivity