Transformers are a large part of the component structure of the electricity system. Knowing the condition is essential to meeting the goals of maximizing return on investment and lowering total cost associated with transformer operation. In order to reconcile both decreasing maintenance spending and reliable service, condition based maintenance (CBM) is often proposed. The successful application of CBM lies in obtaining information transformers, so that, on the one hand, a critical condition will be noted early enough to take measures and on the other hand, so that only minimal maintenance is being applied to transformers still in good condition.
The paper covers the following areas.
· General aspects of aging in transformers.
· An overview of condition monitoring methodology with partial discharge patterns in focus.
· Partial discharge measurement and diagnostics on power transformer using a multi channel digital PD detector.
The need for improved performance from expensive industrial plant has in recent years necessitated the application of condition-Monitoring methodologies. These can provide early warning of potential failure with the opportunity of organizing avoidance strategies to minimize lost time and unexpected costs, thus greatly improving manufacturing efficiency.
The same is the case with power industry. Here maintenance expense as well as increased liability for non-delivered energy increases the costs of sudden failure of a component. E.g. Transformers.
The continuing drive for improved efficiency demands that a more educated analysis of monitored signals provides an indication of or even diagnoses the cause of a fault. The level of urgency of a condition can then be determined, thus enabling the necessary action to be taken over an appropriate fine scale. Maintenance can be scheduled based in plant condition rather than on a time based regime.
This need for condition based maintenance has encouraged the development of adaptable and cost effective diagnostics for HV transformers in addition to regular measurements like gas in oil analysis, several measuring tools have been introduced.
To simplify the task of PD testing, the digital PD detector LDS-6 was adopted to automate there measurements and perform sensitive measurements under electrically noisy conditions.
Condition Monitoring of Transformer
Condition based maintenance (CBM) for
Ø Cost reduction
Ø Life time extension
Ø Increased reliability
GENERAL ASPECTS OF AGING
The insulation system of a transformer consists mostly of oil and paper which are subject to aging. Aging is defined as the irreversible changes of the properties of electrical insulation systems (EIS) due to action of one or more factors of influence.
The aging factors produce electrical, thermal mechanical or environmental aging mechanism that eventually leads to failure.
When aging is dominated by one aging factor this is referred to as single factor aging. In multifactor aging more than one factor substantially affects the performance of the EIS. Aging factor may act synergistically, that is, there may be direct interactions between the stresses. Interactions may be positive or negative.
The aging of a practical EIS can be complex and failure is usually caused by a combination of aging mechanisms, even though there may be only one dominant aging factor.
The following aging factors are important
1. Thermal Aging: Thermal Aging involves the process of chemical and physical changes as a consequence of chemical degradation reactions, polymerization depolymerisation diffusions etc. It also involves the thermo mechanical effects caused by the forces due to thermal expansion or contraction. The rate of thermal aging is very much influenced by the operating temperature.
2. Electrical Aging: Electrical Aging either AC, DC or impulse involve the following.
· The effects of partial discharges
· The effects of tracking
· The effects of treeing
· The effects of electrolysis
· The effects of increased temperature produced by high dielectric losses.
Electrical Aging is influenced by field strength.
3. Mechanical Aging: Mechanical Aging involves the following:
· Fatigue failure of insulation components caused by a large number of low-level stress cycles.
· Thermo mechanical effects caused by thermal expansion and/or contraction.
· Rupture of insulation by high levels of mechanical stress such as may be caused by external forces or operation condition of the equipment.
· Abrasion wear caused by relative motion between equipment components.
· Insulation creep or flow under electrical, thermal or mechanical stresses.
4. Environmental Aging:
Environmental factors influence, in various ways, the kind and degree of degradation caused by other stresses to which on EIS is exposed. Other important aspects are the redistribution of stresses by changes in the environment and the influences of dust and other contamination on electrical behavior.
AGING IN THE TRANSFORMER INSULATION SYSTEM
1. Winding insulation :
Most windings in large power transformers consist of paper wrapped insulation on the windings. The paper is prepared form wood pulp and contains 90% cellulose. The latter is a natural polymer of glucose and consists of around 1200 monomer units. Once in its paper form there is some destruction of the structure during manufacture, winding and processing. Paper on a new transformer would start with an average chain length in monomer units or degree of polymerization (DP) of around 1000. Over most of the aging range there in a linear relationship between time and the logarithm of the chain length. The aging rate is enhanced by increasing temperature, increasing moisture and presence of air. The strength of paper-in turn is critically dependent upon the DP. For DPs of between 1000 and 500 the strength is directly related to DP. At a value of DP=150 and below, the strength is inadequate to withstand any winding movement. The aging under air rather than vacuum introduces and additional acceleration factor of 2.5. The role of moisture is more complex. Water may enter the transformer from outside but it is also produced by the degradation process itself. Rate of degradation when the paper has 4% moisture in 20 times greater.
2. Oil insulation:
The essential requirement for the oil is to maintain dielectric performance in the oil gap and across solid surfaces, to age very slowly and to have adequate thermal and viscosity properties to achieve better performance. A low quality oil or one with poor aging rate, is often associated with low transformer life. The use of additives can allow a poorer oil to achieve adequate initial properties for better performance. Hence, if used, the additive content must be monitored and maintained since the loss of property value can be very rapid.
CONDITION MONITORING
Every dynamic system – electrical, hydraulic, mechanical or thermal – possesses a normal characteristics ‘signature’ when operating in the desired fashion. When the signature changes. Even in a very subtle fashion, it may herald the onset of a failure mode. The small difference, between normal and abnormal ‘signatures’ have often been hidden by (or even assumed to be) ‘noise’ in the system. Modern transducers and associated signal analysis techniques can now discriminate between truly random variations and significant trends which, with knowledge of the system parameters and normal characteristics, can be used to predict time to failure. Such an approach is called Condition Monitoring.
Condition monitoring is now being applied to an ever-growing range of industries; if process variable such as critical dimensions, speed, temperature, potential difference, electrical current etc, can be measured, then a process signature can be identified which can lead in turn to the application of condition monitoring techniques. Systems have been developed in which the measurement signal from all the instruments are gathered on regular basis and compared with previous measurements and their trends as well as being validated against models of the process. Deviations and trends detected in this way provide warning of incipient faults in either the measurement system themselves or in the associated process plant and so assist avoiding unscheduled shutdowns.
A qualitative assessment of the health of the equipment often referred to as its condition, and this is used to assess the expected stability.
MEASUREMENT TECHNIQUES
PARTIAL DISCHARGE
Partial discharge is an electrical phenomenon that occurs within a transformer whenever the voltage load in sufficient to produce ionization, and partly bridges the insulation between conductors. Although the magnitude of such discharges usually small, they cause, progressive deterioration and may lead ultimate failure. Therefore partial discharge phenomena occurring in transformer can indicate incipient breakdown in the insulating dielectric.
Each partial discharge occurring within the transformer produces an electrical pulse, mechanical pulse and TEM waves.
Several detection techniques exist for partial discharge. Conventional partial discharge detection, is suited for sensitive carboratory, but cannot easily be used in normal environment. Important information about PD process, PD patterns, and fault mechanisms can be obtained under laboratory condition. To apply all this experience under service conditions, the so called VHF PD detection has been developed.
Multiple sensors are placed on the tank of the transformer and by comparing propagation time of partial discharges for each of the sensors and estimated location for the PD source is generated.
In case of VHP PD detection (500-1000 MHz) an antenna is inserted in the transformer tank through an oil valve. The TEM waves generated by partial discharge are picked up by the antenna and broad band filter is used to pick or select a suitable frequency area between 500 and 1000 MHz.
Important conclusions are made regarding the insulation condition of power transformers based on a 1-h voltage induced test. The HV apparatus are not discharge free and in general a discharge level <500 pC is tolerated.
After the PD patterns of the test setup are localized and identified. PD patterns originating from the power transformer are recorded during the enhanced voltage test. In practice two types of patterns are measured
1) Regular PD patterns which are patterns that are characterization of power transformer in good shape.
2) Irregular PD patterns that represent intolerable PD sources that can relate to insulation defects after manufacturing or aging effects during service life.
When a large population of power transformer is available for testing, a PD database can be developed for classification of such defects in other transformers by their PD patterns. Further more, such a PD database also provides information about general trends in regular or irregular PD patterns as they occur during induced voltage tests. This information contribute to a clearer insight when assessing the insulation condition of power transformers and scheduling maintenance.
PARTIAL DISCHARGE MEASUREMENTS USING A DIGITAL PD DETECTOR
To simplify the task of PD testing in the lab and on-site, the digital PD detecter LDS-6 was adapted to automate these measurements and perform sensitive measurements, under electrically noisy conditions
This special detector has following advantages.
Ø Permanent scanning.
Ø Digital, highly advanced, multi-application Partial Discharge and Diagnostic System.
Ø Automatic recording.
Ø Special noise suppression and improved sensitivity.
SETUP
The Partial Discharge signals will be taken from the tap of the primary or between primary and secondary. The measuring cable connects to an automatic multiplexer, which selects one of up to eight input channels for the digital PD detector.
OVER VIEW OF THE LDS -6
Important components are
a) Input Multiplexer
The input multiplexer selects the actual channel to test.Up to eight inputs is possible. When testing the transformer, the LDS-6 automatically scans the inputs in user-programmable sequences. E.g. - The inputs one to three can be tested repeatedly for one minute each.
b) Input Attenuator
The input attenuator consist of a switched resistive Pi-network to adapt the measuring sensitivity to the input PD level.
c) Wide band processing unit
Wide band processing unit is the heart of the detector. It integrates the PD signals to create an output signal equivalent to the apparent charge of the PD event. The term wide band processing unit indicates that it uses a wider part of the electro magnetic spectrum (100 KHz to 400 KHz).
NOISE SUPPRESSION
PD tests are very sensitive to external electro magnetic signals. There signals can affect the measurement. They are not the desired PD signals but noise. Especially in unshielded test laboratories and on site the noise level can be so high that it prevents successful Pd measurements if the noise is not reduced.
a) Noise Filter
One kind of electromagnetic noise is created by the local AM radio station or other narrowband HF transmitter.
The noise filter LDF-5 can be used to filter out there narrowband frequencies.
b) Noise gating
Some devices (e.g.-Variable frequency drives power tool set) create short electromagnetic impulse. The spectrum of there pulses is similar to the PD pulse spectrum, so they cannot be filtered out.
They can be distinguished by an antenna at a place where only the noise pulses can be received. But not the PD Pulses.
Then the noise gating unit suppress the measurement for the time of occurrence of the there noise pulses. Because the noise and the PD signals are not related all PD measurements and analysis can still be done.
VOLTAGE MEASUREMENT
A characteristic of transformer testing is the use of induced test voltages with a frequency higher than the normal power voltage. The voltage measuring circuit of the LDS-6 is able to measure accurately and synchronize to a test frequency of up to 500 Hz.
ANALYSIS OF THE PD DATA AND TEST REPORT
While testing every PD event of the active channel will be written to the hard disk together with the actual voltage, phase angle, time and channel number. This allows a replay of the test and all state of the art analysis methods can be employed. The LDS-6 also contains diagram functions, which compares the finger print of the PD measurement with the characteristics of previously recorded pattern and gives the most likely reason for the PD. This help with the interpretation of the results.
After the measurement, the data will be written to an Excel spreadsheets, with adaptable templates, the report is fully customizable and can also be saved to database.
ADVANTAGES OF THE SYSTEM
· Life enhancements of transformers and there by providing reliable and quality electrical power
· Condition based maintenance rather than time based maintenance
· Cost effective solutions
· Completely modular and wireless solution
· Universally available and user friendly platform
FUTURE EXPANSION
· This system in future will be expanded for dissolved gas analysis which will be capable of analyzing individual fault gases.
· This product will be launched as a solution package with the power apparatus resulting in high potential in market.
CONCLUSION
The system is developed and implemented at customer sites. This product is launched as a solution package with the power apparatus resulting in high potential in the market. Moreover this would make a strong place in the growing market. The online data was useful to arrive at suitable decisions.
REFERENCES
- IEEE Transactions on Power Delivery.Vol.17.No:2,April-2009
- CONDITION MONITORING www. Brighton-ac-uk/engineer
- PARTIAL DISCHARGE MEASUREMENT http://sun1.rrzn-user.uni-hannover.de
- IEEE International Symposium on Electrical Insulation, Araheim, ca, 2.-5.April-2000, Paper 13-2.
- High Voltage Technology , LL Alston (Oxford University)
- 2008 International conference on Condition Monitoring and Diagnosis (Beijing, China April 21-24 2008).
No comments:
Post a Comment
leave your opinion