By Brian Sparling
The challenges faced by electric utilities over the past years are unrelenting and can be summed up in one sentence: "Reduce operating costs, enhance the availability of the generating and transmission equipment, and improve the supply of power and service to the customer base." And this, in an environment where the available resources are inexorably decreasing and the pressure from the shareholders and the competition mount steadily.
The concerted actions of all involved must impact every line of the income statement and of the balance sheet as costs and requests for new capital must be greatly reduced.
Critical oil-filled electrical equipment, such as transformers, shunt reactors, current transformers and bushings, are critical elements of an electrical power system. Their reliable and continued performance is the key to profitable generation and transmission. Their costs of acquisition, replacement, transportation, installation and repairs are among the highest on the system. Their failures, and resulting unavailability, create losses of revenues. These failures, when catastrophic, will generate substantial costs in terms of peripheral equipment destruction, environmental damages and unplanned emergency utilization of human resources and alternative power sources.
The early detection of incipient faults in transformers, shunt reactors, current transformers and bushings will create economic benefits that will have a measurable impact on the results required to meet these formidable challenges.
THE OVERALL BENEFITS OF MONITORING AND MANAGING TRANSFORMERS
Use and Load Your Transformer for Maximum Economical Efficiency
Monitoring creates the opportunity to strategically plan and schedule outages and to manage equipment utilization and availability.
In a deregulated environment, fast response to sudden requirements for overloading and taking advantage of market opportunities will be key to the success of any supplier of electrical power.
Manage and Extend the Life of the Transformer With Efficient and Cost-Effective Maintenance
Operating costs will be reduced as the equipment will be repaired within a scheduled repair plan, often on-site and many times under warranty.
The equipment can be kept in service, sometimes under reduced load conditions, when the fault is evolving at a moderate and predictable rate. This condition will prevent a loss of revenues and provide time to plan orderly action to repair or replace the equipment. These actions will often result in repair on-site because the time factor will improve the availability of the necessary resources. Site repairs create substantial economic benefits for transformer owners and for manufacturers when they work under warranty.
Managing and extending the life of critical power equipment require reliable and continuous monitoring as the validity of any action taken, to affect the life of a transformer, needs to be tested periodically. Successful life management and extension of power transformers produce financial benefits that impact the balance sheet in freeing capital for other requirements.
Detect the First Signs of Failure Conditions and Monitor the Evolution of On-Going Failure Conditions
Once a fault has been detected and its evolution is monitored, the severity of that fault can be evaluated and a decision can be made on the course of action to take.
Damages to the equipment will be restricted when incipient faults are detected and timely action is taken. Early detection limits the amount of adjacent damages and confines the area requiring repair and maintenance.
Reduce and Possibly Eliminate Unscheduled Outages and Failures
The early detection of incipient faults in transformers will greatly reduce unplanned power outages and improve the reliability of the power and service supplied to customers.
Fault conditions often lead to catastrophic failures. Their early detection will limit these events and enhance the safety of substation personnel.
Monitoring a fast developing fault and evaluating its progress provide the necessary information to marshal all the essential resources to react on time and reduce the overall damages.
TRANSFORMER MONITORING AND MANAGEMENT
Oil-filled power transformers constitute a significant asset for any owner and the impact of an unexpected failure or outage can have consequences of enormous proportion.
Power transformers have proven to be reliable in normal operation with a global failure rate of 1 to 2 per cent. The large investment in generating capacity after World War II and continuing into the early 1970s has resulted in a transformer population which in theory is fast approaching the end of life. The end of life of a transformer is typically defined as the loss of mechanical strength of the solid insulation in the windings.
With advances in testing techniques of the insulation system (oil and paper), by means of detecting the early signs of deterioration of the insulation system, it is becoming possible to extend the life of certain units, by continuous on-line monitoring.
With continuous on-line monitoring of the acknowledged key indicators or parameters of the operation of the transformers (such as gases dissolved in oil), it is possible to apply industry-recognized diagnostics of this operational data to determine the condition of the unit.
Thermal and electrical faults that develop in power transformers are always associated with the formation of gases dissolved in the oil. Analysis of these gases is a well-recognized method for the detection of incipient faults, the identification of the type of fault and monitoring its evolution with time. The sensitivity of this method is such that a developing fault can be detected long before the problem has become serious enough to bring in an alarm from the gas accumulation relay. This early warning allows the owner to do some advanced planning for repair or replacement of the equipment, before an unexpected failure occurs.
Distributed Intelligence Versus Centralized Data Storage and Analysis
Every transformer is different (much like a human). Each has its own peculiar attributes and operating characteristics. Even though transformers will be of the same design, manufactured in the same factory at the same time, and tested in the same manner, each is unique.
This arises from the manufacturing tolerances that are allowed in the factory. As an example, each unit will have different No Load and Load Loss values. These values have some small part in the loadability of a transformer.
An example of a system that takes advantage of computing power of a PC with real-time inputs and engineering test data of the transformer, is the FARADAYª Transformer Monitoring and Management System by GE Syprotec.
Acoustic Monitoring of OLTCs
On-load tap changers are critical components of power transformers. However, although they are designed with great care and undergo stringent acceptance tests, they remain a major cause of transformer failures. Appropri-ate on-line monitoring of this component would therefore be fully justified, provided a dependable diagnostic method is developed.
The acoustic-monitoring technique is non-invasive, reliable, suitable for new and old transformers, and can be implemented at a reasonable cost. It has been shown that acoustic monitoring of tap changers can detect arcing and contact wear as well as timing and mechanical malfunctions.
In order to take full advantage of this technique, however, we believe that the monitoring of the acoustic signature should not be limited to the diverter operation but also should cover the selector and reversing-switch operations. Figure 1 shows a recording of a diverter-type OLTC passing through the neutral position, which takes 9 seconds. Some OLTCs will take even more time to complete this operation. The system therefore has a 25-second recording capability, although the actual recording time must be set by the duration of the current in the motor.
Long-duration recordings will allow the detection of malfunctions of the selector switch or of the reversing switch. As can be seen in Figure 2 (for a different type of tap changer), a spurious noise is observed at 1.4 seconds, a peak which is not present when the same operation is recorded in the de-energized condition. The detection of this problem triggered an internal inspection of the tap changer, which confirmed severe arcing of the reversing switch. The contacts were therefore repaired, preventing further degradation that could have led to flashover in the tap changer.
Among other important characteristics of an OLTC acoustic monitoring system is the positive identification of the tap changer position, although this can be a problem for older designs where the driving mechanism has no provision for an additional rotary switch or diode matrix.
Equally important is the mathematical treatment applied to the acoustic signature before it is compared with the reference acoustic signature stored in the system memory.
Each one of the above-referred characteristics is required in a comprehensive OLTC monitoring system. The OLTC monitoring system developed in a Hydro-Quebec/GE Syprotec joint venture addresses each one of these characteristics.
The first three industrial prototypes have been in operation since October 1998 on three high-voltage transformers, each equipped with a different type of OLTC.
APPLICATION OF DISTRIBUTED INTELLIGENCE FOR TRANSFORMER MONITORING AND MANAGEMENT
As the importance of DGA is well established, many owners have decided to apply on-line gas-in-oil monitoring as a "first line of defense" in alerting them to a potentially damaging fault developing in the unit. When a unit has given the alarm, further DGA testing can then be carried out to determine the nature and severity of the fault.
Manitoba Hydro has installed HYDRAN¨ 201Ti Incipient Fault Monitors on all their oil-filled HVDC equipment.
Recently they experienced a series of alarms from the monitors on a HVDC converter transformer after the transformer experienced a consequential arc back (from the operation of the mercury arc valves) which resulted in a sudden increase in dissolved gases at a very steep rate (refer to Figure 3).
Three internal inspections were performed until a full understanding of the fault was gained.
The first internal inspection revealed a minor overheating spot with traces of arcing on the top clamp. This was repaired, but it was acknowledged that this spot was not very significant to explain the amount of gassing. Immediately after returning to service, the monitor detected that gassing had started again, which indicated the problem had not been found.
Once gases reached intolerable levels, a second internal inspection was performed. It was found that a number of bolts on the bottom clamps were arcing and overheating. These bolts were replaced and connections were improved with copper strips.
A few days after the unit was returned to service the gassing began again. A third internal inspection focused on examining the core packets, and it was found that four sections of core lamination packets were shorted somewhere in the oil channel. Another two sections were found to be sparking through the oil gap at 1000 V. This sparking explains the high content of hydrogen and elevated amounts of acetylene. The final picture was formed; they had core circulating currents due to shorted laminations and arcing, and had clamps circulating currents through bonding bolts on the bottom clamps.
As no spare unit was available to replace this single-phase transformer, a decision was taken to keep it in service, but to carry out intensive monitoring of the unit at the same time.
A continuous, on-line multiple-gas monitoring system, and a portable, on-line degasser were installed. This combination allows the resident expert at Manitoba Hydro to monitor the rate of individual gas generation and concentration, to determine when to start the on-line degasser to remove the gases from the oil. After an initial period of time, the degasser was put into continuous operation.
Without the application of this technology, it was felt that sampling and DGA would have had to be performed twice a day, seven days a week, in order to keep a serious situation under control. The substantial costs for this service were avoided.
The other option was to remove this bank of transformers from service, and wait the many months for a replacement transformer to arrive. This would represent a great loss of revenue for the owner. As the technology has proven successful in terms of reliability and repeatability of gas measurements, this transformer has been kept in service from June 1997 to the present date.
The transformer is scheduled to be removed from service, and replaced with a new transformer at an appropriate time.
Figure 4 details actual gassing history of the T21A transformer with the degasser in continuous operation. The rise in gases is indicative of overloading of the transformer, and details how the rate of generation of the gas exceeded the capacity of the degasser. The 'fall' of gases is indicative of the lower loading of the transformer, and the degasser removing the gas from the oil.
The economic benefit to Manitoba Hydro by condition-based monitoring, and taking action to degas the oil when it reached dangerously high levels, amounts to an average of $80,000 CDN per day in continued revenue.
Brian Sparling is Product Manager for GE Syprotec Inc. ET