- July 13, 2022
- Posted by: admin
- Categories: Condition Monitoring, Remote Monitoring
A transformer is a critical asset as it is directly responsible to manage electricity, which is the core product. The nature of the asset and its distributed set-up mandates a reliability-centered maintenance strategy.
The operational life of a transformer starts with its provisioning/commissioning, after its installation at the substation. Once in operation, a maintenance strategy needs to be in place for effective operations.
Transformer Operation and Maintenance Cycle
A well-planned maintenance strategy is essential to provide the required reliability, performance, and availability to the transformer over its lifetime at a minimum cost. The objective of a good maintenance strategy is to identify any abnormalities before they cause avoidable damages.
Once an abnormality is detected, predetermined diagnostic techniques are applied to identify the severity of the problem, localize it, and determine if the transformer can be brought back into service without stopping the operation.
If necessary, the applicable corrective action can then be taken, or depending on the transformer condition, it may be appropriate to perform a more intensive intervention on the transformer.
Finally, it may be decided that it is time to refurbish or repair the transformer or even replace it, depending on the results of the final finding that will include consideration of the safety (both to engineers and the public), the possible environmental consequences and the system reliability aspects of continuing the operations.
Elements of Transformer Maintenance Strategy
Here is a transformer operation and maintenance cycle, from the time of commissioning the transformer to the end of life.
If we study the above diagram, we see the five elements of maintenance strategy. Let’s understand them briefly.
Time-based Condition Monitoring
These are certain checks and tests to evaluate the transformer condition which are carried out at regular and pre-planned intervals.
These are most often carried out in parallel with other maintenance tasks, particularly with tasks that require an equipment shutdown, to avoid redundant downtime. The results of time-based condition monitoring are used to decide on the set of maintenance activities required at the time as well as in the future.
Time-based maintenance is planned and done at a predetermined frequency based on the age of equipment and types of tests. This frequency ranges from daily to annually.
The aim is to reduce the likelihood of a part of the equipment failing in service. This includes maintenance tasks to improve the condition of the transformer such as oil change, lubrication, and preventive replacement of parts.
The maintenance checks are carried out according to the planned schedule and frequency, irrespective of the equipment condition. This maintenance method can offer very good risk coverage if the schedule provided by the OEM is followed strictly.
This method has the advantage of being easily planned, but it’s not the most affordable maintenance strategy.
In this method, maintenance is subject to transformer conditions to reduce the chances of the equipment or a part of the equipment failing in service.
Also known as conditional preventive maintenance, it considers the actual physical condition of the transformer and takes many factors into accounts such as equipment usage, the occurrence of events, possible wear or tear of moving or current switching parts, and the performance of similar equipment (performance modeling).
To implement this method, it is required to assess the transformer condition using other methods such as time-based condition monitoring or online condition monitoring. Time-based management or online condition monitoring helps assess the equipment’s condition. This assessment is used to decide if maintenance is required or not.
Condition-based maintenance helps reduce costs by performing maintenance only when needed. The need is defined based on a change in equipment conditions. Identifying such needs requires a robust planning process.
Online Condition Monitoring
With the rise of IoT and other technologies, it has become easier to connect and monitor industrial assets. This method helps to perform some checks while the transformer is in operation using sensors or existing communication systems like SCADA.
Some examples are DGA, infrared thermal scanning, or as simple as oil level monitoring.
This monitoring could be done frequently like daily/hourly or can be continuous. Continuous online monitoring enables real-time visibility of an asset and generates alarms or notifications for thresholds or anomalies. For example, if the oil temperature goes beyond defined thresholds, it triggers an alert and notifies the concerned person.
This process assesses the transformer condition by considering various factors that could affect future performance.
The process output helps to determine whether a transformer can carry out the operations as expected or not. If not, then the action trigger will be the kind of work/repair it requires, minor or major.
The factors could be testing and measurement results, operating history, knowledge of the failure mechanisms and processes, observations, previous experience with similar equipment, and other relevant data and information.
The output could be simple normal or abnormal labels or slightly complex numerical output identified as ‘asset health index’.
The asset health index is a scoring system that denotes asset condition and helps with decisions on future maintenance or replacement.
Maintenance strategies are normally defined based on operation time (like carrying out DGA tests annually) or transformer condition (asset health index).
Reliability-centered Maintenance is an optimized strategy that considers the asset’s position in the network, its operational importance, any potential safety or environmental risk arising from its failure, and any likely consequence of its potential outage.
This strategy takes the overall impact of transformer failure into account. For each asset a criticality index is calculated which indicates the need for reliability of the asset in the system. The criticality index is calculated by combining the transformer’s safety, environmental and operational criticality factors.
This index is used to determine future maintenance tasks, maintenance frequency, and priority. This leads to transformers in risky or important positions being maintained more intensively compared to transformers in a position where less reliability is okay.
Reliability-centered maintenance strategy also takes risk score into account for maintenance decisions. A risk score is a product of the likelihood of failure and failure consequences.
Risk = likelihood of failure * failure consequence
This helps to give more attention to transformers which have high failure consequences. An important transformer serving many power consumers may require a higher level of maintenance if a backup unit is not immediately available.
The likelihood of failure can be determined using the asset health index.
For example, Transformer “K” is in a poor technical condition having a health index of 15 (i.e. a higher likelihood of failure) and serves a critical area (as a result, in case of an outage or failure, there are severe consequences). This will lead to a higher risk index. Due to the risky nature, the transformer should be monitored using continuous online monitoring and maintenance should be more frequent.
For efficient maintenance optimization, Reliability-centered maintenance recommends taking into account all maintenance strategies. This strategy doesn’t only consider a failure in planning but also the consequences of failure.
Implementing real-time online monitoring helps in assessing new developments in your transformers efficiently and without human intervention. It detects anomalies very early, which helps to plan the repair or replacement of the transformer with high priority to reduce the risk.