Reliability centered asset management for power transformers in Turkish national power transmission system

Restructuring process of electric power industry from conventional past to competitive and smart future has increased the importance of maintenance and brought the necessity of improved cost-effective maintenance management tasks. Conventional maintenance methods were generally concentrated on the physical conditions of the asset, its performance and environmental conditions. The new maintenance programs, however, are so organized to include the impacts of maintenance interruption and asset deterioration on system reliability degradation. The main task of electric power utilities is to provide sufficient, reliable and high-quality electrical energy for their customers with the least costs. Deregulation in electric power systems during the last two decades has increased the competition in electricity markets. This in turn forced the energy suppliers to optimize the utilization of their equipment, focusing on technical and cost-effective aspects. The transmission network as being the back bone of the power system should be well planned, properly designed and well maintained to satisfy the expectations. In addition to high quality asset usage and proper design methodology of the planning phase, improved cost-effective maintenance process should be applied during the operation phase of the system. Transmission lines/underground cables, towers, insulators, fittings, transformers, reactors, circuit breakers, instrument transformers and protection equipment constitute the major assets of an electric power transmission system. The strength of the components should be increased by appropriate maintenance strategies in order to increase the overall system reliability as well as to decrease the total costs. Reliability Centered Maintenance (RCM) is the assignment of optimal maintenance strategy for each component in a system, taking into account component importance/priority as well as component failure modes and their effects on system reliability. It can be considered as an optimization of reliability and maintenance tasks with respect to operational requirements. In conventional RCM scheduling, failure modes resulting in loss of performance or lack of functioning of the components are first identified. Then, an effective maintenance method is developed thereby prioritizing those components and failure modes. The aim is to achieve cost effective maintenance procedure by controlling the maintenance performance, which implies a trade-off between corrective and preventive maintenances. RCM experience has provided a basis for Reliability Centered Asset Management (RCAM) procedure. RCAM aims to develop an optimum maintenance process providing the maximum reliability with minimized component costs. RCAM has evolved throughout the years by removing the inadequacies of conventional RCM methods and/or by improving it to cover some additional maintenance tasks. Power transformers are the most important assets of transmission infrastructures as their outages due to unexpected failures may create severe problems both from economic and from technical point of views. They are operated generally close to their ratings and even above it for short periods. Moreover, some of them are operated near or beyond the intended life cycles in our national electric transmission system. Therefore, RCAM studies for power transformers have been of great interest for the researchers. However, none of the aforementioned studies could be applied for a practical RCAM of all power transformers due to some site-specific operating conditions and maintenance procedures. This study presents RCAM process for electric power transmission system and RCAM based maintenance strateg for 154 and 380 kV transformers at te Thrace part of Turkish National Power Transmission System. The proposed RCAM plan aims to improve the transmission system reliability and the cost of the components by adjusting maintenance parameters instead of revising the mainenance tasks of system components. The study consists of two main phases. The first phase aims component-level reliability improvement using individual reliability parameters of a power transformer by Markov state model. In the second phase, this improvement is expanded to the system-level. All the power transformers of the transmission system are classified with respect to their criticality/importance for the system and a strategic maintenance plan is constructed with respect to their criticallitty measures. The second phase also includes the verification of the adventages of the proposed maintenance plan using expected parameters. In the first step of the first phase of the study, the importance of the main components of the transmission system; i.e. transformers, circuit breakers, transmission lines, protection systems, and substations were quantified by the weighting factors using a decision table. Those components are then prioritized to determine the appropriate maintenance tasks by quantifying the failure effects and the maintenance necessity. At first, transmission system is divided into components and the components are then divided into main subcomponents. A decision-logic is applied to the subcomponents and the resulting weighing coefficients are used for component prioritization. Appropriate maintenance process for each main component was determined with respect to the resulting weights. Then current maintenance procedure of a power transformer was modeled by a semi-Markov state diagram with respect to past failure data and maintenance performance indices provided by Turkish National Power Transmission Company (TEIAS). At the second step, the concordance between the past performance and the constructed model was validated by optimizing state transition rates of a simplified 3-state model. Then sensitivity analysis was performed to determine effective transition rates of the model. At the third step, reliability indices and total maintenance and interruption (failure) costs of a power transformer were calculated for the existing maintenance procedure. Revised (optimal) maintenance rates providing technical and economic benefits were then determined. Finally, reliability indices, total cost and total life time for the proposed maintenance rates were calculated and compared with those of the existing ones to show the validity and the advantages of the proposed asset management strategy. At the first step of the second phase of the study, RCAM study for individual transformers is extended to transmission system level. Transformer criticality assessment is used as a tool for this enhancement. Among several decision making methods, fuzzy-logic based methods were found to be more appropriate due to uncertain and stochastic behavior of power transmission system. Many fuzzy decision making tools provide hierarchic enumeration of components with respect to their importance of criteria. However, cost-effective and applicable maintenance strategy for a large-scale power transmission system involves using limited number of transformer classes instead of their hierarchical enumeration. Fuzzy Multi-Attribute Scoring (FMAS) Method provides useful results to classify the transformers with smooth transitions between their criticality classes. This procedure consists of three stages. At first, appropriate criticality criteria are defined and calculated. The criteria are then fuzzified and primary decision is made using the FMAS process at the second stage. Final decision is made by integration of primary decision and qualitative criteria at the third stage. The second step of the second phase is devoted to the verification of the expected benefits from the RCAM study. At first, several reliability indices such as unavailability, failure frequency, average outage duration and life time, and total operation and maintenance cost are calculated for the current maintenance procedure. Then the same indices are calculated for the system where the proposed AM procedure is applied to the power transformers. The results are then compared to state the benefits of the proposed AM scheme. It is clear that an exact verfication involves the collection of a data at least for 5-10 years for the applied maintenance plan. Following the data collection process, system-level reliability indices should be recalculated and are used to validate if the expected benefits are satisfied or not. The results of component-level studies (the first phase) showed that the proposed maintenance rates provided up to 19%, 36% and 19% improvements for unavailability and average annual outage duration, total lifecycle and cost of the transformer, respectively. However, those improvements were the expected uppermost values and the exact values can only be determined from the field data following the application of the proposed maintenance periods. Therefore, the improvements were given as parameter-dependent values. The second phase of the RCAM study also provides an enhancement of the component-level improvements to transmission system-level thereby constructing a useful criticality assessment method and an Asset Management Plan. According to the assessment of the expected benefits of this plan, overall reliability indice improvement with respect to existing maintenance procedures was found to be 12,9%. However, this was a probabilistic expectation and depends on the strategic targets of the company. The real benefits of this application will be actualized following appropriate application. Note that this study is the first one that aims to achieve cost-effective maintenance process for the power transformers by optimizing the controllable transition rates of the maintenance proces. When combined with improved maintenance tasks, it will probably offer the best RCAM procedures to the transmission utilities.