SDG&E Takes Proactive Approach to Maintenance

Across North America, electric utility infrastructure is aging. As infrastructure reaches the end of its life, utilities are faced with the inevitable expense of replacing costly equipment, such as substation transformers. However, San Diego Gas & Electric (SDG&E) has embarked on a condition-based maintenance (CBM) program that can extend the life of valuable infrastructure by remotely “sensing” potential problems and alerting the SDG&E control room that repairs are necessary. This not only saves money by reducing maintenance trips in the field, it can prompt repair of vital equipment before a failure occurs.

The utility began the program in 2009 with the plan to install CBM equipment on all substation transformers by 2015. So far, the utility can evaluate the health of the transformer bushings, cooling fans and pumps, load tap changers (LTCs), core temperatures and dissolved gases of more than 65% of its substation transformers.

The system consists of monitors placed physically at strategic locations on the transformer main tank, high-voltage bushings and LTCs. The data is transmitted to the input and output ports of a central processor unit, mounted adjacent to each transformer. From there, the data is transported via SDG&E enterprise gateways to analytic software in the control room configured to generate alerts and send notifications with levels of criticality.

The dissolved gas analysis (DGA) monitor, also mounted adjacent to each transformer, extracts oil samples through lines that tap directly into the transformer oil valves. The new system provides oil sampling data and dissolved gas concentration levels, and measures the rate of change in those levels over time, which is then transmitted back to grid manager for more detailed analysis.

To enhance its ability to analyze the large amount of data collected, SDG&E invested in a software visualization tool. This program, which recently went live, allows CBM operators to access data from multiple diagnostic sources. These sources are used to monitor and analyze the health of the asset through a single application. The system was designed to be flexible enough to continually refine the analytics, and to include additional data and industry best practices.

Reaping the Benefits

SDG&E moved to a CBM strategy for several reasons. The discrete alert annunciation sends individual notifications directly to the maintenance group. These notifications provide crews with more detailed information about the necessary maintenance before they even arrive on site and allow for better productivity.

For example, the CBM system allows technicians to detect the loss of cooling fans or pumps. If an element of the cooling system is not working, the loss of current to the fans triggers an auxiliary relay contact tied to an input on the CBM monitor. This input triggers a notification to the maintenance group with an alert level based on criticality. A fully functioning cooling system is essential to preventing the transformer from overheating, therefore, lowering load capacity.

Additionally, sensors generate notification to the utility if there is a breach in the integrity of the conservator tank bladder. This is vital to ensure the transformer remains a closed system and is void of outside contaminants.

The LTC is another important component of the transformer being examined. The LTC motor energy is monitored to determine the necessary amount of current to advance the LTC through each tap position. This provides insight into the health of the motor and operating mechanism. Temperature differentials between the LTC oil and main tank oil are indicative of the LTC contact condition. For more critical banks, DGA monitors are used on the LTC tank and the main tank.

However, one of the most critical benefits of CBM is the fact SDG&E engineers, technicians and maintenance professionals can obtain real-time data, view the analysis and make timely decisions to save the asset from premature failure and extend the life of the asset. In the past, technicians manually extracted transformer oil samples on an annual basis and sent them to the lab for DGA. Now, SDG&E performs DGA automatically every six hours and as frequently as every one to two hours if certain thresholds are reached.

Viewing the rate of change in the gases, to understand the nature of the problem and correlate it to conditions on the system, allows for more informed decisions as opposed to reacting with only limited information available. As an added benefit, the risks of error introduced in the manual sampling and testing procedures are virtually eliminated. The same was true of the cooling system, which also used to require a manual inspection to determine if it was functioning properly. Now the cooling system is exercised once a month.

The utility also has realized benefits when it came to monitoring bushings. In the past, technicians had to take an outage to test bushings. Now tests are automatically run every hour, and an outage is only required for the initial monitor installation.

The physical design and installation of CBM equipment has offered other benefits such as correcting station prints and bringing them up to current standards. The utility also can identify and correct broken devices, such as fans and winding gauges, and identify and correct malfunctioning alarms.

Preventing Potential Problems

The CBM program already has saved the utility a significant amount of expense. For example, in August 2010, an energized 12-kV capacitor caused a 6-MVA transformer to overload and produce heating gases in the oil. Fortunately, the CBM system alerted the utility to the condition. The capacitors were de-energized through supervisory control and data acquisition (SCADA), resolving the overload condition and preventing premature aging of the transformer.

A few months later, in another 69-kV distribution bank, the CBM system alerted a technician to an escalating current imbalance on the high-side bushings as it exceeded the 3% (level 3) and even the 6% (level 4) alert modes. These alarm levels occurred at higher temperatures, so the technician took the bank out of service and replaced the bushings. Through the analysis, SDG&E discovered the partial discharge was from the bushing and not from the transformer. The technician was able to detect the problem early on before it translated into a major failure.

During the summer 2011, a distribution bank signaled a bushing health monitor alert on a C-phase bushing. In 2008, the bushing was tested to have a 0.3 power factor, but it had climbed to a 2.24 power factor three years later. As it started to escalate, SDG&E took it out of service because the power factor was high. The technicians ended up changing out all three bushings because they were all of a similar vintage and had a high power factor. As a result, the utility was able to save the transformer, as the bushings had not been slated for testing for another five years.

SDG&E also has realized benefits with transmission banks. In March 2011, technicians installed new bushings and processed oil in a bank. Just a few weeks later, the bank tripped out as a result of the activation of a sudden pressure device. Technicians checked the relays but did not find any faults. A current sample was requested from the DGA monitor and it verified no heating gases had been created since the last sample taken just two hours earlier. This real-time verification allowed the bank to be re-energized much sooner than previously was possible.

Analyzing Data

With the new visualization tool, SDG&E highlights each substation on to the embedded Google map, as well as its health and risk factor attributes. Integrated with data from SCADA, the maintenance management system and geographic information system, the utility can easily visualize what is going on with the unit. The technicians can read dashboards, trend SCADA data and be flexible enough to do an analysis under one application.

For data analysis on a real-time basis, recorded gasses from the DGA monitors are sent to a Web-based service provider. The configurable software is continually under development through collaboration with the utility and other subscribers. Detected abnormal DGA or moisture conditions result in e-mail alerts and are updated within the visualization application.

The utility is also collaborating with an industry leader in diagnostic instruments and power systems engineering on enhanced analytics to estimate daily loss of life from the effects of moisture, oxidation and thermal heating, bubble formation risk, and dynamic emergency loading capacity and detection of LTC problems for parallel operated transformers.

Using these algorithms to continuously analyze data and provide an up-to-date health assessment is invaluable for both replacement strategies as well as prioritizing CBM schedules.

Looking Ahead to the Future

Transformers can cost up to $3.5 million, but proper and proactive maintenance can save the equipment from premature failure, extend its useful life and reduce unplanned customer outages. The utility considered the customer impact and field operations when building the business case for its CBM program. These programs also coincide with the utility's smart grid initiatives and are helping SDG&E become a national leader in effectively analyzing and using increased amounts of data to benefit customers and the region overall.

As the program expands in the future, the utility is considering using the CBM strategy across other pieces of equipment, such as gas circuit breakers and substation batteries. This will help SDG&E reduce SF6 greenhouse gas emissions and prepare for new battery regulations on distribution substations of 110 kV and above. The utility is installing sensors on 500-kV circuit breakers to test this concept in the field.

SDG&E has a dedicated crew for preventive maintenance to review data and perform analysis, and a CBM team to manage the deployment. This endeavor requires committed personnel and will result in long-term benefits, including better care of assets and more effective maintenance that extends the life of transformers. The data helps the utility avoid unnecessary replacement expense while keeping its transformers and substations in top-performing condition. SDG&E will continue to install new CBM and other smart grid technology that enhances reliability and creates a more responsive and resilient grid for the benefit of customers and the community.

Janet Lonneker ([email protected]) is a project manager for San Diego Gas & Electric's asset management and smart grid projects implementation group. She manages a variety of smart grid projects for SDG&E and holds an advanced degree in electrical engineering.

Brent Baker ([email protected]) is a construction supervisor for San Diego Gas & Electric, responsible for supervising the condition-based maintenance program, transformer voltage and vegetation management crews. He has helped with the piloting and implementation of monitoring equipment and transformers since 2008.