Presented By:
Christopher E. Root
COO Vermont Electric Power Company
TechCon 2021
Summary
As the electric power system is rapidly evolving, the need for more data further into the power system has become even more critical for the reliability of the system. This paper will address several aspects of these growing needs and requirements. They include:
- Increasing data demands of the utility from its field assets.
- Private utility network expansions for operations, security and cost reasons.
- Future consideration of enhanced high-speed networks for both reliable operation and maintenance reasons.
Background on Vermont Electric Power Company (VELCO)
The state of Vermont has 16 distribution utilities which include one large investor owned utility (Green Mountain Power) and two electric coops, one large municipal utility and many smaller municipal electric departments. All of these groups are joint owners of the transmission company in Vermont (Vermont Electric Power Company or VELCO). VELCO owns over 700 miles of transmission (115kV, 230kV and 345kV), 52 high voltage substations, 1700 miles of fiber optic communications lines, one HVDC terminal and maintains a HVDC line with Canada. VELCO is the transmission operator for the state and operates under the direction and guidelines of the independent system operator in New England (ISO-NE).
The peak load in Vermont is approximately 1000 MW which occurs both in the summer and in the winter. These seasonal peak hours occur after dark. It is traditional that northern utilities have winter peaks after dark, but unusual that a northern state would have its summer peak hours after dark. This is a result of over 350MW of installed photovoltaic generation (35% of peak load).
Almost all of this generation is behind the meter distributed generation on the distribution system or is connected to the sub-transmission system. Most of the generators are less than 5MW in size with a few in the 20MW range. This results in most generators acting as a net load reducer. As the sun goes down, however, the load rapidly increases. Understanding the state of this generation is one reason for collecting more data from inside the distribution system so transmission operators can have better situation awareness. Most operational computer models do not accurately model this generation, which is distributed and is not directly telemetered. VELCO is working with the distribution utilities to send generating data for all distributed generators (greater the 150kW) back to VELCO via various channels.
The Vermont “Duck Curve”
High renewable penetration has fundamentally changed the timing of daily peaks. Traditionally, load peaked around 2 p.m. when commercial energy demand reached its highest point. Today load peaks after sundown.
More critical to grid operators, the daily load shape changes dramatically on sunny days, as shown in the below figure. This is the infamous “duck curve.” Where loads used to ramp up in the morning, peak in mid-afternoon, and drop gradually, on a sunny day, mid-day loads drop to the bottom of the duck’s belly, and then rise to their daily peak when the sun goes down.
The chart below shows the increasing levels of photovoltaic generation in the state and its impact on the net load over a 24-hour period over several progressive years. 2017 was the first year which the midday net load was lower than in the middle of the night. In 2019, the state set a new low net load for a sunny Sunday afternoon in the spring. It can be seen that the peak time of electrical consumption is after dark when the photovoltaics shut down.
A final point to make is to notice the flat curve during the day. This is a day when there were thick clouds and it was raining. This significantly reduces the level of solar generation and the difference between a sunny day and a rainy day is significant. The below chart from two days in the same week in 2019 shows how different it can be.
Increasing Data Demands Require New Collection Methods
Utilities across the country have discovered that data about the conditions of the power system assets is increasingly important. For many years, SCADA data has been collected and stored these in historical databases. As substations became more sophisticated, the amount of protection and control, and equipment monitoring data available rapidly increased. The old SCADA systems were being overtaxed with bringing back information to the control rooms which was not required to operate the system in real time. This led to discussions of how to bring this non-operational data back from substations in parallel to the SCADA communications and having it stored in a different database, which was more accessible to maintenance and planning engineers.
At VELCO, a large OSI-Soft PI database is used to collect data from everything from transformer and circuit breaker monitors to system protection information such as distance to fault calculations. After the physical attack on the PG&E substation in California a few years back, most utilities significantly improved physical security, which in many cases included high definition cameras which monitored the substation in real time. All these new applications, required increased bandwidth and high speed.
Within substations, most utilities have significant programs to modernize their protection and control schemes. Gone are the days of “dumb” electromechanical relays. These relays have been replaced with versatile multi-purpose smart relays. They have the capability to capture significant amounts of data that can be made available to engineers. In the past, when an engineer needed to analyze a post fault scenario, a technician was required to go the substation and download the relay information and bring it to the engineer. Today this can be done automatically but requires strong, secure high capacity communications. Asset management science has been embraced by the utility industry to lower costs, anticipate equipment failures, reduce field visits and improve reliability. One of the predicates of this strategy is to make better decisions based on the asset’s data. This is another major driver for data collection.
The cost of sending technicians to a substation to check on an alarm or to retrieve data continues to increase. Increasing the ability to do some of these functions remotely lowers the cost to the utility and can improve power system reliability. After a circuit breaker failure alarm and/or trip, for example, allows the control room operator to swing the camera’s view over to the failed piece of equipment to assist in the initial condition of the station. This could change the magnitude and type of response which may be needed (i.e. Fire Department).
Making the Case for Private Networks
Many years ago, the telecommunications systems were regulated, and utilities relied heavily on phone companies to collect primarily SCADA data from substations and provide local telephone services. In the years since de-regulation, costs have lowered for most services, but network reliability has had some lingering issues and the demand for communications bandwidth has increased. Additionally, the new cyber regulations for the utility’s transmission systems has become very stringent.
All of these data demands have led many utilities to transition away from contracted telecommunication services to internally owned and operated private fiber optic systems. Much of these systems are built on the utilities power line poles and with their underground power cable systems. By controlling these systems, the incremental cost of collecting increasing amounts of data is lowered and the schedules for construction of these systems have been under the control of the utility.
As a privately owned communication system, the utility can then apply all its cyber security systems both electronically and physically to protect these systems from hackers. These systems are generally capitalized and lead to a lowering of the operational costs while eliminating contracted communications costs. This approach also opens the possibility for utilities to sell additional fibers to others for a profit.
There are several ways for which fiber optics can be installed on transmission and distribution systems. They are slightly different.
The two types of overhead lines fiber optic options which can be installed in the power system are All Dielectric Self Supporting (ADSS) and Optical Ground Wire (OPGW). ADSS is usually strung under the phase conductors mounted to transmission towers. If there is enough clearance to ground and to the electrical phases for workers, then the fiber can be installed without replacing structures (assuming the structure strength has sufficient design margin). In Vermont, the decision made was to use ADSS fiber as it can installed fairly quickly and provides for easier installation of the splice boxes under the cable.
The second style of fiber optic cable that is very popular to use with transmission systems is OPGW. In this technology, fibers are imbedded in the stranded steel static wire at the top of the transmission tower. These are very reliable, but line outages are usually needed to replace conventional static wires with OPGW.
On distribution systems, some utilities will place ADSS fiber cables in the power space on a pole which the utility owns and controls or it can be installed in the communications space below the power space. One advantage of putting it in the power space, is that there is usually minimal “make ready” work required to keep the safety space between the electric and communications parts of the pole. When the ADSS is installed in the power space, all workers who have to work on it must be qualified electrical line workers as defined by OSHA. This would include having insulated bucket trucks and using the appropriate Personal Protection Equipment to be in the power space. Similar issues occur when the fiber is installed in urban manholes which contain power cables.
Future Consideration for Communications
As the needs for greater data and bandwidth mount, there does not seem to be an end in sight to the increases. This makes for an easier case to justify improving the transport systems for this data. With the major emphasis on reliability and resiliency in the power system, being able to assess the situation quickly and respond with the correct type and numbers of responders can mean minutes if not hours of delay in restoring customers. Having the ability to gather data and analyze it will only get more important in the future.
The cyber security of the power system in the last few years has been one of the highest priorities to utilities, regulators, CEO’s and governmental agencies. There are many Critical Infrastructure Protection (CIP) standards which have been issued by NERC that include the consequence of large fines for non-compliance. Defending the operation and data of the system is paramount to the security of the utility and the country. There are many bad actors, who on a daily basis attempt to penetrate the utility systems and there have been several cases where they have been successful. One vulnerability can be the communication system, which communicates between the control centers and the substations. Having control over this system makes it easier to design the complete fence around critical data. This improves cyber compliance and control during emergencies.
It is believed that NERC will continue to identify the communication networks as vulnerable to disruptions and critical to getting more information about the system (i.e. Distributed Generation levels). Thus, the importance for utilities to take control into their own hands and build, maintain and protect the communication systems they rely on to operate the grid continues to increase.
Summary
This paper has tried to present the case that as our grid evolves, data will be an even more important aspect for the operation of the power system. Security, volume of data, control and costs are all factors to consider when analyzing the decision to build a private utility data and communications network.
