For all the opinions voiced about the electric power sector, there is one fact that is often overlooked: the grid works, and it does so with a remarkably high degree of certainty.

Unlimited access to high-quality power has enabled industries to flourish in this country. From aluminum smelting to financial markets to cloud computing, industries have grown due to the delivery of high-quality power. Our confidence in the grid is built on over 100 years of successful voltage management by distribution utilities and successful frequency management by balancing authorities.

The foundation to this has been the implicit deal between retail customers and their distribution utilities and balancing authorities: that customers will help cover the fixed costs of maintaining system stability and meeting peak demand in exchange for high-quality power.

Two recent developments have changed this deal: the deregulation of wholesale markets and the rapidly falling costs of distributed generation. While these developments have positive impacts in the long term, they have system implications that have put utilities in a tight spot.

Distribution utilities earn revenue by billing retail customers according to fixed rates. At the same time, these utilities must purchase power in wholesale markets with volatile price changes. The proliferation of advanced metering infrastructure is enabling utilities to shift toward time-of-use and real-time pricing and create more complex bills. This helps align revenue and costs, but doesn’t address the second issue related to the rapid rise of distributed generation, in particular solar PV systems.

Historically, electrical current flows from central generators across the transmission grid to distribution circuits and ultimately serves load at the end customer site. Grid assets are built to accommodate peak customer load and to manage the volatility in demand at each level of the grid. The system was not built to manage uncontrolled two-way traffic safely or to import significant generation at distribution voltages.

Further, PV output is highly volatile, capable of changing output by over 80 percent within a few seconds.

Imagine a solar customer that is exporting power at 2 p.m. despite a large HVAC load. If a cloud interrupts solar production, the customer rapidly becomes a large power consumer; and when the cloud passes, the customer swings back again to being a generator. This volatility disrupts the flow of current on the distribution grid unpredictably and at high speeds, challenging the distribution system to manage voltage within mandated quality levels. While the effects of these swings can be smoothed out in aggregate, customers located on the same distribution feeder may experience power quality issues. The speed of volatility is important because traditional voltage regulating assets (load-tap changers, voltage regulators, capacitors, etc.) are electromechanical devices that generally react slowly to manage unidirectional demand volatility.

When voltage deviates from mandated limits, it can result in fines to the utility, end-use equipment damage, or even grid asset failure. The costs of these consequences are ultimately borne by the entire rate base.

There are a few possible solutions to address the technical challenges from distributed PV volatility: curtail PV production, invest in more robust distribution grid infrastructure to support greater volatility, or solve the problem at the source by making customers accountable for the system effects of highly variable loads.

Limiting distributed PV development should not be an option, but without a solution to the volatility challenges, distribution utilities may have no choice. PV curtailment could work, but introducing such a mechanism would reduce PV production and jeopardize solar’s hard-earned bankability.

Utilities could also invest heavily in distribution grid equipment to guard against PV volatility. However, Germany offers a cautionary tale. While the wholesale price of energy in Germany is falling due to high adoption of wind and distributed PV, the retail price to customers has increased. The disparity comes from the infrastructure investment required to deliver high quality power in the face of increased volatility. The investment required to support high penetration of PV in the U.S. would be greater than in Germany because of the lower U.S. geographic density, long radial network topology and other technical characteristics.

The third option is to solve the volatility problem at the source. If customers with erratic load or distributed generation were incentivized under a new “deal” to either smooth their load or pay for increased volatility, they would have the option to solve the problem themselves or pay the utility to do so. Customers already have a suite of products available to smooth their net demand if incentivized to do so through utility demand charges or even a newly defined “volatility charge.”

Load management and intelligent energy storage solutions are commercially available today to lower peak demand, smooth volatility, and, in so doing, create value on both sides of the utility meter. Distributed energy storage can be used to mitigate voltage issues along a circuit by intelligently injecting and absorbing power to level demand. This ensures that demand swings from one customer do not affect the power quality for their neighbors.

By addressing volatility at the edges of the network, where the problem originates, we can enable continued expansion of PV and other distributed generation beyond today’s limits, without jeopardizing grid stability.


Joe Matamoros is Vice President of Commercial Operations at Stem. Will Fadrhonc is Manager of Grid Solutions at Stem.