New Forms of Demand Side Management Provide Value to Emerging Grid

by Bob Shively, Enerdynamics President and Lead Facilitator

Historically the role of demand side management (DSM) programs has been three-fold: reduce overall usage through energy efficiency (EE) efforts (for example, weatherization or more efficient light bulbs); reduce usage at times of system peak or system shortages (for example, direct load-control switches on air conditioners or hot water heaters that allow interruption by the utility); and shift demand from peak periods to off-peak periods (for example, ice storage systems for building cooling). The overriding purpose of such DSM programs has been to offer an alternative to building costly new power plants. But as the grid changes with the rise of renewable and distributed energy resources (DER), the role of DSM itself is positioned for dramatic change.

Increasingly, the future of DSM involves an integrated approach to DERs for managing energy demand and shifting load not only on the grid as a whole, but in specific locations to help defer the cost of distributed related upgrades.

~ Arizona Public Service statement in its Preliminary 2017 Integrated Resource Plan, October 2016 update


As the above quote from Arizona Public Service (APS) indicates, DSM will increasingly be asked to provide more flexible and even locational benefits. In some cases, it may have to address a net load curve that is rapidly shifting due to the growth of solar power. Over the next eight years, APS expects its net load curve[1] to change significantly:

Responsive image

source: Arizona Public Service Preliminary 2017 Integrated Resource Plan, October 2016 update

Absent DSM options or battery storage, APS must depend more and more on flexible gas plants that require a capital investment in a resource that is used for just a few hours in a day. [2]

Responsive image

source: Arizona Public Service Preliminary 2017 Integrated Resource Plan, October 2016 update

With California's renewable goal of 50% by 2030, the grid issues there are even more extreme. To determine how DSM might best help support the grid, the Lawrence Berkeley Lab recently worked with consulting firms E3 and Nexant along with various market participants to perform the study 2025 California Demand Response Potential Study – Charting California’s Demand Response Future. Over a two-year period, the team used customer-specific data to evaluate end-use and technology capabilities while focusing on two questions:

  1. What types of demand response services can meet California’s future grid needs?
  2. What is the expected resource base size and cost for demand response services?


A key part of the study was to move away from just thinking about traditional types of DSM and to ask what specific services does the grid need to best utilize DSM resources. The study identified four demand response (DR) service types that will be most helpful [3]:

Responsive image

The shed service is similar to traditional load management programs where load such as air conditioners or hot water heaters are curtailed during peak times.

 

Responsive image

The shift service is similar to traditional peak demand shifting, except that in the future grid it is expected that it will be necessary to shift load into the middle of the day to utilize the significant solar generation that will come onto the grid.

 

Responsive image

The shape service accomplishes the same load movement as shed or shift, but instead a service where the load is moved only when needed, the shape service results in permanent changes in load shapes.

 

Responsive imageThe shimmy service moves loads up or down quickly in response to specific system needs, possibly in time increments as small as every 5 minutes or even less.

 

To summarize the time frames in which each service would interact with the grid:

 

Responsive image

 

Likely end-use technologies included in California programs are electric vehicles (EVs), behind-the-meter batteries, air conditioning and HVAC systems, pool pumps, commercial lighting, commercial refrigeration, large industrial processes, agricultural pumping, data center loads, and wastewater treatment and pumping. 

The study’s medium demand response scenario for California found that the current time-of-use (TOU) rates and critical peak pricing (CPP) programs provide 1 GW of shed and 2 GW of shift. New programs could provide up to 10 to 20 GWh of daily shift, 2 to 10 GW of cost-effective shed, and 300 MW of load-following or regulation shimmy. So clearly there is significant potential here. 

What is needed to make this happen?

The study cover letter from the California Public Utilities Commission (CPUC) suggests that key steps include:

  • Investing in the integration of demand response into wholesale markets where it can be dispatched consistent with locational marginal prices
  • Enabling a new generation of demand response aggregators capable of delivering tailored options that work for customers with unique needs
  • Committing to default TOU rates for all customers
  • Committing to a greater differentiation of incentives based on relative locational value (meaning that DSM provided at one location might be paid more than the same DSM provided at a less valuable location)

As the grid continues to evolve, it will be necessary to reformulate traditional DSM programs to get the most potential from available flexible customer loads. Our expectation is that load resources will be increasingly important in allowing grids to integrate large amounts of renewables at the least cost possible.


Footnotes:

[1] The net load curve is the amount of load to be served by traditional generation after renewable output is subtracted from customer demand.

[2] For more discussion of changing load shapes and the need for more flexible resources, see our recent blog Renewables Require System Operators and Designers to Rapidly Respond to Changing Load Curves.

[3] These graphics are from the Lawrence Berkeley National Laboratory November 30, 2016 Final Draft Study Results Presentation.