Flexibility

Motivation for flexibility to be considered more specifically in planning studies

Increasing shares of renewable generation, particularly wind and solar power, as well as the proliferation of new demand side resources such as battery energy storage, demand response and electric vehicles are challenging the power system planning paradigm. Whether these are driven by climate and other energy policies, customer choice or cost reductions in technology, there will be a need to change power system planning processes to more efficiently and reliably integrate these resources. One of the key characteristics of wind and solar resources is their variability and uncertainty [1]. The output of these resources is weather-driven, and a significant portion of solar photovoltaic (PV) resources is expected to be behind the meter. As such, they are difficult to forecast perfectly (uncertain), and their output tends to be variable as well as only partly dispatchable. They can be dispatched if control capabilities exist, but even so are likely to only be dispatched down – while they have the potential to be curtailed to provide upwards dispatch capability, this is not often done. This has increased the need for and value of operational flexibility in power systems [2]. Flexibility here refers to the operational maneuverability of the set of resources available to the system. This includes the ability to ramp and dispatch resources to manage variability and uncertainty of supply and demand, which may become more challenging and relevant in systems with high shares of renewable generation.

System flexibility, like many other aspects of the bulk electric system, can be examined as a supply and demand problem. A holistic assessment of flexibility will examine the resources available to provide flexibility and the factors driving the demand or requirements for flexibility. These can either be assessed separately or at the same time within a power system dispatch simulation tool.

As an example of the need for flexibility, Figure 1‑1 shows the variability seen in Germany due to wind and solar production. Net load, the difference between the total load and the load less variable generation, is an important concept here, since it exhibits greater variability when compared against the load. The net load is typically served by dispatchable resources, although other factors also come into play, such as non-dispatchable generation like geothermal and nuclear, interties with neighboring regions, and the fact that wind and solar can be dispatched if control systems are enabled. While the focus is on managing load net of wind/solar, one may also want to consider these other aspects in planning studies, particularly if the inflexibility related to these aspects will impact on the ability to balance supply and demand. Figure 1‑2 shows the uncertainty associated with solar power, based on multiple different forecasts provided for the same day for a solar plant in Texas [3]; this shows how different forecasting models predicted the same day. This uncertainty also needs to be accommodated together with the variability of net load. The combined impact of variability and uncertainty of wind and solar power have been covered extensively in the past; in the context of these guidelines, the main point is that there is a need to ensure sufficient operational flexibility to manage this variability and uncertainty.

[[File:./flex-assets/media/image2.png|509x394px]]

Figure 1‑1
Variability of Wind and Solar Power in Germany over 3-day period

[[File:./flex-assets/media/image3.emf|460x356px]]

Figure 1‑2:
Uncertainty in Solar Power Output for a Plant in Texas, based on different solar forecasts

Another manner to examine the variability and uncertainty is to calculate the ramping mileage of the net load on the system. This measure, defined as the absolute ramping summed over the year, allows one to understand how much flexibility is required in different systems. To calculate mileage, the absolute ramps are determined across each interval of data and summed. For example, for a net load of [10 MW, 12 MW, 15 MW 10 MW], the total mileage would be 2 MW + 3 MW + 5 MW = 10 MW across the time period, even though the largest up ramp is 3 MW and down ramp is 5 MW. This measure allows for an understanding of how much additional ramping can be required, even if the largest ramps don’t always get significantly larger. EPRI calculated this for several European and US utility regions, as is shown in Figure 1‑3 and Figure 1‑4, and plotted against annual demand.

[[File:./flex-assets/media/image4.png|633x270px]]

Figure 1‑3
Ramping Mileage for Selected Balancing Areas in Europe for Demand Only and Net Load

[[File:./flex-assets/media/image5.png|637x252px]]

Figure 1‑4
Ramping Mileage for Selected Balancing Areas in the US for Demand Only and Net Load

As can be seen when looking only at demand, the variability is proportional to the size of the system; some systems such as the UK tend to exhibit slightly higher variability compared to most other regions when considering size. However, when net demand is considered, most regions see an increase in ramping mileage. This would be expected due to the variability of wind and solar PV. It can also be seen that certain regions such as the UK, Denmark and ERCOT appear to see a greater additional mileage than others; this may be due to their specific load shapes, or the nature of the VER.

Examining this mileage measure may provide useful insights into the flexibility needs of the system, but is still not widely understood; therefore EPRI is continuing to examine whether and how such a measure could be helpful when considering flexibility issues. Factors such as the data resolution, normalization methods and the flexibility in the wind and solar themselves will also need to be considered when assessing and comparing this metric.

EPRI’s Integrated Energy Network initiative

EPRI launched the Integrated Energy Network (IEN) initiative in 2017. This is a large initiative, focused on a range of issues crucial to the future of the energy system. In particular, focus is put on integrating different sources of energy together such as electricity, heat and transport. This will require increased coordination in planning and operating energy systems. As such, one aspect was to develop a concept called Integrated Energy Network Planning (IEN-P). This white paper, published in mid-2018 , identified the main challenges associated with the planning of future energy systems, with a particular focus on multiple aspects of the electricity system, namely generation, transmission and distribution.

Of these ten identified challenges, several are directly related to the guidelines presented here. This includes the following:

• Incorporating Operational Detail – As emerging power system resources (e.g. storage, DER, PV, etc.) replace synchronous generators, which traditionally have provided needed operational reliability services, resource planners will need to consider the potential reliability impacts and operational reliability capabilities of candidate resources. From a flexibility perspective, the need for capturing operational flexibility will become increasingly important.
• Increased Modeling Granularity – In terms of flexibility needs, increasing modeling granularity to represent intra-hour ramping, as well as increased spatial granularity to represent location issues (as discussed related to deliverable flexibility later), will allow flexibility to be better considered in planning tools.
• Integrating Generation, Transmission and Distribution Planning – Future resource planning will require closer interaction of planners across the entire electricity supply chain to understand how decisions at one planning level may impact other levels and the ability to tradeoff potential investments between systems to optimize the future electric power system. When obtaining flexibility from distribution systems in particular, models will need to be able to consider how these resources are limited in providing flexibility.
• Expanded Analysis Boundaries and Interfaces – Electric companies are beginning to be asked by regulators and external stakeholders to address issues outside of electric company service territories and even beyond the electric sector of the economy as part of their resource planning activities. Many of these changes are driven by the fact that large areas can reduce the need for or availability of flexibility resources, while flexibility may also be available from other energy systems.
• Incorporating New Planning Constraints – Future resource plans will need to be optimized to achieve objectives beyond traditional least-cost resource adequacy; flexibility metrics are clearly starting to be incorporated as discussed in the examples above.
• Integrating Wholesale Power Markets – Increasingly, planners will need to consider the evolution of wholesale power markets that provide opportunities for companies to buy and sell capacity, energy and ancillary services, and the impact of markets on the economic viability of resources that provide reliability services and other desired system attributes. This factor includes the need to consider market changes such as the recent development of flexible ramping products, or concepts such as that described in California above.
• Addressing Uncertainty and Managing Risk – There is a growing need for resource planners to explicitly account for key uncertainties when developing resource plans and to adopt new approaches to manage evolving corporate risks. Uncertainties related to short term operations, and the flexibility required to manage that uncertainty should be included.

Other challenges were also identified related to stakeholder engagement, long term forecasting and customer modeling; these do not directly refer to flexibility issues, but still should be considered in planning. Therefore, as the IEN-P progresses, the flexibility guidelines described in this update will be coordinated with that activity to ensure consistency across the efforts.

Motivation for guidelines document

These guidelines are designed to help planners integrate the concept of flexibility and flexible resources into the planning processes. They would likely augment or improve upon the methods described in previous section. The term ‘planners’ here includes a variety of different types of planning functions, including resource adequacy assessment, transmission planning, resource planning (e.g. Integrated Resource Planning or other resource planning activities), as well as high level policy-type analysis. All of these are linked, and in some cases may involve the same analysis. However, depending on the given regulatory regime, how different aspects link may differ. For example, a vertically integrated utility may have all of these functions very tightly linked, an Independent System Operator (ISO) region may have the ISO make decisions based on proposals from stakeholders, and could include capacity markets or not, and other entities such as Generation & Transmission companies may have other processes. The guidelines here are intended to be applicable to a wide range of different end users, and as such focus more on the engineering fundamentals.

Given the range of different tools available, and the trend to model operational type issues in planning that traditionally were outside the scope of planner’s functions, this guideline document is an attempt to gather the most recent thinking, at EPRI and elsewhere, on this process and to make recommendations for the choices a planner will face. It is not intended to be a definitive process, but instead a set of potential analyses and studies that should be considered when integrating renewables. This is likely to continue to evolve over time, and will be updated accordingly. Individual planning functions within different entities will also need to adopt this general document to their own processes. As described later, the amount of work done by EPRI in this area in the past several years has been extensive. The main aim of the guidelines document is to collate the main lessons learned and outline the methods developed in one place.