Manufacturing processes represent large, concentrated loads that can provide significant flexibility benefits by modulating their power demand. The advantage of focusing on such industrial-type users is that—unlike residential and commercial users—their electricity consumption is largely independent of human preferences and schedules. This means that their loads may be shaped and changed in time to assist with grid operations much more readily than, say, those of commercial or residential buildings.
Broadly speaking, manufacturing processes can assist grid operations by carrying out two complementary actions. In one, production is increased during off-peak hours, along with the grid load of the process. This means that the process temporarily generates products in excess of product demand, and the excess is stored. In the other action, process operators reduce the production rate (and associated load) during peak electricity demand hours. Since production dips below nominal levels at this point, some of the products that were previously stored are used to meet the demand of the customers of the process.
While some manufacturing companies are already providing flexibility to the grid, many are not yet engaged in this type of grid support. This blog will discuss two key questions concerning increasing this important source of flexible loads. First, what are the physical requirements and limitations on carrying out these actions? And, second, how can the operators of a manufacturing process be incentivized to act and provide grid flexibility?
Physical Requirements and Limitations
There are two considerations regarding physical requirements, related to the type of processing carried out and the design of the process itself. The type of processing matters given that the product(s) must be stored, and this needs to be done safely and in a cost-effective way. As an example, storing toxic chemicals such as chlorine is tightly regulated (some jurisdictions cap the storage capacity for every unique location). This limits the potential number of chlor-alkali plants that are otherwise very strong candidates for providing flexibility.
In addition, excess production capacity (that is, beyond the nominal demand) must be available. This is a decision that is typically made at the design stage. Excess capacity may be added at a later time, but the cost (on a per-unit product basis) tends to be higher. Another important factor is process agility, that is, the speed with which the process can change its production rate. While conceptually similar to the ramp rates used to characterize generators, the notion of process agility is quite a bit more complex. Changing the production rate is subject to many constraints related to process safety, to maintaining the quality of the product(s), and to avoiding equipment fatigue. As such, the rate of change of the production rate is typically a nonlinear function of the operating point, and its value may be larger or smaller depending on what the starting point is and the direction of the change (up or down).
Incentives for Participation
The discussion of the second point—incentives for participation—starts from process agility as well. The characteristic time constants of manufacturing processes may be quite long, often in the order of hours. This suggests that chemical processes are best suited to provide flexibility for grid operations over rather long time scales and can be incentivized via price-based demand response schemes in the day-ahead market. Additionally, interruptible power can be provided (where the process or parts of it are completely shut down for a while), but this can be quite expensive, particularly when restarting the process takes a long time and/or has high costs.
Focus for the Future
These constraints notwithstanding, some manufacturing processes are well suited as flexibility providers. Continuous systems include aluminum smelting, air separation, chlor-alkali, and hydrogen production via electrolysis. A well-known example of batch process is steel manufacturing.
Many process operators are already participating in flexibility schemes. Some companies have entire departments dedicated to this engagement, which is regarded as an additional revenue stream. Others are not as far along in this journey and have proceeded in a more ad-hoc manner. Future fundamental (and applied) research must therefore focus on creating models and optimization methods that allow operators to confidently engage in flexibility schemes, along with market structures that provide an equitable distribution of the derived benefits.
McKetta Department of Chemical Engineering
Oden Institute for Computational Engineering and Sciences
The University of Texas at Austin