Optimal Uniform Pricing for Multi-Interval Dispatch without Make-Whole Uplifts

summary

The gist

Optimal uniform pricing rule that minimizes demand payments, eliminates OOM make-whole payments, preserves LMP-based congestion charges and ensures revenue adequacy In a network with ramp-limited…

In short

The research developed LMP+U, an optimal uniform pricing rule designed to minimize demand payments while eliminating out-of-market (OOM) make-whole payments. This rule preserves congestion charges and ensures revenue adequacy, even when generator ramp constraints cause market prices to fall below their bids. It means a single uniform price adjustment across all locations is the most efficient way to manage dispatch incentives.

Key concepts

LMP+U
This is the proposed optimal pricing rule. It sets the price at LMP plus a uniform adder across all locations. The goal is to minimize demand payments and remove the need for OOM make-whole payments, which are often unfair or opaque.
OOM Make-Whole Payments
These are extra compensation payments made to generators outside the main market settlement when dispatch prices (LMPs) are lower than what generators expect. The paper aims to eliminate these because they create non-transparent signals and discriminatory incentives for bidding.
Congestion Charges
These charges are based on the Locational Marginal Price (LMP). The proposed rule ensures that these essential LMP-based congestion charges remain preserved, meaning the pricing structure still accurately reflects network constraints like transmission limitations.

Terminology used across episodes

This episode discusses

The paper

Optimal Uniform Pricing for Multi-Interval Dispatch without Make-Whole Uplifts · Read on arXiv

Cornell University · Dartmouth College

Transcript

Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.

Rosa: I'm Rosa.

Dev: And I'm Dev, here with Taro.

Rosa: We're looking at the paper: "Optimal Uniform Pricing for Multi-Interval Dispatch without Make-Whole Uplifts".

Paper summary: Rosa: So we're looking at this paper called Optimal Uniform Pricing for Multi-Interval Dispatch without Make-Whole Uplifts. It deals with a situation where generators have ramp limits and demand forecasts are not perfect, which can cause the locational marginal prices to drop below what the generators are willing to accept in market.

Dev: That's the core problem, right? When that happens, you need these out-of-market make-whole payments to keep things running smoothly for the generators. But those payments have some real headaches with transparency and fairness.

Rosa: Exactly what this paper does is propose an optimal uniform pricing rule called LMP plus U. The big claim here is that this rule minimizes the total demand payment while getting rid of those out-of-market make-whole payments and keeping the LMP based congestion charges intact.

Dev: So, it's not just a tweak to how you calculate prices; it’s a whole new way of pricing that relates back to existing schemes. The paper claims this rule is derived in closed form and takes the form of the LMP plus a uniform price adder across all locations.

Taro: From my side, what interests me is the incentive analysis part. They claim that under this new LMP plus U rule, a generator who is just taking prices won't have any reason to change their bids without that out-of-market uplift.

Rosa: That’s interesting because it suggests a generator doesn't have an incentive to lie or inflate its offer if the system is set up with this uniform price adder. It means they are bidding truthfully within the tested range.

Dev: And that’s important for real-time loops because it simplifies things; you don't have to worry about generators manipulating bids just to get that extra compensation. The paper also shows that this approach can reduce price volatility by up to twenty-nine percent in some cases.

Taro: A twenty-nine percent reduction in volatility is a solid number when you're dealing with real-time dispatch, especially when things are running on tight constraints and inaccurate forecasts. It suggests a more stable system operation under this pricing structure.

Rosa: So, to wrap up this part of the paper, they show that LMP plus U generally gives higher aggregate generator net revenue compared to the old LMP with make-whole payments because that common price adjustment boosts every active generator's profit.

Dev: But we have to remember the trade-off, which is that those higher generator profits come at a cost in terms of increased demand payments, reflecting this uniform allocation of ramping costs in the market.

Taro: So, what this means for the world is that we might be able to manage these complex network constraints and forecasting errors more efficiently without needing those potentially messy out-of-market settlements that complicate real-time signals.

Rosa: Right. And that brings us to where they are in the paper's title and authors. The paper is Optimal Uniform Pricing for Multi-Interval Dispatch without Make-Whole Uplifts, written by Valentina Norambuena Guzman, Cong Chen, Lang Tong, and Timothy D Mount.

Dev: It’s interesting how these authors approach it; they take something that already exists in the pricing structure—the LMP—and modify it systematically to solve a specific operational problem involving ramp limits and forecasting errors.

Rosa: They are essentially saying that if you want to keep the congestion charges working correctly, you have to adjust the base price in a very specific, uniform way across all locations.

Dev: I'm curious about the practical implications for someone running these real-time control loops; how does this uniform adder translate into something we can implement reliably on our systems?

Taro: From an autonomy researcher standpoint, it’s about graceful recovery; the system doesn't break down because of a temporary forecast error or a ramp limit issue. This paper suggests a pricing mechanism that supports that stability.

Rosa: So, to bring this segment to a close, this work suggests that by using LMP plus U, we can maintain the integrity of congestion charges while smoothing out the incentive structure for generators in challenging dispatch scenarios.

Dev: We’ve covered the core thesis and what it promises regarding revenue adequacy and price stability. Next up, we'll look at what these results actually mean for how we manage power systems in practice.

Conclusion: Rosa: So we've been looking at how this new pricing rule works for multi-interval dispatch without those extra make-whole payments.

Dev: Yeah, the paper is called "Optimal Uniform Pricing for Multi-Interval Dispatch without Make-Whole Uplifts." It’s by Valentina Norambuena Guzman and her team.

Rosa: They found a way to minimize how much demand you have to pay while completely cutting out those out-of-market make-whole adjustments and keeping the LMP congestion charges.

Dev: The main idea is that instead of paying those extra bits when things go wrong, you just add one uniform price across every location. They call it LMP plus U.

Rosa: It turns out this rule is derived in a closed form, which means they figured out the exact math for the optimal price structure.

Dev: That’s cool because it shows that you can relate this new pricing back to what we already use in our system, just with a little adjustment.

Rosa: What’s really interesting is how they show that a generator who just takes prices doesn't have an incentive to change their bid if you use this LMP plus U setup.

Dev: That’s good for real-time control because it means generators aren't trying to inflate their offers just to get that extra compensation.

Rosa: They actually found that this approach can reduce price volatility by up to twenty-nine percent in some scenarios.

Dev: A twenty-nine percent reduction is a solid number when you’re dealing with real-time dispatch, especially when things are running on tight constraints and inaccurate forecasts.

Rosa: But it's important to remember that this uniform addition means the higher demand payments reflect the actual ramping costs in the market.

Dev: Exactly, so you’re paying for the physical cost of ramping across all locations, which is different from those random make-whole payments.

Rosa: It preserves those LMP properties they're used for congestion charges and keeps that non-negative merchandising surplus as long as there are real-time market obligations to cover it.

Dev: The paper shows this works even with ramp-limited generators and when the net demand forecasts aren't perfectly accurate.

Rosa: It suggests a more stable way to manage these complex network constraints without needing those complicated out-of-market settlements that complicate real-time signals.

Dev: It’s about maintaining the integrity of congestion charges while smoothing out the incentive structure for generators in challenging dispatch situations.

Rosa: Next up, we’ll look at what this actually means for how we manage power systems in practice.

More episodes

← Home