Single-Level DER Allocation Model for Multi-Aggregator Pricing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to effectively model and optimize the dynamic interactions between multiple aggregators and subscribers in a multi-aggregators multi-subscribers (MAMS) environment, leading to inefficiencies in energy resource allocation and profitability, particularly due to the lack of a comprehensive framework for volume and price optimization.
Innovation Solution
A non-linear single level optimization model is developed, incorporating a game-theoretic framework and Karush-Kuhn-Tucker (KKT) equivalents, to estimate distributed energy resource responses, optimizing energy resource allocation and trading strategies for both aggregators and subscribers, using a single model that integrates subscriber and aggregator layers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing predictive modeling techniques leverage historical data and machine learning algorithms for accurate forecast of energy production and consumption, then measurement precision of energy resource availability is improved, but the system fails to address multi-aggregator multi-subscriber scenarios with dynamic interactions between multiple aggregators and subscribers
Solution Approach 1:
The patent segments the complex MAMS environment into distinct layers: a subscriber layer that models individual subscriber responses to pricing, and an aggregator layer that models aggregator pricing strategies and resource allocation. This segmentation allows the system to handle multi-aggregator multi-subscriber dynamics by breaking down the interactions into manageable components while maintaining overall system accuracy.
Solution Approach 2:
The patent transitions from traditional single-level prediction models to a multi-level hierarchical model that incorporates both subscriber-level and aggregator-level dynamics. By adding this dimensional layering, the system can simultaneously capture accurate energy forecasts and the complex strategic interactions between multiple aggregators and subscribers that were previously unmodelable.
2Productivity
If multiple aggregators compete for the same pool of distributed energy resources to maximize aggregated energy volume and flexibility, then productivity of energy resource aggregation is improved, but the complexity of modeling and optimizing the competitive dynamics between aggregators and subscribers increases
Solution Approach 1:
The patent divides the competitive dynamics into two sequential optimization problems: first optimizing at the subscriber level to determine resource allocation responses, then optimizing at the aggregator level to determine pricing strategies. This segmentation reduces the overall complexity by solving simpler sub-problems rather than attempting to model all interactions simultaneously.
Solution Approach 2:
The patent employs a two-stage approach where subscriber responses are determined first (preliminary action), and then aggregator pricing strategies are optimized based on those responses. This preliminary determination of subscriber behavior simplifies the subsequent aggregator optimization problem, making the overall complex competitive dynamics more manageable.
3Reliability
If dynamic pricing strategies incorporating game-theoretic frameworks are employed by aggregators to maximize profits, then profitability of aggregators is improved, but the computational complexity and time required for real-time optimization increases
Solution Approach 1:
The patent segments the game-theoretic optimization into two independent stages: first solving the subscriber layer optimization to determine resource allocation responses, then solving the aggregator layer optimization to determine pricing strategies. This segmentation reduces computational time by breaking down the complex simultaneous game-theoretic problem into sequential simpler problems, while still capturing the strategic dynamics necessary for maximizing aggregator profits.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of the present disclosure herein provide a method and system of a non-linear single level optimization model for estimating DER responses in MAMS environment. Existing methods do not have a framework for volume allocation optimization and price-volume optimization especially in multi-aggregator multi-subscriber systems. Moreover, the price-volume optimization methods in existing literature consider a system having multiple subscribers and single aggregator only and using these methods it is difficult to assess how multiple subscribers will behave with multiple aggregators. Further, in a multi-aggregator system, determining the prices offered by each for an amount of energy resource allocated by the subscribers is another challenge, since the prices offered are dependent on several factors such as day ahead market prices, market risks and the like. The disclosed non-linear single level optimization model enables to estimate equilibrium optimal values between an amount of energy resources allocated for each aggregator among the one or more aggregators and the amount of energy resource traded by each aggregator in one or more energy markets.