Energy Bidding Apparatus Using Reinforcement Learning and Linear Programming
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current energy bidding methods for renewable energy sources face challenges in accurately determining electricity prices due to unstable power generation, requiring assumptions about profit models and leading to privacy leakage issues, and often rely on limited quadratic function assumptions for convergence.
Innovation Solution
A reinforcement learning based method and apparatus that uses linear programming to optimize energy supply configurations between multiple energy suppliers and demanders by calculating total demand and supply amounts, and employing reinforcement learning tables to determine optimal electricity purchase and sale quotations, ensuring privacy and maximizing profit.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If iterative calculations with quadratic function assumptions are used to negotiate prices, then convergence can be ensured, but the profit model assumptions are difficult to achieve in practice and privacy leakage occurs
Solution Approach 1:
The patent changes the mathematical approach from quadratic function assumptions to linear programming formulation. This parameter change in the optimization method allows the system to handle general profit models without requiring specific functional form assumptions, thereby improving adaptability while maintaining convergence through the guaranteed optimality of linear programming solutions.
Solution Approach 2:
The patent introduces an energy aggregator as an intermediary that collects supply and demand information, formulates the pricing problem as a linear programming problem, and determines optimal prices. This intermediary structure allows the system to handle complex profit models privately without requiring direct sharing of sensitive profit function information between buyers and sellers.
2Measurement precision
If iterative calculations are used to negotiate prices, then a balance point can be obtained, but privacy leakage of profit models occurs
Solution Approach 1:
The energy aggregator acts as a trusted intermediary that receives only supply amounts and demand amounts from participants, not their private profit models. The aggregator formulates and solves the linear programming problem internally, then returns only the optimal price and quantity allocations. This preserves participant privacy while achieving precise price determination.
Solution Approach 2:
Instead of requiring participants to share their actual profit models, the system uses the linear programming formulation with objective function coefficients representing the aggregator's optimization criteria. This copying approach replaces the need for private information sharing with a public optimization framework that achieves the same pricing objective.
3Productivity
If reinforcement learning is used to determine quotations, then profit optimization is achieved, but system complexity increases
Solution Approach 1:
The patent replaces complex iterative negotiation mechanisms with a direct linear programming formulation. By substituting the mechanical iterative process with a mathematical optimization program, the system achieves profit optimization through a more straightforward computational approach, reducing the complexity of the bidding system architecture while maintaining or improving efficiency.
Data Source
AI summary
A method and an apparatus for reinforcement learning based energy bidding, adapted for an energy aggregator to determine the energy supply configuration between multiple energy suppliers and multiple energy demanders, are provided. In the method, a supply amount of each energy supplier and a demand amount of each energy demander are acquired. A total demand amount of the energy demanders is calculated and replied to each energy supplier, and a total supply amount of the energy suppliers is calculated and replied to each energy demander. An electricity purchase quotation determined by each energy demander according to respective demand amount and the total supply amount, and an electricity sale quotation determined by each energy supplier according to respective supply amount and the total demand amount are received. A linear programming method is adopted to determine the energy supply configuration between the energy suppliers and the energy demanders according to information.