Distributed Operation Planning System for Power Grids
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current operation planning systems in multi-unit settings, such as electric power systems and manufacturing, face challenges in efficiently generating and updating operation plans due to centralized management, which leads to delays and increased costs when changes occur, making it difficult to adapt to changes in demand or unit failures.
Innovation Solution
A distributed operation planning system where planning units and operation units are paired, allowing for decentralized generation and adjustment of operation plans, using genetic algorithms, reinforcement learning, and greedy algorithms to select unit-groups that meet output requirements while minimizing costs and adhering to restrictions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If operation plan generation is concentrated in the central management server, then centralized control is achieved, but the system response time increases and scalability decreases
Solution Approach 1:
The patent segments the operation plan generation function from the central management server and distributes it to multiple planning units. Each planning unit independently generates operation plans for specific operation units, eliminating the bottleneck of centralized plan generation and reducing system response time while maintaining coordinated control through shared demand information.
Solution Approach 2:
The patent introduces a new dimensional structure by pairing planning units with operation units in a many-to-many relationship matrix. This dimensional transformation allows parallel plan generation across multiple planning units simultaneously, while the pairing structure maintains systematic coordination without centralized sequencing.
2Stability of the object's composition
If operation plan generation is concentrated in the central management server, then centralized control is achieved, but the system scalability is limited
Solution Approach 1:
The patent segments the monolithic centralized planning function into multiple independent planning units that can be dynamically added or removed. This segmentation enables the system to scale by simply adding more planning units without redesigning the entire centralized architecture, thereby improving adaptability while maintaining controlled operation through the pairing mechanism.
Solution Approach 2:
The patent creates universal planning units that can handle multiple operation units through the pairing structure. Each planning unit serves multiple operation units, and each operation unit can be served by multiple planning units, providing flexibility and scalability without requiring dedicated planners for each unit.
3Loss of information
If all information for each unit is collected before generating operation plans, then comprehensive planning is achieved, but the planning process becomes time-consuming
Solution Approach 1:
The patent segments the information processing load by distributing demand information to multiple planning units simultaneously. Each planning unit receives and processes the same demand information independently, eliminating the sequential information collection bottleneck while maintaining comprehensive information availability for all planning decisions.
Solution Approach 2:
The patent performs preliminary distribution of demand information to all planning units before the actual operation plan generation begins. This preliminary action ensures all planning units have the necessary information ready, enabling parallel processing and reducing the overall planning time while maintaining information completeness.
4Stability of the object's composition
If the system operates according to an old plan, then operational stability is maintained, but operation costs increase when unit behaviors change
Solution Approach 1:
The patent implements feedback mechanisms where planning units continuously monitor changes in unit behaviors and demand information. When changes are detected, the system automatically triggers regeneration of operation plans, ensuring cost-optimality is maintained while operational stability is preserved through systematic re-planning rather than ad-hoc adjustments.
Data Source
AI summary
According to one embodiment, an operation planning system includes a demand information database and a plurality of planning units. The demand information database stores demand data including an overall output required for a plurality of operation units for a planning period in an operation plan. The plurality of planning units adjusts output settings of the plurality of operation units. Any of the planning units selects a combination of operation units which provide the overall output required for the planning period, and based on operation costs depending on the outputs of the operation units, a process is repeated to generate the operation plan, the process including: selection of the operation unit belonging to the combination of operation units; and adjustment of the output setting for the selected operation unit by the planning unit paired with the selected operation unit.