Smart Energy Scheduling With Hierarchical Real-Time Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current smart energy management systems face challenges in balancing power production and consumption in real-time, particularly with intermittent renewable energy sources, leading to sub-optimal decisions due to the lack of short-term feedback and predictions, which affects energy efficiency and increases grid power purchases.
Innovation Solution
A hierarchical optimization method that uses long-term demand and power generation forecasts to generate coarse-grained schedules, with a lower layer adjusting in real-time to short-term power generation profiles, allowing for fine-grained optimization and balancing of power imbalances between renewable energy sources and consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a two-layered optimization method with hourly MPC advisory layer and real-time controller is used, then power supply optimization is achieved, but the real-time controller decisions are sub-optimal due to lack of short-term feedback and predictions
Solution Approach 1:
The patent implements a feedback mechanism where the lower layer optimization results are fed back to update the upper layer optimization. The lower layer detects short-term imbalances and provides feedback signals that trigger re-optimization in the upper layer, ensuring that short-term predictions and feedback are incorporated into the overall optimization strategy, thereby improving decision optimality.
Solution Approach 2:
The patent segments the optimization process into two distinct layers: an upper layer that performs long-term planning with coarse temporal resolution, and a lower layer that handles short-term real-time control with fine temporal resolution. This segmentation allows each layer to specialize in different time scales, with the lower layer capturing short-term feedback that was previously lost, while the upper layer maintains overall optimization strategy.
2Stability of the object's composition
If mid-term energy storage is used to balance renewable energy variability, then power supply stability is improved, but overall energy efficiency decreases
Solution Approach 1:
The patent employs dynamic temporal resolution adjustment where the system transitions from coarse temporal resolution in the upper layer to fine temporal resolution in the lower layer. This dynamic adaptation allows the system to respond to short-term renewable energy variability without relying on mid-term energy storage, thereby maintaining power supply stability while reducing energy losses associated with storage and retrieval operations.
Solution Approach 2:
The upper layer performs preliminary optimization with coarse temporal resolution to establish long-term scheduling decisions. This preliminary action allows the system to anticipate and prepare for renewable energy variability in advance, reducing the need for reactive energy storage and improving overall energy efficiency by minimizing storage cycle operations.
3Device complexity
If coarse temporal resolution is used in optimization, then computational complexity is reduced, but real-time balancing capability is insufficient
Solution Approach 1:
The patent segments the temporal resolution into two levels: coarse resolution in the upper layer for long-term planning and fine resolution in the lower layer for short-term real-time control. This segmentation allows the system to maintain low computational complexity in the upper layer while achieving high-speed real-time balancing responses in the lower layer, effectively resolving the contradiction between computational complexity and real-time capability.
Solution Approach 2:
The patent introduces a vertical dimension to the optimization architecture by creating a two-layer hierarchical structure. This dimensional change allows the system to handle different time scales simultaneously, with the upper layer managing long-term trends and the lower layer managing short-term fluctuations, thereby achieving real-time balancing without proportionally increasing overall computational complexity.
Data Source
Figure 1~2
Figure 3~4
AI summary
A localized smart energy management system comprises a plurality of controllable loads (11), at least one intermittent energy source (13a-13b), a selectively connectable dispatchable energy source (14), and optionally but preferably an energy storage system (12). A method for balancing power production and power consumption of such localized smart energy management systems (10) in real time comprises performing a coarse-grained optimization in a first layer (103) of a hierarchical optimization structure to generate a predicted schedule, based on long-term load demand profiles (108) and long-term power generation profiles (109). A second layer (105) iteratively refines the predicted schedule upon receiving a new forecast of a short-term power generation profile (101) for the at least one intermittent energy source.