Microgrid Controller Using Dual-Time-Scale Predictive Power Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing microgrid controllers face challenges in accurately managing power distribution due to the need to account for both long-term and short-term power dynamics, leading to less efficient models that fail to consider discrete and nonlinear system constraints effectively.
Innovation Solution
A microgrid controller with a two-layer predictive control architecture, utilizing stochastic and nonlinear programming, which determines optimal power characteristic levels at different time intervals using a processor and database to manage power distribution efficiently across the microgrid, including energy storage devices and generators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single-time-scale control model is used, then the model complexity is reduced, but the accuracy of power management deteriorates because it cannot simultaneously capture long-term periodic patterns and short-term rapid dynamics
Solution Approach 1:
The control model is segmented into two distinct time scales: a long-term model capturing periodic patterns (e.g., daily load cycles) and a short-term model capturing rapid dynamics (e.g., solar PV fluctuations). Each model operates at its appropriate time scale, avoiding the need for a single overly complex model while maintaining comprehensive accuracy.
Solution Approach 2:
The solution introduces a time-scale dimension by operating controllers at different frequencies. The first controller operates at a lower frequency for long-term optimization, while the second controller operates at a higher frequency for short-term responsiveness. This dimensional approach allows simultaneous capture of multiple dynamics without exponential complexity increase.
2Measurement precision
If multiple controllers operating at different time scales are implemented, then the power management accuracy is improved, but the system complexity increases
Solution Approach 1:
Multiple controllers operating at different time scales are merged into a unified hierarchical architecture. The first controller (long-term) and second controller (short-term) are combined such that their control actions are coordinated through a common optimization framework, reducing overall system complexity while maintaining the benefits of multi-scale operation.
Solution Approach 2:
The control system is designed with universal components that can operate at multiple time scales. The optimization engine and constraint management mechanisms are multi-functional, handling both long-term periodic patterns and short-term rapid dynamics through a unified mathematical framework, thereby reducing the need for separate specialized subsystems.
3Speed
If fast-response control is implemented for solar PV and wind turbines, then the responsiveness to rapid power dynamics is improved, but the ability to account for long-term utility rates and load patterns deteriorates
Solution Approach 1:
The control system is segmented into two functional layers: a fast-response layer that handles rapid power dynamics of solar PV and wind turbines, and a slow-response layer that optimizes for long-term utility rates and load patterns. Each layer operates independently at its appropriate time scale, allowing fast response without sacrificing long-term optimization capability.
Solution Approach 2:
The solution addresses the time horizon conflict by introducing a time-scale dimension. The first controller operates over a longer time horizon to capture utility rate structures and load patterns, while the second controller operates over a shorter time horizon to respond to rapid renewable generation fluctuations. This dimensional separation allows both objectives to be simultaneously achieved.
Data Source
AI summary
A microgrid controller includes a database in communication with a processor. The processor is operable to receive at least one microgrid input, to determine a first plurality of optimal power characteristic levels at a corresponding one of a plurality of first time intervals for a first time period, and to determine a second plurality of optimal power characteristic levels of a device determined at a corresponding one of a plurality of second time intervals for a second time period. The first time intervals are found at a first frequency different than a second frequency of the second time intervals. One of the second plurality of optimal power characteristic level corresponds to one of the first plurality of optimal power characteristic levels at each first time interval. The processor is configured to control a device optimal power characteristic level in response to the second plurality of optimal power characteristic levels.