DER Extremum Seeking Control with Decaying Dither Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Optimization-based approaches for managing Distributed Energy Resources (DER) in electric distribution grids require knowledge of the underlying distribution system topology and real-time load information, which is not always available, leading to inefficiencies and potential grid reliability issues as the number of DERs, such as solar PV systems, increases.
Innovation Solution
A 2-dimensional Extremum Seeking (2D-ES) control system with an exponentially decaying probing signal is used to manage DER active and reactive power contributions, allowing for substation real and reactive power target tracking without requiring exogenous information about the grid, using an equilibrium-based switching criterion to optimize power contributions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If optimization-based approaches are used to manage DER, then DER participation in grid operation is improved, but the requirement for knowledge of distribution system topology and real-time load information increases
Solution Approach 1:
The extremum seeking control algorithm enables DER to autonomously optimize its own power contribution by locally searching for the optimal operating point that maximizes grid utility, without requiring external information about system topology or real-time load conditions. The DER essentially serves itself by performing local optimization based on feedback signals from the grid.
Solution Approach 2:
The patent introduces an intermediary signaling mechanism where the grid provides feedback signals to DER through the power flow, and DER responds by adjusting its power contribution. This intermediary approach allows optimization without direct information exchange about system state, using the power flow itself as the communication medium.
2Measurement precision
If persistent probing signals are used in extremum seeking control, then optimal power contribution is achieved, but system losses increase and opportunity costs are incurred
Solution Approach 1:
The patent employs periodic probing signals (dither signals) that are superimposed on the DER power output to enable the extremum seeking algorithm to detect the slope of the utility function and converge to the optimal operating point. The periodic nature of these signals allows for continuous optimization while maintaining system stability.
Solution Approach 2:
The probing signal amplitude is made dynamic rather than constant, allowing the system to adapt the intensity of exploration based on convergence status. This dynamic adjustment reduces energy losses by decreasing probing amplitude once optimal conditions are approached, while still maintaining the ability to track changing optima.
3Adaptability or versatility
If multiple DERs are deployed in the distribution grid, then grid reliability is enhanced, but adverse effects on grid reliability and resiliency increase due to lack of coordination
Solution Approach 1:
The extremum seeking control system implements feedback mechanisms where DER continuously monitor the effect of their power contributions on grid utility and adjust their output accordingly. This feedback-driven approach ensures that multiple DERs coordinate their actions implicitly through the shared grid environment, preventing adverse interactions while maintaining deployment flexibility.
Data Source
AI summary
Embodiments relate to an operating system coupled to and controlling at least one Distributed Energy Resource (DER) in a smart grid and includes an electricity distribution network and a 2-dimensional Extremum Seeking (2D-ES) controller coupled to network. The electricity distribution network has first and second inputs and one output and maps all active power and reactive power inputs to the smart grid providing measurements of an objective function. The 2D-ES controller controls the DER and includes an active power loop and a reactive power loop. The active power loop is in communication with the first input and the output, governs active power contribution, and receives measurements of the objective function which contain static and oscillatory components. The reactive power loop is in communication with the second input and the output, governs reactive power contribution, and receives measurements of the objective function which contain static and oscillatory components.