Distributed Reactive Voltage Control for Fast Grid Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional centralized optimization methods for reactive voltage control in power grids face challenges such as single-node failure, high communication and computing burdens, data privacy issues, and inability to efficiently manage high-speed resources in high-penetration power grids.
Innovation Solution
A fully-distributed reactive voltage control method that partitions the power grid into areas, uses a robust recursive regression algorithm to solve linear regression models, and employs the gradient projection and alternating direction multiplier algorithms for optimal control, allowing for fast and model-free reactive voltage control without requiring accurate system model parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If centralized optimization method is used for reactive voltage control, then voltage quality can be improved and network loss reduced, but the system faces single-node failure risk, high communication burden, high computing burden, and data privacy issues
Solution Approach 1:
The patent divides the centralized power grid system into multiple distributed control areas, each with its own control unit. The global optimization problem is segmented into local sub-problems that can be solved independently using local measurements and models, eliminating the need for a single centralized controller and reducing communication burden to only boundary variable exchanges between areas.
Solution Approach 2:
The patent introduces boundary variables as intermediaries between different control areas. These boundary variables (voltages and power flows at interface nodes) serve as the only communication channel between areas, allowing each area to maintain independence while still achieving coordinated global optimization through iterative exchange of boundary information.
2Ease of operation
If centralized optimization method is used, then reactive voltage control can be achieved, but computing burden and communication requirements increase significantly
Solution Approach 1:
The computational task is segmented from a single centralized computation into multiple parallel distributed computations across different areas. Each area solves its own local optimization problem using only local measurements and a linearized power flow model, dramatically reducing the computing time and complexity at each node while maintaining overall system optimization.
Solution Approach 2:
The patent performs preliminary linearization of the power flow equations to create a simplified model that can be solved quickly in real-time. By pre-processing the complex nonlinear power flow constraints into linear relationships, the system enables fast distributed optimization without requiring heavy computational resources during actual control operations.
3Adaptability or versatility
If centralized method is used, then optimization control can be implemented, but data privacy of stakeholders cannot be protected
Solution Approach 1:
The patent segments the power grid into independent control areas where each stakeholder retains control over their local data and resources. Only anonymized boundary variables (voltages and power flows at interface nodes) are exchanged between areas, allowing optimization control to be implemented while protecting the private operational data of each stakeholder from being accessed by others.
4Measurement precision
If traditional power flow equations are used in optimization models, then accuracy is maintained, but computational complexity increases and real-time control becomes difficult
Solution Approach 1:
The patent performs preliminary linearization of the nonlinear power flow equations around the operating point to create simplified linear relationships between voltages, powers, and reactive power injections. This pre-processing step maintains sufficient accuracy for voltage control applications while enabling real-time distributed optimization without the computational burden of solving full nonlinear power flow equations iteratively.
Data Source
AI summary
A fully-distributed reactive voltage control method, includes: establishing a power grid reactive voltage optimization model of a power grid; parting the power grid reactive voltage optimization model into a plurality of area reactive voltage optimization models of a plurality of areas of the power grid; converting a power flow equation constraint in each of the plurality of area reactive voltage optimization models to a linear regression model; solving the linear regression model by using a robust recursive regression algorithm to obtain a solution result of the linear regression model; solving each of the plurality of area reactive voltage optimization models by using the solution result of the linear regression model, a gradient projection algorithm, and an alternating direction multiplier algorithm, so as to realize a reactive voltage optimization control of each of the plurality of areas.