Neural Network Control for Grid-Connected Converter Stability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Grid-connected converters (GCCs) with LCL-filters face instability and control challenges due to their third-order nature, making decoupling of d- and q-axis control loops difficult, and conventional damping strategies fail to effectively address these issues.
Innovation Solution
A neural network-based control system is implemented, which includes a nested-loop controller with an inner loop featuring a neural network that optimizes dq-control voltages, accounting for resonant circuit dynamics of the grid filter, and optionally uses dynamic programming and Levenberg-Marquardt algorithms for training, eliminating the need for passive or active damping structures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-generated harmful factors
If LCL-filter is used in GCC systems, then harmonic attenuation capability is improved and cost is reduced, but system stability deteriorates and control difficulty increases
Solution Approach 1:
The patent implements a feedback-based control system that continuously monitors the GCC output and adjusts control parameters to maintain stability. The control system uses feedback signals from the grid-connected converter to dynamically compensate for the oscillatory tendencies introduced by the LCL-filter, thereby resolving the stability issue while preserving the filter's harmonic attenuation benefits
Solution Approach 2:
The patent employs parameter optimization techniques to adjust critical system parameters such as damping factors and control gains. By carefully selecting and tuning these parameters, the system achieves stable operation with LCL-filter while maintaining effective harmonic suppression, thus resolving the contradiction between stability and harmonic attenuation
2Object-generated harmful factors
If LCL-filter is used in GCC systems, then harmonic attenuation capability is improved, but decoupling of d- and q-axis control loops becomes difficult
Solution Approach 1:
The patent divides the control system into separate d-axis and q-axis control loops with distinct control strategies. By segmenting the control architecture and implementing independent control for each axis with appropriate decoupling terms, the system manages the complexity of LCL-filter control while maintaining effective harmonic attenuation
Solution Approach 2:
The patent introduces decoupling terms as intermediary elements between the d-axis and q-axis control loops. These intermediary components compensate for the coupling effects introduced by the LCL-filter, allowing independent control of active and reactive power while preserving the filter's harmonic suppression capabilities
3Device complexity
If capacitor dynamics are neglected to simplify control, then vector control problem is simplified to first-order L-GCC system, but description precision deteriorates and oscillatory behavior increases
Solution Approach 1:
The patent applies partial action by selectively modeling only the most critical dynamic components of the LCL-filter while neglecting less significant ones. This approach maintains adequate system description precision for control purposes without requiring the full complexity of the third-order system, thus resolving the contradiction between simplicity and accuracy
Data Source
AI summary
An example system for controlling a grid-connected energy source using a neural network is described herein. The example system can include a grid-connected converter (“GCC”) operably coupled between an electrical grid and an energy source, a n-order grid filter (e.g., where n is an integer greater than or equal to 2) operably coupled between the electrical grid and the GCC, and a nested-loop controller. The nested-loop controller can have inner and outer control loops and can be operably coupled to the GCC. A d-axis loop can control real power, and a q-axis loop can control reactive power. Additionally, the inner control loop can include a neural network that is configured to optimize dq-control voltages for controlling the GCC. The neural network can account for circuit dynamics of the n-order grid filter while optimizing the dq-control voltages.