Modular Multi-Level Converter Control via Predictive Communication Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modular multi-level converters face challenges in control due to the limitations of existing control systems, particularly under unreliable communication conditions which can lead to delays, packet losses, and bandwidth limitations, degrading control performance and potentially destabilizing the system.
Innovation Solution
A method and controller that utilize model predictive control, accounting for communication delays, data losses, and quantization by incorporating models of both the converter and communication media, allowing for remote control and compensating for these issues through predictive techniques to ensure accurate and flexible control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If remote control is implemented via communication media, then controller flexibility and computational capability are improved, but communication delays, packet losses, and bandwidth limitations degrade control performance and system stability
Solution Approach 1:
The controller predicts future states of the converter system based on current and past data, allowing it to prepare control actions in advance that compensate for anticipated communication delays. This predictive approach enables the system to act proactively rather than reactively, maintaining control performance despite time-varying communication latencies.
Solution Approach 2:
The system continuously monitors actual converter states and compares them with predicted states, using this feedback to refine predictions and adjust control actions. This closed-loop feedback mechanism compensates for packet losses and communication uncertainties, ensuring reliable control performance while maintaining remote operation flexibility.
2Measurement precision
If model predictive control is used to compensate for communication issues, then control accuracy under unreliable conditions is improved, but computational load increases
Solution Approach 1:
The controller dynamically adjusts prediction horizons, update frequencies, and model complexity based on current communication conditions and system requirements. This adaptive parameter tuning allows the system to maintain high control accuracy when communication is reliable while reducing computational load during periods of poor communication or limited resources.
3Weight of stationary object
If the controller is located remotely from the converter, then weight and space requirements at the converter site are reduced, but instantaneous and perfect control signals cannot be achieved
Solution Approach 1:
The remote controller predicts future converter states and prepares control signals in advance, compensating for transmission delays. By acting proactively rather than reactively, the system eliminates the detrimental effects of time-varying communication latencies while maintaining the benefits of remote controller placement.
Data Source
AI summary
A method for controlling a modular multi-level converter comprises the steps of: collecting control input variables from the converter; transmitting the control input variables to a controller of the converter via a first communication medium; determining, in the controller, and actual state of the converter and at least one control output variable based on a model of the converter; and transmitting the control output variable to the converter for controlling the converter via a second communication medium. The model of the converter accounts for the first and/or the second communication medium.