Modular Multilevel Converter Control via Signal Model Extrapolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for controlling modular multilevel converters are limited in their ability to adjust the update rate of control signals independently of the input data stream, leading to suboptimal performance and potential noise increases when trying to enhance the update rate.
Innovation Solution
A method that generates a compressed output data stream with a higher temporal update rate than the input data stream by inserting additional setpoint values using extrapolation or prediction methods, allowing for independent control of the update rate and reducing noise components, particularly by determining model parameters from the input data stream and using them to form a signal model that describes the fundamental and harmonic components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the update rate of control signals is increased to improve control precision, then control performance is improved, but noise components increase undesirably
Solution Approach 1:
The patent applies parameter changes by transforming the input data stream parameters through a signal model that separates fundamental and harmonic components. The model parameters (amplitude, frequency, phase) are adjusted to generate output setpoints at a higher update rate while maintaining signal quality and avoiding noise amplification.
Solution Approach 2:
The patent uses preliminary action by pre-determining model parameters from the input data stream and storing them in a signal model before generating the output data stream. This preliminary modeling allows the system to predict future setpoints at higher update rates without introducing noise, as the model captures the essential signal characteristics.
2Speed
If digital-to-digital conversion is used to increase data rate, then update rate is improved, but device complexity increases
Solution Approach 1:
The patent replaces the mechanical/digital conversion approach with a mathematical signal modeling approach. Instead of using digital-to-digital converters to increase data rate, the system uses a signal model with predetermined parameters to generate output setpoints at the desired update rate, thereby avoiding additional conversion hardware and reducing device complexity.
3Productivity
If additional setpoint values are inserted to increase update rate, then control responsiveness is improved, but processing complexity increases
Solution Approach 1:
The patent uses copying by generating additional output setpoint values based on the predetermined signal model parameters rather than directly processing additional input data. The model parameters are copied and applied to generate multiple output setpoints at the higher update rate, reducing processing complexity compared to direct interpolation or prediction methods.
Data Source
Figure 1
Figure 2~4
Figure 5
AI summary
The invention relates, inter alia, to a method for operating an electrical device (20) taking into account an input data stream (Se) which is formed by input desired values (Ue) and the input desired values (Ue) of which have a temporal update rate (f1). The invention provides for the input data stream (Se) to be used to form a compressed output data stream (Sa), the temporal update rate (f2) of which for output desired values (Ua) is greater than the temporal update rate (f1) for the input data stream (Se), and components of the electrical device (20) are controlled on the basis of the output desired values (Ua) of the output data stream (Sa).