Electrical Converter Pulse Pattern Control for NP Balance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing control schemes for the 3L(A)NPC+AF topology struggle to maintain balanced Neutral Point (NP) potential, leading to reduced output voltage quality, especially at zero power factor and low speeds, due to reliance on tolerance bands and repetitive drifting.
Innovation Solution
A linear NP potential control method compatible with OPP-MP3C, which adjusts optimized pulse patterns online to balance AF cell voltages and NP potential without tolerance bands, ensuring equilibrium across all operating conditions, including zero power factor.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If legacy NP control approaches with tolerance bands are used, then the control scheme is compatible with MP3C, but the NP potential drifts repeatedly and output voltage quality is reduced
Solution Approach 1:
The patent implements a feedback mechanism where the neutral point potential is continuously monitored and used to adjust the switching instants of the optimized pulse pattern. The controller calculates the deviation from the desired neutral point potential and modifies subsequent switching events to correct this deviation, ensuring continuous balancing without tolerance bands and without compromising output voltage quality
Solution Approach 2:
The patent makes the switching instants dynamic by allowing online adjustment of the pre-computed optimized pulse patterns. Instead of using fixed switching patterns with tolerance bands, the controller dynamically shifts switching instants based on real-time neutral point potential measurements, enabling adaptive balancing across all operating conditions including zero power factor
2Reliability
If NP balancing is implemented with redundant voltage vectors, then the NP potential can be controlled, but the output voltage quality is reduced due to repetitive drifting
Solution Approach 1:
The patent pre-computes optimized pulse patterns offline that are specifically designed to balance the neutral point potential while maintaining high output voltage quality. These pre-optimized patterns eliminate the need for real-time redundant vector selection and tolerance band management, providing both reliable NP control and superior voltage quality from the outset
Solution Approach 2:
The patent changes the approach from discrete redundant vector selection to continuous switching instant adjustment. By modifying the timing parameters of the optimized pulse patterns online based on neutral point potential feedback, the system achieves reliable NP control without the quality degradation associated with repetitive drifting in traditional methods
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A method for controlling an electrical converter (10) is provided. The electrical converter (10) comprises a main stage (12) adapted for converting a DC voltage into an intermediate voltage having at least two voltage levels and a filter cell stage (14) with a filter cell (30) for each phase of the main stage (12), with the main stage (12) comprising a DC link (20) and with each filter cell (30) being adapted for adding or subtracting a cell voltage of the filter cell (30) to the intermediate voltage. The method comprises the steps of: determining a first optimized pulse pattern (OPPMC,abc) for the main stage (12) and a second optimized pulse pattern (OPPAF,abc) for the filter cells (30) from an offline-computed optimized pulse pattern (OPPbase), wherein the first optimized pulse pattern (OPPMC,abc) comprises switching instants for the main stage (12) over a next computation window (αwin) and wherein the second optimized pulse pattern (OPPAF,abc) comprises switching instants for filter cells (30) over the next computation window (αwin); determining a neutral point potential (VNP) depending on a measured lower DC link voltage (Vbot) and a measured upper DC link voltage (Vtop) of the DC link (20); determining a first balancing signal (UNPbal,abc) for the main stage (12) depending on the determined neutral point potential (VNP); determining a second balancing signal (UAFbal,abc) for the filter cells (30) depending on measured cell voltages (VAFC,abc) of the filter cells (30); determining a first differential mode voltage (UNPbal,DM,abc) from the first balancing signal (UNPbal,abc); determining a second differential mode voltage (UAFbal,DM,abc) from the second balancing signal (UAFbal,abc); determining a first voltage change (ΔUMC,abc) from the first balancing signal (UNPbal,abc) and the second differential mode voltage (UAFbal,DM,abc); determining a second voltage change (ΔUAF,abc) from the second balancing signal (UAFbal,abc) and the first differential mode voltage (UNPbal,DM,abc); adjusting the first optimized pulse pattern (OPPMC,abc) by moving its switching instants depending on the first voltage change (ΔUMC,abc) over the next computation window (αwin) and adjusting the second optimized pulse pattern (OPPAF,abc) by moving its switching instants depending on the second voltage change (ΔUAF,abc) over the next computation window (αwin); determining a reference flux trajectory (Ψref) over the next computation window (αwin) depending on the adjusted first and second optimized pulse pattern (OPPMC,abc, OPPAF,abc); and controlling the electrical converter (10) depending on the reference flux trajectory (Ψref) and on the adjusted first and second optimized pulse pattern (OPPMC,abc, OPPAF,abc).