Multi-Converter Pulse Pattern Control for Harmonics and Transients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for controlling electrical converters fail to efficiently manage power delivery, with the harmonic requirements, especially in medium-voltage systems, where high pulse numbers are not an option due to high switching losses, and traditional linear controllers are sluggish during transients.
Innovation Solution
A model predictive pulse pattern control (MP3C) method is used to control electrical converters, which estimates the effect of pulse patterns on converter output currents, adjusts switching times to compensate for ripple currents, and uses a fast closed-loop control mechanism to manage harmonic distortions and transients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If carrier-based pulse-width modulation (CB-PWM) with linear control is used, then control simplicity is maintained, but switching frequency must be increased to meet harmonic requirements, resulting in high switching losses
Solution Approach 1:
The patent changes the control parameter from continuous linear control to discrete optimized pulse patterns with specific switching sequences. OPPs use offline-calculated switching sequences that yield optimal harmonic distortions for a given pulse number, allowing low switching frequency operation while meeting harmonic standards.
Solution Approach 2:
The patent pre-calculates optimized pulse patterns offline before operation. OPPs are offline-calculated switching sequences that are stored and then applied during operation, eliminating the need for real-time calculation and allowing the system to benefit from pre-optimized harmonic performance at low switching frequencies.
2Loss of energy
If optimized pulse patterns (OPPs) are used to reduce switching losses, then switching frequency is reduced, but the controller becomes sluggish during transients due to slow response
Solution Approach 1:
The patent implements a model predictive control mechanism that uses feedback from the actual system state to adjust the pulse patterns in real-time. The controller predicts future states and adjusts switching instances to achieve desired flux trajectories, providing fast transient response while maintaining the low switching frequency benefits of OPPs.
Solution Approach 2:
The patent makes the pulse patterns dynamic by allowing real-time adjustment of switching instances based on system conditions. While the base pulse patterns are fixed and optimized offline, the actual switching times are dynamically adjusted through model predictive control to respond quickly to transients and disturbances.
3Stability of the object's composition
If traditional linear controllers are tuned slow with additional filters to avoid reacting to ripple, then stability is improved, but the controller response during transients becomes sluggish
Solution Approach 1:
The model predictive control uses feedback from flux measurements and predictions to adjust switching instances in real-time. This allows the controller to respond quickly to transients without being destabilized by ripple, as the control action is based on predicted future states rather than raw instantaneous measurements.
Solution Approach 2:
The controller performs preliminary prediction of future flux states and adjusts switching instances proactively to achieve desired trajectories. This predictive approach allows fast transient response without needing to filter out ripple, as the control action is based on predicted rather than measured instantaneous values.
4Power
If multiple electrical converters are used to increase power delivery capacity, then power capacity is improved, but coupling between converters complicates control and requires managing circulating currents
Solution Approach 1:
The patent controls each electrical converter independently using individual optimized pulse patterns and model predictive control mechanisms. By segmenting the control of each converter while maintaining coordination through the multi-converter MPC framework, the system achieves high power capacity without excessive control complexity.
Solution Approach 2:
The multi-converter model predictive control uses feedback from all converters to coordinate their operation and manage circulating currents. The controller predicts and adjusts switching instances for each converter based on system-wide conditions, enabling independent converter control while maintaining overall system coordination and managing coupling effects.
Data Source
AI summary
A method is disclosed for controlling an electrical converter system with at least two electrical converters. The method comprises determining a pulse pattern for each electrical converter from a reference voltage of the respective converter. The method further comprises determining a sum current reference and at least one difference current reference, along with determining a measured sum current and at least one measured difference current. The method further comprises determining a sum current error and at least one difference current error, and determining a sum flux modification by multiplying a gain to the sum current error and at least one difference flux modification by multiplying a gain to the respective difference current error. The method additionally comprises mapping the sum and difference flux modifications to a converter flux modification for each electrical converter; modifying the pulse pattern for each electrical converter, and applying the modified pulse patterns to the converters.