MMC Heat Balancing via Loss Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In Modular Multilevel Converters (MMCs), heat generation due to conduction and switching losses leads to uneven semiconductor switch temperature distribution, resulting in reduced converter performance and increased wear, with hot spots limiting overall performance.
Innovation Solution
A method for heat balancing in full-bridge cells of MMCs, where a loss and temperature model estimates real-time heat losses at each semiconductor switching device, allowing the local cell controller to select between two zero output voltage switching states to balance heat among devices, without interrupting regular operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional PWM control is used in MMC, then the converter can operate with simple control logic, but heat losses are unevenly distributed among semiconductor switches leading to hot spots and reduced reliability
Solution Approach 1:
The patent applies preliminary action by calculating and storing the on-state losses for each semiconductor switch before normal operation begins. These pre-calculated loss values are then used during runtime to determine optimal switching states, eliminating the need for complex real-time thermal modeling while achieving heat balancing.
Solution Approach 2:
The patent implements feedback by continuously monitoring the thermal state of semiconductor switches and using this information to adjust switching decisions. The controller selects switching states that balance the thermal conditions across all switches, creating a closed-loop control system that actively manages heat distribution.
2Measurement precision
If real-time temperature measurement is implemented for each semiconductor switch, then accurate temperature balancing can be achieved, but the system complexity and cost increase significantly
Solution Approach 1:
The patent uses an intermediary approach by introducing a thermal model that estimates switch temperatures based on easily measurable electrical parameters (current and voltage) rather than direct temperature sensing. This thermal model acts as a mediator between electrical measurements and thermal state estimation, providing accurate temperature information without physical sensors.
Solution Approach 2:
The patent replaces the mechanical/physical temperature measurement system (thermocouples or RTD sensors) with an electrical-based thermal modeling approach. By substituting physical sensing with electrical measurement and mathematical modeling, the system achieves temperature monitoring without the complexity of installing and maintaining physical temperature sensors on each semiconductor switch.
3Duration of action of stationary object
If heat balancing control is implemented to optimize temperature distribution, then semiconductor wear is reduced and reliability improves, but the control algorithm becomes more complex
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing the on-state losses for each semiconductor switch configuration before normal operation. During runtime, the controller simply retrieves these pre-computed values and uses them to make switching decisions, transforming a potentially complex real-time optimization problem into a simple lookup and comparison operation.
Solution Approach 2:
The patent changes the control parameter from direct temperature measurement to estimated thermal state based on electrical parameters. By transforming the control variable from physical temperature (requiring sensors and complex processing) to electrical-based thermal estimation, the control algorithm becomes simpler while maintaining effectiveness in balancing heat distribution.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method effectively balances heat between semiconductor switches within a cell, improving the overall capacity of the converter by optimizing temperature distribution without the need for actual temperature measurements, using a mathematical model based on electrical current and voltage data.
Implementation Method 1
There are losses in the form of heat generation in the cells of the MMC, both conduction losses and switching losses
Data Source
Figure 1~2
Figure 3
AI summary
The present disclosure relates to a method for heat balancing performed in a full-bridge cell (3) of a phase leg of a Modular Multilevel Converter (MMC). The method comprises obtaining real-time measurements of parameters of the cell. The method also comprises, based on the obtained measurements, by means of a predetermined mathematical model, estimating real-time heat losses at each semiconductor switching device (V1,..., V4) of the full-bridge. The method also comprises, based on the estimated losses, selecting one of a first and a second zero output voltage switching states for balancing heat between the switching devices, when a modulator of the MMC requests a zero output voltage of the cell.