Sensor Offset Calibration via Polarity Flipping in Dynamic Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional flipping techniques are inadequate for calibrating sensors in multi-phase systems, as they fail to determine offset impact on individual channels and are unsuitable for highly dynamic systems where fundamental frequency overlaps with flipping frequency, leading to inefficiencies in control loops due to unpredictable current distribution.
Innovation Solution
The proposed method involves flipping the polarity of one sensor while keeping others unchanged, allowing for independent determination of offset and gain mismatch in each sensor, using a sequence of sampling intervals to generate compensated signals, and applying this technique to each sensor in a multi-phase system to account for gain-matched or unmatched conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional flipping techniques are used to calibrate sensors, then offset calibration is performed, but the techniques fail to account for gain mismatch and are unsuitable for highly dynamic systems where fundamental frequency overlaps with flipping frequency
Solution Approach 1:
The patent segments the calibration process into multiple distinct phases: a first calibration phase for determining offset values when sensors are stationary, and a second calibration phase for determining gain mismatch values during dynamic operation. This segmentation allows each phase to address specific calibration aspects independently, enabling the system to handle both offset and gain errors while adapting to highly dynamic conditions where conventional single-phase techniques fail.
2Measurement precision
If sensor polarity is flipped to determine offset, then offset calibration is achieved, but gain mismatch between sensors cannot be determined
Solution Approach 1:
The patent applies preliminary action by first determining offset values during a stationary calibration phase before the system enters dynamic operation. These pre-determined offset values are then stored and used during subsequent dynamic calibration phases, allowing the system to focus on determining gain mismatch values without being confounded by offset errors. This preliminary offset calibration enables accurate gain mismatch determination that would otherwise be impossible.
3Measurement precision
If flipping frequency is used for calibration, then offset can be determined, but in highly dynamic systems the flipping frequency overlaps with the fundamental frequency making calibration unreliable
Solution Approach 1:
The patent employs periodic action by conducting calibration operations at specifically timed intervals rather than continuously. The controller performs calibration during defined calibration periods when the system is stationary or in known states, separating these calibration events from normal dynamic operation. This periodic approach allows the system to perform reliable offset and gain calibration without the flipping frequency interfering with the fundamental frequency during dynamic operation.
Solution Approach 2:
The patent applies dynamics by making the calibration process adaptive to system state. The controller determines whether the system is in a stationary state suitable for offset calibration or in a dynamic state suitable for gain mismatch calibration. This dynamic adaptation allows the system to switch between different calibration modes based on operating conditions, ensuring reliable calibration results regardless of whether the system is highly dynamic or stationary.
4Device complexity
If calibration is performed without accounting for gain mismatch, then offset calibration is simpler, but efficiency is reduced due to uncorrected gain errors
Solution Approach 1:
The patent ensures continuity of useful action by implementing an ongoing calibration process that continuously determines both offset and gain mismatch values during system operation. Rather than performing calibration only during initial setup, the system continuously updates calibration parameters during dynamic operation, ensuring that gain mismatch errors are constantly corrected. This continuous calibration maintains high system efficiency without requiring system shutdown or complex manual intervention.
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 approach enables real-time sensing and offset compensation in multi-phase systems, improving power transfer efficiency by accurately determining and mitigating sensor offsets without integrating over time, even in systems with gain mismatches, thus enhancing the accuracy of control loops.
Implementation Method 1
Magnetic sensors are used to sense current in an alternating current (AC) system and can generate a sense signal
Data Source
AI summary
Techniques are described for calibrating sensors for use in systems in the presence of offset. Sensors may be used to generate sense signals which represent true signals that are part of a system. When the sensors are not calibrated, inefficiency due to offset can be introduced into a system that incorporates the generated sense signal. Flipping techniques may be used to mitigate offset. Applicant has appreciated that when the sensor gains are mismatched, the offset calibration associated with a sensor is not independent from the offset calibration associated with the other sensors. Some of the flipping techniques described herein account for gain mismatch by flipping the polarity of each sensor in a one-at-a-time fashion, and by combining the results in a common system of equations to determine the gain mismatch and the offset of each sensor.