Flux Space Vector Model Correction for Sensorless Motor Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the position of the flux space vector in electric motors without position sensors are prone to integration drift, leading to inaccurate models, especially at low frequencies, and require additional feedback that distorts the flux space vector.
Innovation Solution
A control method for electric motors that uses a feedback mechanism where the difference between a reference flux space vector and the integration result is fed back to the integrator, allowing the flux space vector model to be corrected in both magnitude and angle, eliminating drift and speed-dependent errors, and utilizing known machine parameters and rotation direction as input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If open-loop integration is used to determine the flux space vector from stator voltage and current, then the method can be implemented without position sensors, but the integration result drifts away over time due to small offsets in variable acquisition
Solution Approach 1:
The patent applies feedback by comparing the determined flux space vector with a reference flux space vector and using the difference to correct the integration result. The feedback signal is generated by forming a difference between the determined flux space vector and the reference flux space vector, then feeding this difference back to correct the integration, thereby preventing drift accumulation and maintaining accuracy over time.
2Measurement precision
If feedback is provided to suppress integration drift, then the flux space vector accuracy is improved, but the model value increasingly deviates from the actual flux space vector at low speeds
Solution Approach 1:
The patent applies dynamics by making the feedback mechanism adaptive to operating conditions. The feedback signal is dynamically adjusted based on the difference between the determined flux space vector and the reference flux space vector, allowing the system to maintain accuracy across varying speeds. The feedback gain and application are optimized to work effectively at both high and low speeds, unlike fixed feedback approaches.
Solution Approach 2:
The patent applies parameter changes by using a reference flux space vector that is determined independently of integration drift. The reference vector serves as a stable benchmark that does not suffer from the same drift issues, allowing the feedback mechanism to correct the determined vector accurately across different operating conditions including low speeds.
3Stability of the object's composition
If feedback is used to correct the flux space vector, then integration drift is suppressed, but the feedback variable distorts the observed flux space vector when measurement offsets are present
Solution Approach 1:
The patent applies the intermediary principle by introducing a reference flux space vector as a mediator between the integration process and the feedback mechanism. The reference vector acts as an unbiased benchmark that is not affected by integration drift or measurement offsets. By comparing the determined vector against this independent reference and using the difference as feedback, the system avoids the distortion problems that arise when feeding back the raw integration result directly.
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 ensures accurate flux space vector modeling without speed-dependent errors, even at low frequencies, by using a drift-free reference flux space vector, reducing computational complexity and maintaining model accuracy across a wider speed range.
Implementation Method 1
an induced voltage space vector UI is determined, which is supplied to an integrator, such that a flux space vector is generated
Data Source
Figure 1
Figure 2
AI summary
Open-loop or closed-loop control method for a converter which feeds an electric motor, wherein a current space vector is detected as the motor current and the motor voltage, in particular a voltage space vector, is set, wherein an induced voltage space vector (Formula A) is determined, which is supplied to an integrating element, a flux space vector is produced, the angle of which is at right angles to the voltage space vector, wherein the magnitude of the flux space vector corresponds to a predetermined rated value, and the difference between the integration result and the flux space vector produced in this way is used as feedback for the integration element.