A Solution to Ambiguities in Position Estimation for Solenoid Actuators by Exploiting Eddy Current Variations

Area of Science:

• Electrical Engineering
• Mechanical Engineering
• Control Systems

Background:

• Solenoid actuators require accurate position estimation for various applications.
• Current sensorless techniques often rely on differential inductance, which can lead to estimation ambiguities due to non-monotonic dependencies.
• A unique position estimation, especially in standstill conditions, is crucial for reliable actuator monitoring.

Purpose of the Study:

• To develop a novel sensorless position estimation technique for solenoid actuators.
• To overcome the ambiguities in position estimation caused by non-monotonic inductance-position relationships.
• To enable unique and precise position estimation, including standstill conditions, for low-cost applications.

Main Methods:

• Identification and exploitation of eddy current losses as a parallel resistance parameter.
• Online estimation of differential inductance and parallel resistance with low computational effort.
• Implementation of a data fusion algorithm utilizing a neural network for position estimation.

Main Results:

• Successfully resolved position estimation ambiguities by incorporating eddy current loss analysis.
• Achieved online estimation of differential inductance and parallel resistance.
• Demonstrated unique, precise, and high signal-to-noise ratio position estimation in a solenoid actuator end-position detection use case.
• Validated the approach for standstill conditions.

Conclusions:

• The proposed method effectively addresses ambiguities in solenoid actuator position estimation.
• Eddy current loss analysis provides a robust solution for sensorless monitoring.
• The neural network-based data fusion enhances estimation accuracy and reliability.
• This approach offers a low-cost, low-implementation-effort solution for unique position estimation in solenoid actuators.