Resolver Harmonic Error Correction Without Large Lookup Tables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional resolver devices require large storage areas to correct angle errors, which is inefficient for high-precision angle detection, especially when the resolver is accelerating or decelerating.
Innovation Solution
The resolver device employs n-th-harmonic error estimation units to real-time estimate and subtract error waveforms from the resolver angle, eliminating the need for large storage areas by updating amplitude and phase of angle errors for each frequency component.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional resolver devices store resolver error for each rotation angle in an error information table, then angle error correction is achieved, but storage area becomes excessively large
Solution Approach 1:
The patent transforms the error correction approach from storing complete error values for each angle to storing only amplitude and phase parameters of harmonic components. By representing the error waveform as a sum of harmonic components with specific amplitudes and phases, the storage requirement is dramatically reduced while maintaining correction accuracy.
Solution Approach 2:
Instead of storing the actual error waveform data directly, the patent creates a simplified mathematical model (harmonic representation) that copies the essential characteristics of the error. This model uses amplitude and phase parameters to reproduce the error waveform through harmonic synthesis, achieving compression without significant loss of correction effectiveness.
2Reliability
If large storage areas are used to correct angle errors at all rotation angles, then comprehensive error correction is achieved, but device complexity and memory requirements increase
Solution Approach 1:
The patent reduces the complexity by changing what parameters are stored. Instead of storing complete error lookup tables that require complex memory management and retrieval operations, the system stores simple amplitude and phase parameters that can be easily processed through mathematical operations to generate correction values on-the-fly.
3Quantity of substance
If conventional methods store error information for uniform velocity rotation only, then storage is reduced, but error correction fails during acceleration or deceleration
Solution Approach 1:
The patent creates a dynamic error correction system that can adapt to varying operational conditions. By using harmonic representation with amplitude and phase parameters, the system can generate accurate correction values regardless of whether the resolver is rotating at uniform velocity, accelerating, or decelerating, making the correction mechanism versatile across all motion states.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
An n-th-harmonic error estimation unit that estimates the error component of an n-th harmonic included in a resolver angle θ and a subtraction unit that subtracts an n-th-harmonic estimated angle error from θ to output the corrected angle θ' are included. The n-th-harmonic error estimation unit includes an n-th-harmonic error phase detection unit that obtains a phase difference u such that the integral, for a 1/n period of an electrical angle of the rotor, of the output obtained by synchronously detecting θ by using a rectangular wave obtained by comparison from a COS wave expressed as cos(nθ + u) becomes zero and generates a SIN wave expressed as sin(nθ + u); a synchronous detector that synchronously detects θ' by using a rectangular wave obtained by comparison from the SIN wave; an integrator that integrates the detected output for a 1/n period of the electrical angle; an actual angle integral calculation unit; an amplitude setter that sets an error amplitude from the value obtained by subtracting the integral of the actual angle integral calculation unit from the integral of the integrator; and a multiplier that generates the n-th-harmonic estimated angle error by multiplying the SIN wave by the error amplitude. A large storage area becomes unnecessary, and an angle error can be corrected.