Motor Speed Estimation via Multiple-Point Smoothing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing motor-speed estimation methods face challenges in achieving smooth and real-time speed measurement, particularly at high speeds, due to position quantization errors from encoders, leading to ripple issues and phase delays.
Innovation Solution
A multiple-point smoothing method that over-samples encoder output, averages position differences, and applies low-pass filtering to obtain a final speed estimation, using an over-sampling factor that remains constant regardless of motor rotation speed, thereby reducing ripple and phase delay.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If encoder output is sampled at standard sampling frequency, then device complexity is reduced, but speed measurement smoothness deteriorates due to position quantization errors
Solution Approach 1:
The patent divides the speed estimation process into multiple segments by using multiple past position measurements (P(t-1), P(t-2), ..., P(t-N)) rather than a single position difference. This segmentation of the measurement process allows the system to average out quantization errors across multiple segments, improving smoothness without requiring higher sampling frequency.
Solution Approach 2:
The patent performs preliminary smoothing by calculating a weighted average of multiple past position differences before final speed computation. This preliminary action of averaging position differences from multiple previous samples reduces the impact of quantization errors before the final speed calculation, achieving smooth speed measurement at standard sampling frequencies.
2Device complexity
If standard speed estimation method is used, then device complexity is reduced, but measurement precision deteriorates due to ripple errors at specific speeds
Solution Approach 1:
The patent introduces a dynamic weighting mechanism where the weights of past position measurements are adjusted based on the ripple detection result. When ripple is detected at current speed, the system dynamically modifies the weighting factors to exclude or reduce the influence of affected measurement points, thereby maintaining measurement precision across varying speeds without increasing device complexity.
Solution Approach 2:
The patent changes the parameters of the speed estimation calculation by using a weighted average formula with adjustable weights (w1, w2, ..., wN) instead of a simple difference calculation. This parameter change allows the system to optimize the estimation by giving different weights to different past measurements, reducing ripple errors while maintaining computational simplicity.
3Ease of operation
If simple speed estimation is used, then ease of operation is improved, but reliability deteriorates due to phase delay at high speeds
Solution Approach 1:
The patent performs preliminary calculations of multiple position differences and their weighted average before final speed determination. This preliminary action prepares smoothed and more accurate intermediate values that compensate for potential phase delays, allowing the final speed estimation to be more reliable at high speeds while maintaining the simplicity of the overall operation through automated calculation.
Solution Approach 2:
The patent implements a feedback mechanism where the estimated speed is continuously updated based on the weighted average of past position measurements. This feedback approach ensures that the speed estimation adapts to changing motor conditions, improving reliability at high speeds by continuously refining the estimate based on recent measurement history without complicating the operational interface.
Data Source
AI summary
A multiple-point smoothing method for motor-speed estimation. The output of an encoder is over-sampled to obtain over-sampled position differences according to an over-sampling factor M. The over-sampled position differences are averaged to obtain an initial speed estimation. The initial speed estimation is low-pass filtered to obtain a final speed estimation. The final speed estimation is sent to a speed controller to control the speed of a motor. Alternatively, in a composite multiple-point smoothing scheme, the initial speed estimations based on two over-sampling factors are obtained. One of the initial speed estimations is selected based on an average of the two estimations, and the selected initial speed estimation is further processed to obtain a final speed estimation. The method of the present invention can reduce ripple in estimated speed of motor when the motor is operated at high speed.