Analog Input Conversion Using Trimmed Multi-Sample Noise Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional analog input modules face challenges in achieving data stability due to noise interference, requiring increased conversion times which compromise conversion speed and lead to unsatisfactory results, and existing methods either fail to provide fine control or result in inappropriate data fluctuations.
Innovation Solution
The method involves converting analog data into digital data by repeating the conversion process a predetermined number of times, sorting the data by size, and determining the mean value among the sorted data from current and previous conversion periods to minimize noise influence and ensure stability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conversion times are increased to enhance stability of conversion data, then data stability is improved, but data conversion time increases and conversion speed specification cannot be satisfied
Solution Approach 1:
The patent applies partial action by performing a fixed small number of conversions (n=3) and using a trimming algorithm that discards only the maximum and minimum values. This partial processing approach achieves noise reduction without requiring excessive conversion times, thus maintaining conversion speed while improving data stability.
Solution Approach 2:
The patent performs preliminary trimming of conversion data by removing maximum and minimum values before calculating the average. This preliminary action eliminates outliers caused by noise in advance, allowing the system to achieve stable data with fewer conversions and maintain specified conversion speed.
2Reliability
If average value method is used to eliminate noise, then some noise influence is reduced, but stability is still insufficient when multiple continuous values are affected by noise
Solution Approach 1:
The patent uses a trimming algorithm that removes only the extreme values (maximum and minimum) from the conversion data set. This partial elimination approach effectively removes outliers caused by noise while preserving the majority of valid data points, achieving better noise elimination than simple averaging while maintaining measurement precision.
3Reliability
If hysteresis characteristic is applied to recognize data fluctuation, then data stability is improved, but fine control capability is lost due to stair-shaped fluctuations
Solution Approach 1:
The patent applies a trimming algorithm that removes only the extreme values (maximum and minimum) from the conversion data set. This partial elimination approach effectively removes outliers caused by noise while preserving the majority of valid data points, achieving better noise elimination than simple averaging while maintaining measurement precision.
Data Source
AI summary
A method for converting analog data into digital data for analog input module is disclosed, wherein the analog input module adds at least one or more digital data determined at previous conversion periods (k−1th, . . . k−dth, said k and d being a natural number) to n numbers of digital data converted to the current conversion period (kth), and determines a value corresponding to a mean level in size among the digital data as a final digital conversion value of the current conversion period (kth), and outputs the value, and as a result, even if the n numbers of digital data converted to the current conversion period (kth) are greatly deviated from an original value by noise, a stable result having a minimum influence by the noise can be provided in industrial sites where many noises are introduced, by allowing a digital data value approximate to the original value to be outputted.