How Can Scattered Data Impact Tool Performance Analysis?

  • Thread starter Thread starter serbring
  • Start date Start date
  • Tags Tags
    Data
Click For Summary

Discussion Overview

The discussion centers around the analysis of scattered data from measurements of a tillage machine using two different soil engaging tools (A and B). Participants explore the implications of data variability on the assessment of force and power differences between the tools, considering aspects of data collection, statistical analysis, and experimental conditions.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested
  • Mathematical reasoning
  • Experimental/applied

Main Points Raised

  • One participant notes the difficulty in drawing conclusions due to the scattered nature of the data and suggests that outliers may be present, complicating the identification of a suitable regression curve.
  • Another participant proposes the idea of time-based filtering to reject impulse shocks, speculating that rocks could be a source of noise, although the original poster excludes this possibility.
  • Some participants express skepticism about the data quality, suggesting that the experiment may have significant issues affecting the results.
  • A participant mentions that the spread of data for tool B appears greater than for tool A, indicating potential differences in variability.
  • There is a discussion about the need for multiple runs at constant speeds to average down variability, with one participant indicating that six runs may not be sufficient.
  • Concerns are raised regarding the relationship between force and speed, with conflicting observations about expected trends and behavior of the data.
  • Participants discuss the use of linear regression and residual analysis, with one expressing uncertainty about handling data with different variance magnitudes.
  • Clarifications are sought regarding the methodology of converting force vs. time data to force vs. speed data, highlighting confusion about the data collection process.

Areas of Agreement / Disagreement

Participants express a range of views on the data quality and analysis methods, with no consensus on the reliability of the results or the best approach to evaluate the data. Multiple competing perspectives on the implications of the scattered data and its analysis remain unresolved.

Contextual Notes

Limitations include potential outliers affecting regression analysis, the need for clearer definitions of experimental conditions, and unresolved questions about the uniformity of the ground during testing.

  • #31
I'm sorry, without thinking about it, I was using the time-series terminology "cross correlation". That was probably misguiding. It's just the correlation between the two variables that I was thinking of.
 
Engineering news on Phys.org
  • #32
I have also computed the Pearson's correlation coefficient of the average values of draft with respect 'speed and depth, respectively. The correlation is higher with speed than with depth.

Instead, by computing the Pearson's correlation coefficient of the time based draft with respect 'time based speed and depth, respectively, sometimes, the reselts are pretty variable. Sometimes, the correlation is strong, other times the correlation is weak. Sometime, the correlation is stronger with speed, other times is stronger with depth.
 

Similar threads

  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 109 ·
4
Replies
109
Views
9K
  • · Replies 6 ·
Replies
6
Views
6K
Replies
6
Views
4K
  • · Replies 4 ·
Replies
4
Views
2K
Replies
1
Views
2K