In rotating machinery analysis, tachometer signals are often collected. If a 1 pulse per revolution tachometer is sampled at the same frequency as the vibration signal, then there is some jitter in the tachometer signal. Smoothing cubic splines are a preferred method of removing this jitter. This …

In that case, here is a quick example showing how such a fit would be done in the current version of BSplineKit. I may add a higher level interface for this in the future. using BSplineKit using LinearAlgebra using SparseArrays using Random using Plots # Exact function f(x) = exp(-x) * cos(8x) # …

@marius311 , looking at the Loess.jl page, the robust fit smooth line seems to be way off? Tested it below against SmoothingSplines.jl. and good results were obtained using your recommended Loess span parameter. using Loess, SmoothingSplines, Plots xs = sort(10 .* rand(200)) ys = sin.(xs) yn = ys …

Yea that example it looks kind of weird. The main option to play with is span, lowering it does less smoothing, with loess(xs, yn, span=0.5) your example works much better I think.

@marius311 , thanks for the feedback. It looks much better now with your recommended Loess span parameter. I have updated example above accordingly and included some massive outliers to spice it up further.

Here is a very simple constant speed example of the raw data with some spline fits that I did using Matlab and matlab central and other toolboxes. It gives an example of the jitter in the tach signal. I will lookup some other tach signals and post them from non-constant speed sources to give some …

A bit more detail for those curious. If you picture some reflective tape on a rotating shaft and then a digital optical pickup on the shaft and collect the data with a standard digital signal analyser that includes anti-alias filtering you will get a signal much like this. [tach] . Note the signal …

Would it be possible to put in a dropbox your tachometer signal and share it?

This example is from a winder similar to that shown in https://www.valmet.com/board-and-paper/board-and-paper-machines/winders-and-roll-handling/winding/two-drum-winders/ . The data is available for a week at Transfer - Dropbox . Note the the data contains the raw tachometer signals which have been …

@Jake , please confirm which of the datasets you would like to interpolate.

See below results for the pure Julia packages: (1) BSplineKit.jl’s nice solution given by @jipolanco (2) SmoothingSplines.jl (3) Loess.jl suggested by @marius311 They work relatively well for your dataset with following caveats: (1) allows defining the number of knots but is a bit less smooth t…

Smoothing interpolation for tachometer data

New to Julia

Jake March 2, 2021, 4:04pm 20

Perhaps midtime1 and rotatationalspeed1 is a good place to start.

How to smooth zig-zag?

Topic		Replies	Views
Using Interpolations New to Julia interpolations	47	20108	October 5, 2020
Best way to take derivatives of unevenly spaced data (with interpolations? discrete derivatives?) Numerics intervals , interpolations	37	5564	April 30, 2021
[ANN] KissSmoothing.jl - Hassle free & simple data smoothing Package Announcements smoothing	7	2202	April 2, 2022
[ANN] FastInterpolations.jl — Fast & zero-allocation ND interpolation Package Announcements package , interpolations , high-performance	34	2504	June 1, 2026
My code for cubic spline interpolation plots different image everytime and not smooth New to Julia interpolations , splines	4	1397	October 28, 2021

Smoothing interpolation for tachometer data

Related topics