Response time correction of slow response sensor data by deconvolution of the growth-law equation
- 1Centre for Arctic Gas Hydrate, Environment, and Climate,UiT The Arctic University of Norway, Tromsø, Norway
- 2Institute for Physics and Technology, UiT The Arctic University of Norway, Tromsø, Norway
- 3CNRS, University of Grenoble Alpes, IRD, Grenoble INP, 38000 Grenoble, France
- 44H-JENA engineering GmbH Wischhofstrasse 1-3, 24148 Kiel, Germany
- 1Centre for Arctic Gas Hydrate, Environment, and Climate,UiT The Arctic University of Norway, Tromsø, Norway
- 2Institute for Physics and Technology, UiT The Arctic University of Norway, Tromsø, Norway
- 3CNRS, University of Grenoble Alpes, IRD, Grenoble INP, 38000 Grenoble, France
- 44H-JENA engineering GmbH Wischhofstrasse 1-3, 24148 Kiel, Germany
Abstract. Accurate, high resolution measurements are essential to improve our understanding of environmental processes. Several chemical sensors relying on membrane separation extraction techniques have slow response times due to a dependence on equilibrium partitioning across the membrane separating the measured medium (i.e., a measuring chamber) and the medium of interest (i.e., a solvent). We present a new technique for deconvolving slow sensor response signals using statistical inverse theory; applying a weighted linear least squares estimator with the growth-law as measurement model. The solution is regularized using model sparsity, assuming changes in the measured quantity occurs with a certain time-step, which can be selected based on domain-specific knowledge or L-curve analysis. The advantage of this method is that it: 1) models error propagation, providing an explicit uncertainty estimate of the response time corrected signal, 2) enables evaluation of the solutions self consistency, and 3) only requires instrument accuracy, response time, and data as input parameters. Functionality of the technique is demonstrated using simulated, laboratory, and field measurements. In the field experiment, the coefficient of determination (R2) of a slow response methane sensor in comparison with an alternative, fast response sensor, significantly improved from 0.18 to 0.91 after signal deconvolution. This shows how the proposed method can open up a considerably wider set of applications for sensors and methods suffering from slow response times due to a reliance on the efficacy of diffusion processes.
Knut Ola Dølven et al.
Status: final response (author comments only)
-
RC1: 'Comment on gi-2021-28', Anonymous Referee #1, 13 Dec 2021
- AC3: 'Reply on RC1', Knut Ola Dølven, 08 Apr 2022
-
RC2: 'Comment on gi-2021-28', Anonymous Referee #2, 03 Feb 2022
In this paper, the growth law is used to model the slow response measurement process, and a new technology of deconvolution of the measurement signal of the slow response sensor based on the weighted least square is proposed, so as to achieve the measurement effect similar to that of the fast response sensor.
Some views:
1ï¼There are too many curves in Figure 3ï¼it is difficult to see clearly;
2ï¼The experimental results of the paper show that the real response signal can be extracted from the measurement signal of slow response sensor to eliminate the influence of transmembrane effect, which is in good agreement with the measurement results of DTB sensor. However, the experimental results of the algorithm are introduced in the summary. It is not understood that the correlation R has increased from 0.18 to 0.91. Because the slow response curve is very different from the fast response curve, the correlation between the two must be very low. The correlation between the fast response signal extracted from the slow response signal and the fast response signal measured directly must be very high. It doesn't feel that it can be said to be "improved", nor can it reflect the advantage of this algorithm to obtain the fast response signal;
3ï¼ This paper mainly analyzes the influence of time step on the stability of the algorithm. Are there other factors?
4ï¼ Based on the relevant knowledge of slow response and fast response sensors, is it a good way to directly measure fast response signals? Or is it better to extract from slow response?
- AC1: 'Reply on RC2', Knut Ola Dølven, 08 Apr 2022
Knut Ola Dølven et al.
Knut Ola Dølven et al.
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
784 | 115 | 23 | 922 | 9 | 5 |
- HTML: 784
- PDF: 115
- XML: 23
- Total: 922
- BibTeX: 9
- EndNote: 5
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1