Selecting local constraint for alignment of batch process data with dynamic time warping

Spooner, Max Peter; Kold, David; Külahçı, Murat

doi:10.1016/j.chemolab.2017.05.019

Cited by 13 publications

(16 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Local constraints: Local constraints are illustrated in Figure 5d. Given a node (p i , q j ) from the warping path, the subsequent node should be chosen from (i + 1, j), (i + 1, j + 1) or (i, j + 1) such that the warping path is continuous and monotonically nondecreasing [38,39].…”

Section: Dtwmentioning

confidence: 99%

Spatio-temporal Index Based on Time Series of Leaf Area Index for Identifying Heavy Metal Stress in Rice under Complex Stressors

Tang

Liu

et al. 2020

IJERPH

View full text Add to dashboard Cite

Crops under various types of stresses, such as stress caused by heavy metals, drought and pest/disease exhibit similar changes in physiological-biochemical parameters (e.g., leaf area index [LAI] and chlorophyll). Thus, differentiating between heavy metal stress and nonheavy metal stress presents a great challenge. However, different stressors in crops do cause variations in spatiotemporal characteristics. This study aims to develop a spatiotemporal index based on LAI time series to identify heavy metal stress under complex stressors on a regional scale. The experimental area is located in Zhuzhou City, Hunan Province. The situ measured data and Sentinel-2A images from 2017 and 2018 were collected. First, a series of LAI in rice growth stages was simulated based on the WOrld FOod STudies (WOFOST) model incorporated with Sentinel 2 images. Second, the local Moran’s I and dynamic time warping (DTW) of LAI were calculated. Third, a stress index based on spatial and temporal features (SIST) was established to assess heavy metal stress levels according to the spatial autocorrelation and temporal dissimilarity of LAI. Results revealed the following: (1) The DTW of LAI is a good indicator for distinguishing stress levels. Specifically, rice subjected to high stress levels exhibits high DTW values. (2) Rice under heavy metal stress is well correlated with high-high SIST clusters. (3) Rice plants subjected to high pollution are observed in the northwest of the study regions and rice under low heavy metal stress is found in the south. The results suggest that SIST based on a sensitive indicator of rice biochemical impairment can be used to accurately detect regional heavy metal stress in rice. Combining spatial-temporal features and spectral information appears to be a highly promising method for discriminating heavy metal stress from complex stressors.

show abstract

Section: Dtwmentioning

confidence: 99%

Spatio-temporal Index Based on Time Series of Leaf Area Index for Identifying Heavy Metal Stress in Rice under Complex Stressors

Tang

Liu

et al. 2020

IJERPH

View full text Add to dashboard Cite

show abstract

“…After its development in the field of speech recognition [12], DTW has become widely used in batch process monitoring [8,18,9,19,20] to process the batches such that they have the same number of observations, and so that events within each batch are synchronised. This is done by selecting a single reference batch, and aligning the other batches to this reference batch.…”

Section: Dynamic Time Warpingmentioning

confidence: 99%

“…The local constraints suggested in [12] are parametrised by the value P , which represents the number of diagonal steps the warping function must take for each horizontal or vertical step. A method for selecting the value of P may be found in [20] and is applied in this work.…”

Section: Accepted Manuscriptmentioning

confidence: 99%

“…The variable weights, W used in this alignment are here determined using the method in[18] which was developed to give greater weights to those variables which are most indicative of process time. To avoid extreme warpings, a local constraint is selected according to[20], and a band global constraint of width equal to the range of the batch durations is used.Let Z (I × J × N ) be a three-way data matrix formed by the aligned NOC reference data set (where N is the length of the reference batch). This matrix is unfolded batch-wise to form Z (I × JN ).…”

mentioning

confidence: 99%

“…The resulting weights are shown inFig. 2.Next, using the method in[20], a local constraint of P = 1 was selected.This constraint enforces one diagonal step for each vertical or horizontal step Variable weights for DTW taken by the warping function.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Monitoring batch processes with dynamic time warping and k-nearest neighbours

Spooner

Külahçı

2018

Chemometrics and Intelligent Laboratory Systems

Self Cite

View full text Add to dashboard Cite

 Users may download and print one copy of any publication from the public portal for the purpose of private study or research.  You may not further distribute the material or use it for any profit-making activity or commercial gain  You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

show abstract