Using a Smoothing Algorithm to Process Water Level … a Smoothing Algorithm to Process Water Level...
Transcript of Using a Smoothing Algorithm to Process Water Level … a Smoothing Algorithm to Process Water Level...
QUICK TIPS
(--THIS SECTION DOES NOT PRINT--)
This PowerPoint template requires basic PowerPoint
(version 2007 or newer) skills. Below is a list of
commonly asked questions specific to this template.
If you are using an older version of PowerPoint some
template features may not work properly.
Using the template
Verifying the quality of your graphics
Go to the VIEW menu and click on ZOOM to set your
preferred magnification. This template is at 100%
the size of the final poster. All text and graphics will
be printed at 100% their size. To see what your
poster will look like when printed, set the zoom to
100% and evaluate the quality of all your graphics
before you submit your poster for printing.
Using the placeholders
To add text to this template click inside a
placeholder and type in or paste your text. To move
a placeholder, click on it once (to select it), place
your cursor on its frame and your cursor will change
to this symbol: Then, click once and drag it to
its new location where you can resize it as needed.
Additional placeholders can be found on the left
side of this template.
Modifying the layout
This template was specifically designed for a 48x36
tri-fold presentation. Its layout should not be
changed or it may not fit on a standard board. It has
a one foot column on the left, a 2 foot column in
the middle and a 1 foot column on the right.
The columns in the provided layout are fixed and
cannot be moved but advanced users can modify any
layout by going to VIEW and then SLIDE MASTER.
Importing text and graphics from external sources
TEXT: Paste or type your text into a pre-existing
placeholder or drag in a new placeholder from the
left side of the template. Move it anywhere as
needed.
PHOTOS: Drag in a picture placeholder, size it first,
click in it and insert a photo from the menu.
TABLES: You can copy and paste a table from an
external document onto this poster template. To
adjust the way the text fits within the cells of a
table that has been pasted, right-click on the table,
click FORMAT SHAPE then click on TEXT BOX and
change the INTERNAL MARGIN values to 0.25
Modifying the color scheme
To change the color scheme of this template go to
the “Design” menu and click on “Colors”. You can
choose from the provide color combinations or you
can create your own.
QUICK DESIGN GUIDE (--THIS SECTION DOES NOT PRINT--)
This PowerPoint 2007 template produces a 36”x48”
tri-fold presentation poster. It will save you
valuable time placing titles, subtitles, text, and
graphics.
Use it to create your presentation. Then send it to
PosterPresentations.com for premium quality, same
day affordable printing.
We provide a series of online tutorials that will
guide you through the poster design process and
answer your poster production questions.
View our online tutorials at:
http://bit.ly/Poster_creation_help
(copy and paste the link into your web browser).
For assistance and to order your printed poster call
PosterPresentations.com at 1.866.649.3004
Object Placeholders
Use the placeholders provided below to add new
elements to your poster: Drag a placeholder onto
the poster area, size it, and click it to edit.
Section Header placeholder
Move this preformatted section header placeholder
to the poster area to add another section header.
Use section headers to separate topics or concepts
within your presentation.
Text placeholder
Move this preformatted text placeholder to the
poster to add a new body of text.
Picture placeholder
Move this graphic placeholder onto your poster, size
it first, and then click it to add a picture to the
poster.
RESEARCH POSTER PRESENTATION DESIGN © 2011
www.PosterPresentations.com
© 2011 PosterPresentations.com 2117 Fourth Street , Unit C
Berkeley CA 94710 [email protected]
Student discounts are available on our Facebook page.
Go to PosterPresentations.com and click on the FB icon.
Using a Smoothing Algorithm to Process Water Level Logger Data
Green stormwater infrastructure (GSI) is used to mitigate the issues
associated with the quantity and water quality of runoff. On the
main campus of the University of Toledo (UT), a tree filter was
installed to infiltrate and treat stormwater runoff. As a part of tree
filter performance evaluation, a pressure-based water level logger
has been installed and water depth measurements are being
collected in the tree filter catch basin during rain events. Outlier
data points were observed and attributed to turbulence based on
laboratory experiments. An algorithm was created in Python with
the aim of smoothing data that can eventually be used to calibrate
EPA’s Storm Water Management Model (SWMM).
Background and Introduction
Objective
U20-001-04 and U20L-04 HOBO Water Level Loggers by Onset were
used to collect data in the lab and the field in 2014 and 2015.
Methods
Applied algorithm removed outliers • Outlier removal was seen in all data sets
• Algorithms for both field and lab successfully removed outliers
• Laboratory data required smaller sliding window and threshold than field data
• Small proportion of data points were removed from laboratory data (2.4%) as
compared to field data (28.5% for 2014 data and 9.4% for 2015 data)
• Outlier number can be seen from graphs to be greater during times with more depth
variation
Results and Discussion
Conclusions
SWMM Calibration:
SWMM is a simulation based hydrologic/hydraulic model that
incorporates GSI. Given site data, it can be used to predict
stormwater characteristics including flow rate and depth at various
locations (pipes, catch basins). SWMM is calibrated against flow
meter data (Rossman, 2004). However, flow meters are expensive
and difficult to install. We propose calibrating SWMM with water
depth data.
References
Further Improvement of Outlier Removal Approach:
• Adjust sliding window based on individual data set size and
pattern (i.e., specific rain event).
• Investigate the ideal threshold for larger datasets
• Improve outlier removal for smaller data sets
Acknowledgments
NSF REU Program #DBI-1461124 to The University of Toledo’s Lake
Erie Center, “Undergraduate Research and Mentoring- Using the
Lake Erie Sensor Network to Study Land-Lake Ecological Linkages”
This research site was on
the University of Toledo
Main Campus adjacent to
the Law School Parking Lot.
A tree filter was installed in
2014 as a green stormwater
demonstration project to
treat runoff prior to release
into the Ottawa River.
a35 School House Rd. Swan Lake NY, 12783 [email protected]
Department of Civil Engineering, University of Toledo, Toledo OH, 43606-3390
Lisa Ponce a, Anthony Dietrich
Site Description
The overarching objective was to determine if water level
loggers could be used to calibrate a hydrologic model (SWMM).
Specific objectives:
•Determine cause of erratic data in field studies
•Finding and removing outliers from water level logger field data
The contributing watershed (0.32 hectares) is a parking lot that includes four catch basins
(left). The cross-section of the tree filter (Storm Tree LLC) shows an open bottom design
with an inlet basin, a weir wall and an underdrain with an overflow (right).
The tree filter (Storm Tree LLC).
A weather station was installed
at the site. It collects the
barometric pressure data used
to convert the logger depth data
in pressure (kpa) to depth in
length (cm).
The weather station data is automatically uploaded
onto a HOBOlink data base. The images above are
samples of pressure and rain data taken by the
weather station. Readings are taken every minute.
Data Smoothing with Median Absolute Deviation (MAD)
based Algorithm written in Python (Leys et.al, 2013)
• Sliding window chosen for all datasets
100 minutes for field data , 30 seconds for lab data
• Calculate MAD for every 100 min. or 30 sec.
MAD = mediani (|Xi – medianj(Xj)|) * C
Xj = list of 100 or 30 datapoints
medianj = median of Xj
Xi = # in Xj
median = median from list after subtraction
C = 1/upper quartile of whole data set
• Assign Z-score to each data point
Z-score = |Xi – medianj(Xj)|/MAD
• Determine outliers based on Z-score by defining threshold
If : Z-score > 2.5 (field data), Z-score > 2 (lab data); Then: # = outlier
Applying algorithm to data sets:
• csv files from field and lab data imported into algorithm
• Manipulated csv file is output
• Original data is graphed against manipulated data in MS Excel
Logger deployed in catch basin
U20-001-04 HOBO Water Level Logger
Laboratory studies verified impact of
turbulence on water level logger data.
0.85
0.9
0.95
1
1.05
1.1
1.15
7/10/2015 7/13/2015 7/16/2015 7/19/2015
Depth (ft)
Date
Depth Data Taken in 2015 in the Tree Filter Catch Basin Before and After
Outlier Removal
Original
Manipulated
0.35
0.37
0.39
0.41
0.43
0.45
0 60 120 180 240 300 360 420
Depth
(ft
)
Time (s)
Depth Data Taken by Loggers Submerged in Turbulent and Standing
Water
Turbulent
Standing
Sample data from turbulence tests
1.6
1.8
2
2.2
2.4
6/11/2014 6/15/2014 6/19/2014 6/23/2014
Depth (ft)
Date
Depth Data Taken in 2014 in Tree Filter Catch Basin Before and After
Outlier Removal
Manipulated
Original
0.35
0.36
0.37
0.38
0.39
0.4
0.41
0.42
0.43
0 200 400
Depth (ft)
Time (secs)
Depth Data Taken in Lab Turbulence Tests Before and After Outlier Removal
Original
Manipulated
Water Level Loggers
Lab experiment testing logger
sensitivity to turbulence
# Data Points
Before
Outlier
Removal
# Data
Points After
Outlier
Removal
Total #
Data Points
Removed
Lab Data 420 410 10
Field Data
(2014)
18197 13003 5194
Field Data
(2015)
10174 9220 954
1.7
1.8
1.9
2
2.1
6/11/2014 6/17/2014 6/23/2014
De
pth
(ft
)
Date
Potential for SWMM Depth Data to Match Logger with Calibration
SWMM
Logger
SWMM data from 2014 simulation was
manipulated in MS Excel to demonstrate
how the calibration of SWMM can
improve its prediction.
Screenshot taken of SWM Model created
to predict behavior of tree filter during
storm events
Data Collection • LAB: Testing logger sensitivity to turbulence in a jar tester
• FIELD: Installation of pressure-based loggers in catch basins
• Extraction of data from sensors with HOBOware
• Plotting of control vs. variable data in MS Excel
Weather Station
Tree Filter
Top of catch basin under tree
Outcomes • Turbulence was confirmed to cause variations in water level
logger data.
• A smoothing algorithm was applied to field data to effectively
remove outliers.
• Water level logger data should be able to be used to calibrate
SWMM models.
The water level loggers were hung in the catch basins
with non-stretch cable. The cables were cut long
enough to allow the loggers to reach the bottom of
the catch basin.
https://www.google.com/earth/
http://storm-tree.com/
https://www.hobolink.com/
http://storm-tree.com/
Leys, C., Klein, O., Bernard, P., Licata, L. (2013). Detecting Outliers: Do Not Use Standard Deviation Around the Mean, Use Absolute Deviation Around the Median. Journal of Experimental Social Psychology, 49, 764-766. doi:10.1016/j.jesp.2013.03.013
Rossman, L. A. (2004). Storm Water Management Model User’s Manual
Version 5.0
U20L-04 HOBO Water Level Logger
http://www.onsetcomp.com/products/data-loggers/u20l-04#