Player is loading...
Loïc Puthod - An open-data acquisition toolchain for AI applications
Loïc Puthod, Centre de recherche Crem
Monday, Nov. 15, 2021 · 1:59 p.m. · 11m 02s
In many research projects, meteorological data is required with a fine temporal granularity at specific locations. As the measurements freely available in Switzerland are not always fulfilling these requirements, Crem with the collaboration of Idiap has developed a system for acquiring this data. The system consists of a microcomputer allowing the necessary meteorological measurements to be taken by dedicated sensors before sending them to a server. The measurements can then stored in a database at a defined temporal granularity for diverse applications. A prototype of such a weather station is installed on the roof of the Idiap building and continuously measures temperature, humidity, and direct and diffuse irradiance. This data is used in an on-going research project to control the blinds as well as the intensity of the electric lighting fixtures. The acquisition system in its actual state allows all kinds of measurements to be taken, its only limitation is the existence of a sensor capable of measuring the desired quantity. Further prototypes can therefore be developed for other kinds of applications, from e-scooters to home automation.
Embed
Transcriptions
Note: this content has been automatically generated.
00:00:07
yeah
00:00:08
pop up to her
00:00:19
huh huh huh huh
00:00:28
with uh
00:00:41
yeah
00:00:42
yeah
00:00:59
yep
00:01:07
yes yes
00:01:21
oh
00:01:25
oh okay so we'll start again i hope you had a a nice lunch so now it's
00:01:31
time for the last two bonuses presentations the first one is from a likely to program here
00:01:38
ah what she's going to introduce himself and he's going to
00:01:41
talk about an open data acquisition to chain for a high applications
00:01:56
and i hope it works okay perfect so i think you're wrong i would often everyone and welcome
00:02:03
to my presentation about a hour open that that uh to gina oh but not that occasion system
00:02:09
detection system um first um um i do i work at the crime in um
00:02:16
mainly on simulation tool and um and the application uh uh
00:02:21
systems like yeah like the one uh i will present you
00:02:26
nothing uh why do we need um the data acquisition system
00:02:31
engine of us um for supper project we
00:02:35
need to uh validate a simulation models or we
00:02:39
need to we need like that that to a regulate the buildings or or to do things right
00:02:44
the the issue comes when uh those uh such the data is not available which is which can
00:02:49
be the case um because that that doesn't exist or because the data isn't a a free to use
00:02:56
in that situation you can
00:02:59
you can do two things it fine of how that doing what you
00:03:03
were doing but you can also uh take measurement to create your own database
00:03:08
which is uh in the you can do this with our uh ike isn't that that system
00:03:18
so basically our cousins that i system can measure and and take that
00:03:23
are uh for almost every every kind of thing that can be measured
00:03:28
but for his presentation i will present you uh the
00:03:31
weather station that we uh made out of this uh system
00:03:36
so we take the three main and measurements the first one is
00:03:40
the irradiation so with the with the sense or a gift five
00:03:44
we uh did yeah exactly uh we take 'em diffuse and the tile irradiation
00:03:51
the second one is the temperature nudity since all so protecting me and a comforter and that uh
00:03:58
the last one is the when since our that did us uh the wind screen and the wind direction
00:04:04
all those uh those uh since are are connected to or i mean the p. c. which is
00:04:08
a raspberry by and is the the raspberries then connected to the network with a three g. antenna
00:04:15
that's a bit for the heart for the hardware now uh work on the
00:04:19
software in the protect want every since or would uh communicate with the same protocol
00:04:26
but it just you we are in real life so we have three since ours in three different protocol
00:04:31
that we need to man uh to to understand and to read in um in order to take our present
00:04:37
the um to the opportunity to since or or communicating i quite the right see
00:04:43
which is a great protocol because it it's uh it makes us able
00:04:47
to have more than one son saw since are on the same wire
00:04:52
in fact all since or has a it's it's own address and when they can we don't
00:04:56
need to take measurements you just asked the measurement to do right since always the right targets uh_huh
00:05:03
the second one the the winter nick so these um hum distance or communicating arrest two hundred thirty two
00:05:10
this is i is is simpler it it sent each second string with uh his current values so
00:05:17
we need a we need to be able to read those values process them and then send them
00:05:22
and the last one the the f. five is an analogy each uh since our it means that we
00:05:28
basically measure the voltage that the output targeting ah the
00:05:32
that's the end of the outputs the tote would so
00:05:35
and then we uh we compared to does the voltages to uh evaluation they'd
00:05:42
then that's for this part here then the second the the second part is when the red raspberry
00:05:49
pie read all the data processor that also put them in a in a defined um in defined format
00:05:56
and then the rest there like send all the uh all the all the data to the included the server
00:06:02
imputed to server is a server that allows us to uh send and receive data
00:06:08
on a specific topic which interested because each or whether session will
00:06:13
have it it's it's on a topic to uh send the data to
00:06:18
and then very very last that the the main sarah is reading all that that that that the result station
00:06:24
uh are sending and push all those that are in the digit database that is a store in a camper so
00:06:34
the next point how will the um so you one of our
00:06:38
weather station the one that is uh here up up on the roof
00:06:44
and then uh to uh application of our uh already existing yeah weather stations
00:06:51
so that was a station of martin
00:06:55
mm um something's the um we are measuring or ah okay are
00:07:00
currently the irritation and the temperature and humidity uh of the place
00:07:05
as you can see this was uh stationed was built to be autonomy so it has a uh well i'm
00:07:11
sort of one and we is a battery in the box that he that accompanies the that out of your
00:07:18
because the trend of the station is around one what's when
00:07:21
nothing when it's just taking measurement measurements and sending the the data
00:07:26
and it can rise to fifteen what when the um the b. f. five
00:07:30
cents or start to two hits because uh when the temperature goes to down
00:07:35
the the since our need to to keep 'em to to
00:07:38
say warm so uh it's mounted with a little electric for it
00:07:44
um one direct use of this station is the lighting regulation in
00:07:51
this uh fees b. c. which is you know cases jones of his
00:07:55
um they yep uh created and trained an algorithm that was able to
00:08:01
predict the you mean ounce on the on the desk of the worker
00:08:06
in function of the certain position and the current irradiation
00:08:11
if the algorithm defined that there is too much light
00:08:14
then it can it can move the lines to to get the light or if it's it's there
00:08:20
isn't enough light down on the electric light and uh and improve the come forth of the uh worker
00:08:28
the end of this project is of course to um uh
00:08:32
to to optimise the and that the electric energy the consumption
00:08:38
the and you get to the area the and eclectic
00:08:44
uh one of the application is an addiction on the a good keys project so basically the idea of
00:08:50
the of this project is to be able to predict
00:08:53
the distributing network consumption in function of a meteorological data
00:09:00
so uh we install it three weather station so why one in what they won in dublin one instill
00:09:06
but yeah the idea is to collect data and then at the
00:09:10
end to um to try to to train an algorithm i guess
00:09:16
to um you have to be able to predict when the the the the with uh the with the um
00:09:24
they would uh if you'd see the heating system must be turned on or uh in advance or not
00:09:32
right
00:09:35
uh i wanted to do um hum a live demonstration but
00:09:40
as we are a lot now allow liven you'd you'd and um the
00:09:44
the the the script isn't already on the on this computer what it's
00:09:48
the properties for me to invite you to come and see us up
00:09:51
our um our showroom just outside of this uh of uh does what
00:09:59
but at the
00:10:01
uh i will finish wheeze wheeze does a little slide uh for confusion so we uh we built a um
00:10:11
i that i question the system that is on those
00:10:15
examples use to uh collect a bit origin meteorological data
00:10:21
but it can also very easily use to measure acceleration su to pressures throw
00:10:27
and uh in reality everything that is easily ms
00:10:31
in a miserable and um and yeah but miserable
00:10:36
um the uh some example that force on in those are applications
00:10:41
we are currently mentoring some um batteries in the electrical a scooter
00:10:47
or uh or databases can be used to
00:10:50
uh uh to validate um the simulation models
Conference Program
Workshop introduction by Jérôme Kämpf & Henning Müller + Keynote speech Pr. Dr. Guglielmina Mutani - Place-based energy models for sustainable cities
Pr. Dr. Guglielmina Mutani, Politecnico di Torino
Nov. 15, 2021 · 9 a.m.
333 views
Q&A (Keynote speech: Pr. Dr. Guglielmina Mutani)
Pr. Dr. Guglielmina Mutani, Politecnico di Torino
Nov. 15, 2021 · 9:49 a.m.
Roberto Boghetti - Simulating District Heating Networks’ hydraulics with residual physics neural networks
Roberto Boghetti, Idiap Research Institute
Nov. 15, 2021 · 10 a.m.
Q&A (Roberto Boghetti)
Roberto Boghetti, Idiap Research Institute
Nov. 15, 2021 · 10:13 a.m.
Amine Weibel - Machine learning forecast for photovoltaic power stations (PV) with spatial interpolation on the COSMO-7 model. A real ‘use-case’ on the Czech Republic
Amine Weibel, HES-SO
Nov. 15, 2021 · 10:16 a.m.
Q&A (Amine Weibel, HES-SO Valais-Wallis)
Amine Weibel, HES-SO
Nov. 15, 2021 · 10:27 a.m.
101 views
Dr. Giuseppe Peronato - HeatLearn: a convolutional neural network to assess the building energy consumption at regional scale from open land-use maps
Dr. Giuseppe Peronato, Idiap Research Institute
Nov. 15, 2021 · 10:30 a.m.
Q&A (Dr. Giuseppe Peronato)
Dr. Giuseppe Peronato, Idiap Research Institute
Nov. 15, 2021 · 10:43 a.m.
Pr. Dominique Genoud & Jérôme Treboux - Integrating machine learning and AI in practical energy applications : a review of these last 10 years activities at the Institute of Information Systems
Pr. Dominique Genoud & Jérôme Treboux, HES-SO Valais-Wallis
Nov. 15, 2021 · 10:46 a.m.
Q&A (Pr. Dominique Genoud & Jérôme Treboux)
Pr. Dominique Genoud & Jérôme Treboux, HES-SO Valais-Wallis
Nov. 15, 2021 · 10:59 a.m.
Dr. Marco Tognoli - Fuel-saving potential simulation and estimation when optimally predicting the supply set-point conditions of a District Heating Network with detailed system data-driven modelling
Dr. Marco Tognoli, Politecnico di Milano
Nov. 15, 2021 · 11:31 a.m.
Q&A (Dr. Marco Tognoli)
Dr. Marco Tognoli, Politecnico di Milano
Nov. 15, 2021 · 11:44 a.m.
Pr. David Wannier & Jean-Marie Allder - Research advance in Energy prediction for electrical vehicles and green mobility hotels, the results of the eVIP project
Pr. David Wannier & Jean-Marie Allder, HES-SO Valais-Wallis
Nov. 15, 2021 · 11:46 a.m.
Q&A (Pr. David Wannier & Jean-Marie Allder)
Pr. David Wannier & Jean-Marie Allder, HES-SO Valais-Wallis
Nov. 15, 2021 · 11:58 a.m.
Dr. Kavan Javanroodi - Extending the concept of energy hub to facilitate sector and spatial coupling
Dr. Kavan Javanroodi, Solar Energy and Building Physics Laboratory (LESO-PB) at EPFL
Nov. 15, 2021 · 11:59 a.m.
143 views
Q&A (Dr. Kavan Javanroodi)
Dr. Kavan Javanroodi, Solar Energy and Building Physics Laboratory (LESO-PB) at EPFL
Nov. 15, 2021 · 12:16 p.m.
Pr. Pierre Roduit - Prediction of the electricity flexibility of households using Machine Learning
Pr. Pierre Roduit, HES-SO Valais-Wallis
Nov. 15, 2021 · 12:18 p.m.
Q&A (Pr. Pierre Roduit)
Pr. Pierre Roduit, HES-SO Valais-Wallis
Nov. 15, 2021 · 12:32 p.m.
Loïc Puthod - An open-data acquisition toolchain for AI applications
Loïc Puthod, Centre de recherche Crem
Nov. 15, 2021 · 1:59 p.m.
Q&A (Loïc Puthod)
Loïc Puthod, Centre de recherche Crem
Nov. 15, 2021 · 2:10 p.m.
Cédric Mugabo Serugendo - EnerMaps: The open data tool empowering your energy transition
Cédric Mugabo Serugendo, Centre de recherche Crem
Nov. 15, 2021 · 2:16 p.m.
&Q (Cédric Mugabo Serugendo)
Cédric Mugabo Serugendo, Centre de recherche Crem
Nov. 15, 2021 · 2:27 p.m.
Recommended talks
Workshop - Les services à la personne - AlpICT
Prof. Jean Hennebert, HEIA-FR
Nov. 24, 2015 · 2:47 p.m.
Le projet européen Khresmoi : les résultats
Henning Müller, prof. HES-SO Valais-Wallis (unité e-health)
June 6, 2014 · 10:50 a.m.
102 views