Just before starting, probably if We talk about Nuclenor
because many do not place us
but it is true that if we say that we are the owner and the
company operating Santa Maria Garona, for sure you have heard about
us for everything that in recent years has been talked about
and today we are still talking. My perspective anyway for this
presentation is what we are learning, our trip and also
from the perspective of sector utilities that are suffering the
generation rates and therefore is trying to smarten up to
reduce costs and optimize resources.
First of all introduce ourselves. Nuclenor, 50 percent Endesa and 50 percent
Iberdrola. We manage an American designed plant, General Electric, which
We call a reactor water boiling, 466 megawatts
electrical and connected to the network in 1971. The data here is from
the last ten years we were operating the plant. We stopped in
December 2012 for various reasons, but hey let's simplify economical
matters,
what I do want to show here because then I will link a little to the sector:
we see there are three indicators that are representative of operation which has
had Garoña beyond demagoguery that we can apply, on one hand the
electricity side. We generated from 2002 to 2012 where
you can see a stability indicating a bit reliability of the plant,
we have the capacity factor
meaning the energy generated from or compared to that
where theoretically we could give Garona has always been above
90 percent,
even the green colour you see there, that's the average capacity factor of guano plants
and for which you are not tucked this world, guano is the
World organizer,
is a group of nuclear generators worldwide, and
one that is also important for us is a monitoring matrix
our regulator, in this case the Nuclear Safety Council, established in
Spain in which we compare against a set of values or
indicators and in which the maximum note say, without going into great detail, it is
those green colors you see there,
Garoña is the only plant that from this system has been established
in green and ultimately what comes to say in this slide
is actually the numbers were good and are good but still
we stopped.
Before our present situation, just for
your knowledge in July 2013 we
officially entered in a cessation of operation, but from that date
our owners, who are Endesa and Iberdrola, they asked us better to present
the application for renewal operating license of the plant and
now we are waiting for the technical opinion by the board.
The continuity of the plant has social-political reasons we all
know, but there are other economical aspects that are obviously above
but not only for Garoña, also many generation plants
now in this country and outside this country, and there is one technical
aspect.
I'm going to focus especially on this technical aspect,
but on the technical side that can help us and what we have
deepened to help to make more competitive plants. Probably
all of you are wondering what we have done since we stopped so far,
because it is the question from most of the people, and it is the question that
people do when I'm in different forums. We have the plant, we are doing
mainly three things: first, maintenance. We are keeping the
reversibility of the plant to make feasible
any decisions.
We are or we have presented to the council nuclear safety all our documentation
they asked among other things, which for us has been a milestone, which is
the inspection of the 100 percent of our vessel that has always been in the
newspapers and where it has been shown you can successfully operate
beyond 40 years, and other thing It has been important for us is
to use this time to get ready for what would be an extension of
life to 2031, with many challenges managing the ageing of the plant and the
assets, that here we will not address, but that is part of our
strategy, and the use of new technologies such as
monitoring in supporting utilities, in reducing costs.
I put this chart because I firmly believe that it is important for me,
90% of success is preparation, and in this case it is
one of the things we have done, also from the point of view of
our sector is to take that bag you see there,
We have gone to the United States, we have seen a little of what the sector in
America is doing in the utilities, primarily nuclear but
also seeing other utilities who are suffering,
we have made a benchmarking with Exelon, for those who do not know it, Exelon
is an organization that manages 21 reactors in the United States and is
an example of a management model, we have also been in Luminant, in Texas, which is
probably, if you look on the internet,
the reference in monitoring systems for nuclear power plants,
They have monitored the Comanche Peak plant having two
units 200 kilometers from Dallas, and having a
set up a center where they are able to monitor how the plant is
functioning and critical assets.
We have been doing benchmarking in Oil & Gas with companies both
aeronautical, in the sense that they are regulated companies that are
suffering a lot at the level of costs and so for us it is important to be
regulated and manage these costs. And then we have participated in
different forums. We present our strategy to the American Nuclear Society
and we participated, we have been fortunate to be invited
to participate
in meetings from the American Energy Department, where utilities have been treated
in a sectoral level,
and just so you know where we're
in utilities. These four plants, each one is in a different place,
Vermont Yankee central is between Vermont and
Massachusetts, US,
Kewanee is in Texas, Pilgrim which it is in Massachusetts and of course
us, Garona. What do they have in common? They have in common that all these
plants had large numbers, capacity factor above 90
percent, but still the two below, Vermont Yankee and Kewanee stopped
despite having license approval, approved by the Nuclear Security Council
for 60 years, and Pilgrim already has an end date and Garona
we are where we all know.
This is generating a big debate in utilities here and in the sector
in United States and what it is doing is to give work even
to the American Energy Department because you are seeing that companies can not
support those low prices for energy and what it has done is
focus on four aspects, but which interests us, and we can see
there, is to work in Technology Systems applied to the generation
to gain productivity. And there are, we are in the monitoring day, as the
Monitoring is vital for survival of our actions.
In that sense I think that in all your companies probably have heard
to your CEO and always thig one thing that is scanning.
Another thing is then if it is carried out or not, but everyone talks about scanning,
and in this case we
also try to go to that scanning because certainly
a change of technology has to be done to improve.
Basically what it is being done in the sector is on one hand, we must
guarantee income, that is no more that always taking connected
plants, based on availability and reliability of the
plants, and on the other hand a system that is intensive in human resources and
in stock, because we have to go to optimize our
resources. And all of this only right now
we believe there is a magical number used in the sector, I was before
talking with a partner in the industry and he said another even larger,
but well the one we have managed and handled is to achieve to
reduce 25 percent in operating costs and maintenance of the
plants, without touching the unique value it is important for us, the
security.
Then we have to work on reliability, we need to work on improving and in
process optimization which ultimately improve the productivity of our
means. We are also based in what we have seen,
We have designed a system, we have seen that
monitoring is not monitoring by monitoring but we have to go
to intelligent monitoring, in IT called predictive analytics.
Historically in maintenance
We have moved from a maintenance based on to a condition maintenance
, to a maintenance focused on reliability,
but these three concepts, it has been a progression, now the improvements are
asymptotic and therefore we must go for something else
and the sector has found the opportunity in the software, the
software is all day analyzing in a
continuous way, to analyze the parameters we have in our plants. Later I'm going a bit more deeper
in this analytical prediction, just at the moment
we have said, we have our assets, we have our processes,
we have to manage these data that in most of the cases we already
have. It's funny how if you so think of your facilities, depending on how they are,
but if you see the level of data which is increasing every day thanks to
our intelligent systems and it is also
lengthening the gap, with the actual use we do of data.
That's what predictive analytics tries to avoid. We also
deliver to each of the people involved in our
plants, the data they really need.
I'll explain the example of my plant, but what happens in all plants
in a worldwide level or in United States at least, that is my more direct
relationship.
We have engineers call systems engineers who are engineers
controlling the health status of systems and components. This
people, their mission is to monitor the different variables, different
Data from these teams and from them to test if there may be a
problem or not and take action.
60% of the time this people is collecting data. We have data from
maintenance,
we have data in operation, we have chemical data or data from
other processes we have, but really these data do not flow as
it should in the organization, many are sometimes fiefdoms.
Then the first mission, and that is where you are going to see what we have done with
CIC, is to achieve an infrastructure that centralizes this information and that 60%,
which is not only a fact of Garoña nor a Spanish central data, is data
you can see in Epri
as actual data average from the systems engineers systems dedicated to this. We have
to eliminate such unproductive activities and that gives us the software,
which is capable of feeding from all data available in the plant,
through this software we are able to generate models, this is not science
fiction, it is in plants already, and generates models of behavior that
will be compared always as model real vs variable and when you have a gap you
will tell the engineer; there you have, I have a gap, begin to study why, but there is when
I begin to really pay my engineer, I dont have my engineer to catch data and
that is a change of philosophy We have to do and we're doing
and that's what comes to explain this process of what we call
intelligent online monitoring, the jump to predictive analytics. This
that I am presenting, we presented in
2015 at the American Nuclear Society and what is our model, we see
it is the model of a nuclear power plant to which we have to go now, in which
on one hand we have to contribute to the plant reliability and that is what
that I said, that is, we have a lot of data on the computer, lets put
the software to work to do the sieve of what is important or is not
important.
On the other hand our critical assets transformers which are mainly
generators, emergency diesel, our generator, our turbine, expensive systems,
we have to lengthen his life as much as we can and the best way is
optimize through monitoring it. Now is
being talked and is already implemented what is called the
prognosis maintenance that is, we have a lot of operational
experience because our industry has sued us and we have what we call
fault signatures. These fault signatures, we're getting them
in our databases and from the actual plant data
we are able, because it is being made in transformers, diesel machines,
to identify potential problems and the best moments
to intervene in it.
The difference is what we have in the plants, and if you just look,
five years ago all plants thermocouples were watching the
temperatures in bearings, this is an example of pumps, they had
a vibration which bears the predictive maintenance team went with his
laptop, grabbed a data and from there the month. In many
cases it was looking and comparing, but when you get the alarm you probably
already had the fault in the bearing and you had the fault in the component. What
we are talking about here is before reaching that failure problem
when you have a deviation, pull the alarm, and that's what we talked about for
improve our reliability, and that increase in revenues for plants.
Now let's talk about the other party: productivity. We are intensive as saying
before in human resources, we have many activities, then we will see one of
which have been automated with CIC, which it has a lot of human administrative work
of catching data, interpretations, export data to
Excel, macros ... that can be automated and reduce the time our people
really dedicate to that.
We have rounds and the term mobility, in which we catch data, left in a
separate database, and then if we want that information that is not
compatible with the rest,
we have to go one by one to those fiefdoms to go looking, as
centralize information. And the third, but hey that's no longer an issue
much of our sector, use computer vision in the
meaning that, especially in places where you have radiation, as it is a
process of a nuclear power plant,
if you are already able to have cameras they detect leaks without having
rounds,
that's what you gain from your control installation. As stated objective is the
25% reduce our costs operation and maintenance. What have we
worked with CIC?
The first build an infrastructure that is able to centralize all
our data and we are able to analyze it and make it available to
all members of our organization we want. Here there is
an important thing that I have seen and I think it's
very common, from information systems, IT was usually the
department of information technology who was concerned. All digital
systems, all process control systems of plants or
factories, the instrumentation and control departments, they are departments
more of a plant, and between these two there is a huge world. But it is not going
to be another way but to converge the operating technologies with
information technologies, and on that any
manufacturer who wants to go to a digitalization will have to take the
leap.
We will have to work on mobility to transform in real-time the data
that our rounds are seeing. Ultimately efficient information
management and with them also with CIC
We have worked and are working in pilot projects that allow us
to learn the use of these technologies.
Only one question that I always liked in one of those meetings
I was in the DOE, MS was there. MS is a
company that it is working a lot with us, it has made us response times
for our pressure transmitters, but well, what I was saying was, what
is the most important treasure the plants have?
the data. You can not have an idea, each one in their sector will know,
but we have data to give and take, and often the question is why,
the use or not use of it, that's the question.
Us
with IDBox we have always worked with CIC in the development of our system
information management, but in 2015 we migrated to the platform
IDBox the last in the sense to allow us to be willing to
step into centralize all the information, shared and
distributed in the way it is interesting for us and also the new tool
allows us everything shown in the previous presentations,
about analysis, which is important with everything I came
saying on predictive analytics. There are computer data we have
but hey, the important thing is that we are creating infrastructure for
information flow. What I said about task automation
chores.
This is an example that we have developed. Medidas ambientales is
a company of which Nuclenor owns 50% and the other 50% is Eulen, but
obviously works with us, because we are
owners.
We manage, control, and monitor many temperature points
in the Ebro river, the Ebro is the river that we use the
water for cooling. As well as the control of discharges of our
purifiers and we have to make reports and send them to the
Ebro river federation. What did we do? and what do we do? because we are now
in the validation process because obviously there are official reports and
we have a long period for validation. We have a person in
our environmental department who went monthly and picked data
from our registers because they were located at many points away from the
plant, poured on your computer, in that computer then he was making a
data export to Excel with different configured macros and
get the information and then write the report we sent to the
hydrographic confederation.
All this, valued resource,
imagine that right now, we have to send all data from
different positions via telephone, then we have them on our server,
We convert, analyze them and make the report, and give the report to the
person, thereby productivity improvement is clear.
The other aspect we work now are the operation rounds,
we have always people around the plant taking data from
local instrumentation,
besides instrumentation we always have to our computer
process, which is basically ninety-odd percent information from
important parameters that are all computerized,
and all we do is use the rounds for vision, visual inspection and
collecting information in different points, which now is in an independent
data base, a standalone application, which we
are not able to gather with IDBox, what we are doing are rounds
with Fieldeas, which is the other CIC product for mobility,
preparing to share the information and then export this
pilot project to all that are the others, which is maintainance to
get information. Ultimately use mobility technology in
Real time for our shift leaders, our people operating in decision
so they have all the information available and therefore making decision
is much more conservative and more efficient.
Finally, I would like to at least modestly, what
we have learned, there are four points.
These three who come here and the next we'll talk then. In the three who you see
It is important and is a recommendation to systematize
management innovation technology, I mean, if you
ask the different departments in your companies,
the operation has a vision, the instrumentation has another vision,
electrical maintenance has another vision and to say the least people in
computing.
Therefore the integration of these worlds of IT and OT is fundamental and
sistematize it, I think that's fundamental, manage it through
a visible head that is the one that leads the way. The second, the partner, this is not
that easy, this digitalisation, what everyone's talking about but in my opinion
It is not as easy as to say pum! there you have it, so the
partner
of your choice is important, in our case I think CIC gives us
flexibility, gives us first proximity that requires this type of
projects and something that is important to me, it is prototyped or go to demos, better
to fail early and fail cheap than to do large projects which are sold
internally very well at first, but the costs for
organization are very large.
Prototype, spend and learn to fail, learn the problems but always in
my opinion from the point of view of expenditure control. And finally, I was
once
at a conference of this lord director of MIT and co-authored of a
book called "Second machine age" and goes on to say in his preface, and
you can see on Internet if you search that reference,
when the second revolution started, allt he utilities installed the electric motors,
but do you know when the productivity of the facilities did really improve?
It's funny but so,
thirty years later and it matched when all managers
were retiring heads. I mean, what is important is not the technology,
the important thing is to look for productivity, our workflows and I have
seen on my plant, where you say: I have this flow to a work order
now use my technology but adapted to my flow, adapt the flow
to what the technology is capable of and then you will improve the
productivity.
Thank you very much.
No comments:
Post a Comment