One of those technologies allows machines to discover by themselves the different states an industrial process features. Imagine a computer repeatedly fed with values generated by the sensors installed in an industrial process. Non-supervised machine learning techniques make possible the computer finds out the sensor data belong to, let’s say, three classes and moreover it characterises the classes. What the computer could not do is to name the classes, unless a human operator provides it a clue. That is what the operator does when he examines the computer outcome and assigns the names starting, stopping and running, just to follow the example. But in spite of this limitation, the non-supervised machine learning can be successfully used to detect faults or malfunctions that have never been observed in the past. This is what CARTIF did in the hydroelectric sets of a hydroelectric power station.
Hydroelectric sets are at the heart of hydroelectric power stations. Its role is to transform the energy stored in the mass of water retained by a dam into electric power. Each set is monitored and hundreds of variables are registered: electric current and voltage, temperature measured in the mechanical elements, in the refrigeration and water streams used for refrigeration, flows of water and air, etc. In our case, we have the values recorded along two years during which no fault was detected, and so we had not information about the possible faults. The challenge was to design an algorithm able to detect faults.
The solution developed by CARTIF is based on the SOM (Self-Organising Map) neural network, which is capable of non-supervised learning. The network was fed with all the available data and she was able by herself of discovering the possible states the hydroelectric set could present. The network labels the states in an arbitrary way and to give the correct names a human operator has to collaborate. However, this is not required to detect faults. Since the data used for training represent all the possible non-faulty states, any network input that does not fully fit with those states corresponds to a fault.
This case can be easily identified by checking the similarity between the sensors signals and the prototypes stored by the neural network. When this similarity is too low, it indicates a fault is occurring.
During testing stage, the algorithm implemented by CARTIF was able to detect an overheating twenty minutes before the plant supervision system raised an alarm. It is important to note that our system used already available sensors and no new ones were required.
So, while we wait for the day machines will rule over us, we may use them to implement intelligent algorithms to improve industrial process supervision with no need for high investments.
How to reduce structural conservation task expenses by implementing monitoring systems?
The structures are not everlasting. They are projected to play a role for a certain number of years. Thus, a wind turbine mast lifespan is about 20 years while in the case of a bridge it depends on the type and the material used. According to Guy Grattesat, metal bridge will have a lifespan of 40 years, 100 years for a reinforced concrete, between 15 and 20 years for those made of wood and about 200 years for masonry. Nevertheless, exceeding the life expectancy does not necessarily mean dismantling the structure. Generally, what we do is a more comprehensive monitoring effort and implementing conservation works if necessary.
Wire and accelerations sensor installation process inside the handrail
The European bridge stock is catching its lifespan and it is showing signs of fatigue. According to Eva Lantsoght, Professor of engineering at the University of San Francisco de Quito, “European bridges are old, but their replacement involves a great investment. Only in the Netherlands, there are about 3,000 bridges that could cause problems, being their refurbishment cost around one million euros each”. To mention another example, the first wind turbine parks (1984), whose technological emergence came in 2002 (according to the GWEC, Global Wind Statistics), have great needs for maintenance and the increase of these needs will have the same exponential curve, I can imagine, that its development has had in the past.
18 triaxial MENS accelerometers (sensors developed by CARTIF) installed equidistant along the span inside the handrail.
Having this background in mind, I would like to highlight the importance of implementing new maintenance strategies to reduce structural conservation task expenses. State-of-the-art monitoring systems, developed in recent years at increasingly low costs can be the solution. Usually in this type of structures, the procedure consists on the installation of a network of accelerometers to record the ambient vibration response of the structure. Using identification techniques, the modal characteristics can be estimated and be used to evaluate the structural integrity.
The idea is to know the structural behavior in operation conditions and to determine the valid rangeby controlling only a few parameters (frequency, mode and damping), being these parameters easy to evaluate and to record their trends along the time. Although the meaning of the word ‘monitoring’ refers to the capacity of the acquisition system to record certain values, it is also important to add the ability to process the data and to generate alarm codes, if it would be necessary.
In this aim, researchers require a deep understanding of the technical matter involved as well as a big experience in experimental techniques and data processing. OMA (Operational Modal Analysis) identification techniques are based just on the acceleration records but if some loading signal applied on the structure could be also record, additional information can be obtained using EMA (Experimental Modal Analysis “EMA”). Both approaches should gradually replace, or at least complement, the traditional visual inspections and static loading tests. Monitoring systems could be an emergent business for technology-based companies in cooperation with new maintenance standards or strategies by the infrastructure’s authorities.
In this regard, I can mention the success of low-cost monitoring system of Pedro Gómez Bosque footbridge (built in Valladolid) operated by CARTIF team since 2012. This stress-ribbon structure undergoes no linear effects. The collected data has been useful so far to understand the behavior under dense pedestrian loading, at different met conditions (summer or winter, strong wind…), etc. Analyzing this information, we know what it is normal and what is not, so we can establish an optimal operational range and determine when anomalies appear and why. In these terms, the structure is an ideal benchmark not only for structural monitoring but also for understanding mechanical problems, pedestrian loading and other engineering issues.
The first month of the year is coming and it is time for reflection. The year ended with us still having in our pockets the boarding pass of our virtual trip to a historic event for the climate, the United Nations Conference on Climate Change (COP21), held in Paris last December. And with its conclusions still ringing in our ears, we begin to understand that all of us, as citizens, have the opportunity to do our best to stop this global threat. And for this reason, the researchers at CARTIF step forward and get down to work, looking back for a moment, to remember the road traveled and, on that basis, define our environmental challenges for the new year.
2015 has been a time with lots of environmental milestones for CARTIF. Seven of our twelve ongoing LIFE projects have passed the middle of their execution and there are already very interesting results, with a huge environmental potential for the future (we will tell you in due time). Different waste management technologies, air pollution and Life Cycle Assessment smart tools have been some of our focus this year, along with other interesting international projects, which never forget to include environmental objectives transversely to the challenges they face.
And because the first month of the year is the period where people are filled with purposes, good intentions and To Do Lists, we have also proposed our own challenges, environmental challenges of course, and we want to share them with you.
To work ensuring respect for the environment through R&D should be a natural and routine practice, and we’ll contribute to it with our work whenever we can. We promise to reduce our carbon footprint in this new year, as well as our projects, and, consequently, moving towards environmental sustainability. Because, it is already known, the road is made by walking.
The wearables provide us functions such as view the time and do calculations or improve our capabilities. They are nothing new
Already in 1654, the Qing dynasty miniaturized an abacus on a ring. Today it is considered as the first wearable of the history. At the same time they began to use instruments amplifiers sense of sight and hearing, lends and trumpets. More than a century later appear mechanisms like a marine chronometer by Harrison. In the twentieth century, watches or cameras in vests and helmets are manufactured. Anecdotally, in 1961 Edward O. Thorp and Claude Shannon introduced a computer inside a shoe to cheat playing roulette. But, until 1972 we cannot speak about technologic wearables when the first digital clock was invented by Hamilton.
Concerning these last, today the technological hurdle is connect and measure the broadcast signal by the human body with sensors which can be integrate inside devices easy to wear: for instance bracelets, rings, watches, glasses andclothes.
In terms of functionality, it is focus principality to specialize and group the physiological measures and the vital signs in order to control, detect, and caution physical diseases (cardiological, diabetes, epilepsy) and psychic diseases (stress, anxiety, disorientation)
Although in other areas this devices could be interesting, in the health area could began uncomfortable, even stigmatizing. In times where medical records are so high degree of confidentiality and privacy, who yield to wear a watch that brand, shape or signals emitted, it is telling those around me that I am diabetic? Or in the case of several pathologies, do I have to wear a collection of colorful bracelets to measure related vital signs?
This factor may be one of the biggest problems when entering the wearable in everyday life, so to minimize the visual impact and ease of use are two fundamental reasons why the technology trends are towards devices deriving “merge” with the body (second skins, tattoos) and going a little further, that are integrated into the body. If you can wear the wearables inside the body is not strange if we think that the first external pacemaker, was built in 1957 and only a year after the first internal pacemaker was implanted. Today materials susceptible of being rejected by the human body are numerous and research in the medical, biological and technological fields come together every day new sensors and devices.
Another major inconvenience to the final launch of the wearable comes on the heels of the culture of data privacy and the feeling of control. The challenge is to demonstrate the paradox that the more we control the device, more independence have a device capable of detecting a possible attack, a stressful situation or a seizure (all of them are already), give us greater mobility , increase our ability to travel alone activities without fear of a sudden.
Besides offering the possibility of making calls to emergency services, provide enough medication alarms and storing useful information the next checkup time. Therefore, in the coming years it will be crucial to change the concept of “being controlled” by “security” to receive information in time so as to emphasize the visual discretion devices. This will make the wearables are normal in our lives because we already have the knowledge and technology to develop.
The World Health Organization recommends reducing the red meat consumption and avoid processed meat intake, after the report of the International Agency for Research on Cancer (IARC).
For many years, it has been recommended to consume the red meat in moderation. So far, these suggestions were associated with risk of cardiovascular diseases, diabetes, obesity… However, for the first time, its massive intake has been related with the horrible cancer. This piece of news has spread panic in the society.
According to experts, potentially carcinogenic substances ingested with meat are heme iron (which brings its characterisitic red colour), nitrosamines (substances created when meat amines bind with nitrites and nitrates, used as preservatives), and certain chemicals substances, which are formed in the process of cooking.
What news for our country! Spain is mainly carnivore and the best hams of the world are produced here. According to the Report of Food Consumption in Spain, published in 2014 by the MAGRAMA, the Spanish consume every day about 50 grams of red meat and about 32 grams of processed meat.
Are we at risk?
At the beginning, red meat is not a problem in our country. On the one hand, the consumption does not exceed the amount recommended by WHO. On the other hand, there is not evidence between red meat and colon cancer. However, the situation changes with processed meat. In this case, according to WHO, there are solid evidences of its relationship with this kind of cancer. In addition, they warn that the risk increases with consumption and portions of 50 grams per week mean high risk.
In Spain, the processed meat consumption is almost five times higher than recommended amount. This could increase the danger to suffer a cancer. But it is necessary to know the significance of absolute and relative risk, in order to avoid mistakes. The best way to take in these concepts is with an example:
50 of 100 CARTIF researchers drink a coffee for breakfast. The other half choose a cocoa drink. Throughout the day, 5 of the researchers who have taken coffee and 2 of those have taken cacao feel dizzy. In terms of absolute risk, we would say that 10% of the first group and the 4% of the second have dizziness. So the absolute risk of feeling dizzy after coffee is 6% higher than after cocoa. In fact, the absolute risk is small.
On the other hand, the relative risk is also the difference between the percentages of both groups. Therefore, it is 2,5 times more likely to feel dizzy if we choose coffee rather than cocoa, or the risk of dizziness with coffee is 60% higher than with cocoa. Of course, using relative risk, percentages increase creating a bigger impact. Normally, media use it in their headlines in order to create more interest and excitement, and this is just what has happened wih the WHO information. The headlines have emphasized the relative risk (18%), in spite of the fact that the risk of developing cancer by eating red meat is only 5%.
However, stopping eating meat or eating raw meat are not the best solution. According to WHO, the raw meat can cause infections. Moreover, according to a study published in the journal BioMed in 2013, the relationship between meat consumption and mortality has chart ‘type j’, so it is worse not eat meat than eat a small amount. Meat is a fundamental source of nutrients, proteins, iron, zinc, vitamin B, vitamin A and essential fatty acids.
The meat sector is aware of the negative influence of this kind of information and how it could affect the sale of their products. In the CARTIF area of Agrofood, we are cooperating with manufacturers of pork sausage to improve the nutritional profile of these products. Getting healthier meat products without its losing organoleptics properties is one of our challenge in CARTIF.
Firstly, it is necessary to insist on the fact that, talking about scientific and technical progress is to do about economic and social progress. This is like this since homo sapiens appeared on Earth. But it is absolutely obvious since the late Middle Age, when science started to take off and to apply the knowledge of it to know human needs. The Industrial Revolution took its first steps in the eighteenth century after achieving the use and control of natural sources of energy. Then, the scientific and technological progress has continued unabated. Without it, our lives would be much poorer and the welfare state would be an entelechy.
When so many politicians and all kinds of demagogues filled their mouth promising to improve living conditions of citizens, probably they are not thinking that the basis of their claims, if they really want social progress, is linked to scientific and technical progress.
Unfortunately, we have heard some poor references about this only during the last Spanish electoral processes. It must be that it is a subject without influences on the electorate. This is the first and greatest obstacle that society has to lead in order to start to consider essential that the social welfare is closely linked to advances in science and technology. Opinion makers and those who cause these several opinions have a beautiful but difficult task.
The last major financial crisis, which began in 2008, led to the reduction in research and innovation resources. It is true that they were not the only nor, in the short term, the most painful, but with this reduction is impossible to improve in the short and medium term.
It is necessary that those who govern us and who aspire to do, understand the importance of research and innovation, and try to efficiently undertake the necessary and constant actions to be a social value accepted, avoiding the reduction of these resources.. It is necessary to get a suitable atmosphere for the efficient use of public and private resources, in order to be a richer and more solidary society.
