5 Trends Worth Watching at Virtual electronica 2020

The virtual electronica (Nov. 9–12) will focus on major technical trends in the industry. Show organizer Messe München highlights some of these trends below in the areas of embedded systems, e-mobility, digitalization through the pandemic, and smart energy.

Embedded systems

Embedded technologies are indispensable for dealing with socio-political challenges such as climate protection, demographic development, or increasing resource efficiency. In key branches of the industry, they also form the basis for most innovations. Embedded systems usually operate with a high degree of autonomy. That applies to recording (sensors) the environment, interacting (actuators) with it, and communicating with other embedded systems or computers. The required functions are increasingly provided by artificial-intelligence and pattern-recognition algorithms. Data processing takes place in a central data center (cloud) or, more and more frequently, at the point of origin itself (edge computing). The latter is particularly suitable for time- and safety-critical applications but also helps reduce the huge amounts of data to be transferred. In a largely automated, adaptive production environment or in autonomous vehicles, direct processing of the data in the device even proves to be essential, for example, to avoid impending machine failure or accidents. It does, however, require “intelligence” that is specifically tailored to scarce resources. For example, Artificial Intelligence for Embedded Systems (AIfES), from Fraunhofer Institute for Microelectronic Circuits and System, enables intelligent microelectronics and sensors to be developed that require no connection to a cloud and are capable of learning on their own. AIfES contains a fully configurable artificial neural network (ANN) with a feedforward structure, which makes even neural networks possible. Because processing with AIfES takes place “offline” on the device, no sensitive data has to be transmitted, so privacy is ensured. Preventive and seamlessly integrational security solutions are required to protect the networked and increasingly complex systems against hackers. And that applies not only to companies but also to politics. Just think of attacks on critical infrastructure, for example, in the energy sector or connected car applications. However, classic cybersecurity approaches such as anti-virus software or patches and updates do not work for IoT devices. Security requirements for embedded systems must therefore be taken into account right at the development stage (security by design). Transmission should be much more secure in the future via 5G. However, networking machines and devices with the new mobile standard will gain momentum mainly due to its ultra-low latency and high data transmission rates. Energy consumption will also be much lower. The Gartner analyst firm expects the 3.5 million 5G IoT nodes this year to rise to 49 million by 2023. From 2023, embedded systems in connected cars will then account for the lion’s share. Until then, outdoor surveillance cameras will be the main growth drivers. 5G will, however, also be used in the future in sensor cloud systems to monitor highly dynamic production processes more thoroughly and control them adaptively.

E-mobility

Electric drives, digitalization, networking, and autonomous driving are making up more and more of the electronics in cars. As soon as 2025, electronic components are expected to account for 35% of all material costs. Electric mobility, along with digitalization, networking, and autonomous driving, will drive the number of electronic components up significantly. Electronics and software already account for most of the innovations in the vehicle sector. According to the study “Computer on Wheels/Disruption in Automotive Electronics and Semiconductors” by Roland Berger, in the coming years, the portion of electronic components in the bill of materials (BOM) is expected to increase from the current 16% in partially electrified luxury cars to 35% in fully electrified premium cars. For a premium combustor, the BOM currently sits at about US$3,145. The Berger study claims that by 2025, this price could more than double to US$7,030 in semi-autonomous electric vehicles if the share of semiconductors increases from the current 25% to 35%. About a quarter of this cost can be traced back to digitalization. And the electrification of the powertrain is the cause of more than half of the cost increases. For autonomous driving, computing power and sensors account for the increased automobile electronics cost. These “computers on wheels” are having a dramatic effect along the entire value chain, leading to a redistribution of the roles of all parties involved. Manufacturers (OEMs) will need to commit significant resources to module integration, semiconductor companies will increasingly develop into software providers, and software suppliers will participate in the entire value chain. In light of these and other drastic changes in the automotive industry, the study indicates a significant need for action from all parties. This starts with working out possible competitive advantages arising from the changed distribution of roles, continues with the search for new partnerships, and ends with innovative approaches to the procurement of electronics and semiconductors.

Digital push from pandemic

Like just about everyone else, the medical technology industry and the health-care system have been battered by the coronavirus crisis. At the same time, though, the crisis has lit a fire under the industry, generating the momentum it needed to carry out its long-overdue digitalization. The pandemic has driven demand for medical technology into the stratosphere. But not everyone is jumping for joy about this development. Some companies are knocking themselves out to meet the overwhelming demand they face by their deadlines. In one reflection of this demand, Infineon announced a contract covering 38 million MOSFETs for ventilators last April. Companies like Analog Devices and Maxim Integrated find themselves in similar situations. They have also had to ratchet up the production of components for medical technology on short notice. In contrast, the clear “crisis winners” include pressure sensors, airflow sensors, temperature sensors, and radar sensors used for people screening in rooms and contactless gesture control. But the overall picture is certainly not so rosy for a portion of the medtech industry. Nearly 60% of companies are expecting to experience a double-digit decline compared with last year, according to a Spectaris survey. But the distress that has arisen about the availability of medical technology products during the ongoing Covid-19 pandemic shows just how critically important the medtech industry is around the world. Political leaders will ultimately do everything in their power to strengthen this industry and prepare it to face the future. In Germany, a committee composed of members of the country’s governing coalition has earmarked €3.9 billion in a plan called “Future Program Hospitals” — if the German Medical Technology Association had its way, these funds would primarily flow into medical robotics and digitalization technology for patient care. One other change has also clearly emerged: The coronavirus crisis is giving the industry the boost that it needs to carry out the long overdue digitalization of medicine with all of its exciting new ways to prevent, diagnose, and treat disease. In January 2020, about 1,400 physicians were conducting video appointments with patients. Today, 10× as many are doing so. Nonetheless, many people are highly skeptical of digital medicine. And not just for much-discussed privacy reasons. Many of the respondents in the Continental Study 2019 did indeed acknowledge the chances created by video appointments, AI diagnosis, and nursing robots. But they also cited the risks significantly more frequently. A globally unique, fully connected operating room at the Innovation Center Computer Assisted Surgery in Leipzig, Germany, shows just what state-of-the-art surgery technology can accomplish: minimally invasive medical technology, robots, AI, and real-time 3D imaging. On the other hand, agreement has been reached on one point: Only an increasingly efficient, digital health-care system will be able to meet the needs of its “customers” in the future. In Germany, the focus is being placed — not without reason — on the high level of ingenuity in the medical technology industry. This is an industry that, after all, generates roughly one-third of its revenues with products that were developed in the past three years. It also invests about 9% of its revenues in R&D, well above the industry average (7% worldwide). No other industry generates more patents in Europe. In a worldwide comparison, Germany ranks second with 1,340 patent applications, trailing only the United States (4,872 applications). A large number of small and medium-sized enterprises (SMEs) are highly successful in the global marketplace. To ensure that they remain that way, their medical products must become more “software driven” and network-compatible. But that’s not all: Surgery and nursing robots, prostheses produced by 3D printers, and mobile mini-labs (labs-on-chip) are ready to go to work. In addition, AI algorithms are expected to take on the job of turning the mountains of data produced by sensors, medical wearables, and other sources into information that can be used for diagnostic and therapeutic purposes. Nonetheless, manufacturers face tremendous production-technology challenges in their work to create such miniaturized, multifunctional, and smart products. If that weren’t enough, they also face high regulatory barriers and crushing price pressure. For this reason, only those companies that are capable of efficiently and safely manufacturing this “new” equipment in high quality will succeed in the marketplace.

Smart energy

The energy industry is undergoing a profound process of change. This is because digitization, decentralization, democratization, and decarbonization affect all parts of its value chain, with far-reaching effects on technologies, workflows, and business models. Smart energy, smart grid, smart power, smart home, smart cities — the “clever” adjective has not only infiltrated large areas of our daily lives, it also plays a decisive role in the energy industry. But the road from buzzword to implementation is rocky and long. Data analytics, AI, and the like may have lost their hype status, but they are not self-evident. Yet the use of AI in the energy sector is not new. Already in 2011, the German Research Center for Artificial Intelligence (DFKI), the Karlsruhe Institute of Technology (KIT), AGT Germany, and Seeburger AG together with Stadtwerke Saarlouis started the “Peer Energy Cloud” project. The intelligent processing of the power consumption of individual sockets in private households should optimize the load flow and install a virtual marketplace for power trading in the so-called “micro grid.” However, this project has not yet triggered a flood of imitators. This is also shown by a study conducted by the Association of Municipal Enterprises (VKU) and the BET Office for Energy Management in November 2019. Of the 58 utility companies surveyed, only 13% have already started to implement “smart grid projects,” while 7% are still in the planning stage. In fact, 99% of infrastructures worldwide are not “smart,” as Cedrik Neike, CEO of Siemens Smart Infrastructure, recently said in an interview with the online offering SmarterWorld. But that’s the prerequisite for mastering the issues of resource efficiency, decarbonization, and environmental compatibility in the future. After all, 70% of Earth’s 9.7 billion inhabitants will be living in cities by 2050, according to the UN. Buildings will then account for over one-third of energy consumption. In addition, by 2035, half of the ever-increasing total energy consumption will come from renewable energies from countless small producers. Even now, however, demand and supply in the power grid must be constantly balanced. In the future, this complex task will be manageable only with intelligent energy networks due to the increasing feed-in of renewable energies. This “internet of energy” works with state-of-the-art information and communication technologies (ICT) and measuring equipment from generation to consumption. On the customer side, this is done by intelligent electricity meters (smart meters) and, in the next step, by electrical consumers who, for example, only start their “job” independently when there is an excess supply of electricity. There is still a lot to be done before then. That’s the bad news. The good news is that “smart energy” is proving to be a very lucrative market. Neike predicted in an interview with SmarterWorld that the global core market for smart infrastructure would be worth about €150 billion per year. He sees annual growth in the low-single-digit percentage range. Distributed energy systems and energy storage systems are expected to grow by 10% each and the infrastructure for electromobility by 30% in the next five years. According to a study by VKU in October 2019, distribution network operators in Germany alone will have to invest about €7 billion in the development of smart grids by 2030 in order to be prepared for the “turnaround” with the energy network. Where the journey will lead in the future can be experienced already now at the gates of Vienna. The showcase project Aspern Smart City Research (ASCR) — winner of the World Smart City Award 2016 — is implementing technical solutions for the new energy world with real end customers. They are all based on state-of-the-art ICT with suitable big-data models. In Aspern, for example, the “Green House” is the world’s first passive house plus student residence. It produces its own electricity, forecasts when residents will shower, stores excess energy for colder days, and trades on the balancing energy market. And it does so completely independently. This not only saves energy and thus money, the electricity exchange even promises profits. In the coming project phase “ASCR 2023,” a total of 17 new “use cases” will be the focus of research activities. These include a number of applications relating to electromobility. For example, the charging of car batteries through the use of decentralized renewable energies. In the future, it will also be possible to automatically “refuel” (smart charging) vehicles depending on the current energy price and user behavior. And in a further step, car batteries will then feed electricity into the grid as required — “vehicle to grid.” Another exciting ASCR research topic is the use of digital building twins. These virtual images permanently provide current states of real building units, facilities, or areas — for example, as a basis for predictive maintenance. According to a survey conducted by Sopra Steria Consulting, 60% of energy suppliers saw precisely this as the central benefit of AI in 2017. The more households produce green electricity themselves, the more the boundaries between producers and consumers become blurred. The latter then actively participate in the energy market as so-called “prosumers” and generate the smallest energy flows that have to be managed at low transaction costs. At the same time, new, customer-centric distribution models are emerging. According to a study by Detecon, the topic of blockchaining is thus gaining significantly in importance in the energy industry. This is because the technology, which is known above all in the context of the cryptocurrency Bitcoin, promises direct interaction between players without third parties while at the same time ensuring absolute data sovereignty. Prosumers, for example, can trade their generated solar power with each other in an uncomplicated and secure manner. Similarly, blockchain transactions between charging station and electricity consumer can be automated via blockchain. So-called “smart contracts” already make the provider’s processes between charging and payment much more efficient and cost-effective. However, the majority of Detecon study participants do not expect the technology to be ready for the market for the next five years. Finally, energy supply as a critical infrastructure has always been subject to high-security requirements. This includes not only the prevention of attacks on the IT infrastructure (security) and operational safety (safety) but also data protection (privacy). The increasing networking and digitalization of the energy industry is now leading to a significant increase in the risk situation and thus in the demands on security. After all, even every “smart” electricity meter in households represents a potential gateway.

Energy efficiency

Energy efficiency is one of the most crucial tasks that we face in the 21st century. This effort also includes the use of immense amounts of waste heat — on a large and small scale. “Contradictory” materials play an important role in the process. Most processes used in everyday life and industry produce waste heat that generally vanishes without being put to any good use. The temperatures of this heat are generally too low for energy-recovery purposes. For a long time, one of the most promising options has been thermoelectric materials that can generate power from small differences in thermal energy. The underlying principle is known as the Seebeck effect. This principle shows how a difference in temperature on both ends of particular materials can produce electrical voltage. Of course, such electricity can’t power a factory. But for radio sensors on machines, small and efficient thermoelements could generate sufficient amounts of power from thermal energy. The advantages are obvious: no cables, no battery replacements, autonomy, and permanent availability. This description makes it all sound so simple. But the reality of the process is highly complex. This is not just the case because the electronics have to “handle” electronics with millivolt currents. Furthermore, the material for the thermoelements has to meet demands that actually do not go together. On the one hand, it should conduct electricity well but thermal energy poorly. The problem: Good conductors of electricity are generally good conductors of thermal energy as well. This is not the case with a mishmash of materials made up of iron, vanadium, tungsten, and aluminum that researchers at the Christian Doppler Lab for Thermoelectricity at TU Wien (Vienna University of Technology) have come across. The atoms in this group are generally arranged in a strict, orderly manner, and the distance between the atoms of the individual materials is equally large. This results in a completely regular crystal structure. But this structure changes radically when a thin layer of the material is applied to silicon. The atoms then take on a cubic shape, one that is characterized by a body-centered arrangement. This results in a completely random distribution of the various types of atoms. This mix of regular and irregular patterns of atom order produces very little electrical resistance. At the same time, though, it disrupts the lattice vibrations that handle the job of transporting thermal energy. The thermal conductivity declines, and the temperature difference that is essential for power generation exists for a longer period of time. The amount of electricity that can be generated is expressed by the ZT figure of merit: The higher the figure, the better the thermoelectric properties. Thermoelectric materials that have been measured thus far have ZT levels of about 2.5 to 2.8. The completely new material in use at TU Wien boosts this level markedly, all the way up to 5 to 6.