Systems that never stop learning
The more than 300 researchers in the Information & Automation Technologies (IAT) Cluster develop new solutions for acquiring, modeling, visualizing, and integrating data from various sources so that it can be used for diagnoses and prognoses – whether in the medical or industrial realms or in the energy sector.
Siemens is conducting pioneering research at many locations throughout the world today. The researchers in the IAT Cluster, which has primary sites in the U.S. and Europe for the Imaging & Visualization and Image Analytics & Informatics technology fields, cover the whole range of imaging technologies that are relevant for Siemens. They are developing state-of-the-art systems that range from advanced image reconstruction all the way to database-guided image analysis. Their portfolio of research projects also includes technologies for image content extraction as well as methods of data analysis and data fusion to consolidate the information sent from diverse sources.
Medical practitioners are among the beneficiaries of this research. For example, developers are working on a virtual model of the heart that includes elements of a patient’s anatomy and physiology, such as muscles, valves, hemodynamic features, and reactions to stress. Just as a machine’s functions can be depicted on a computer and its parameters can be changed, so too can doctors use such a model to simulate how procedures such as a heart valve operation will affect the function of a particular patient’s heart. This personalized depiction is possible because the software translates data from imaging modalities and other sources into accurate three-dimensional simulations.
Such models provide doctors with new ways of virtually simulating operative procedures in order to make early predictions regarding their effectiveness. Corporate Technology researchers are also working on the depiction of aneurysms, the effects of stent operations, and methods of predicting the effects of different treatments on cardiac irregularities.
Self-Optimizing Systems
Another important focus area in which the cluster is involved is intelligent systems that perceive their environment, adapt their behavior, learn optimal strategies, draw logical conclusions, and are capable of self-organization and networking with one another.
Some key functions of intelligent systems have already become essential elements in numerous fields of application. Take the production of modern steels, for example, where the material properties of the finished product must be regulated as precisely as possible. An intelligent temperature control system adapts optimally to constantly changing production parameters and sprays precisely the right amount of water onto the steel strip in order to cool it down at exactly the right moment. In the Intelligent Systems & Control (ISC) Global Technology Field (GTF), researchers and development engineers are developing and continuously refining systems such as these.
Intelligent systems also play a very important role when it comes to the operation of gas turbines for power generation. For example, they have made it possible to achieve a new world record of 60 percent efficiency with the world’s largest gas turbine. The 375 megawatt machine has been in operation in a combined-cycle power plant in the Bavarian town of Irsching since 2007.
Simultaneous testing of alternatives
Products and production systems are becoming increasingly complex. The sheer amount of information that is required to effectively implement a technical solution – from the start of its development to the point when it is put into operation – is overwhelming. For this reason, implementation often requires the use of collaborative, intelligent tools, components, and execution systems to manage processes.
The research being done by the Automation Engineering & Run-Time System (ART) GTF covers a product’s entire life cycle. These experts develop powerful tools that make it possible to improve the productivity and reliability of the systems involved – even under the harshest conditions. Thanks to three-dimensional direct modeling and intuitive computer interfaces that are easy to use, it is possible to achieve the rapid conceptualization of different designs at minimal cost. This feature enables the user to simultaneously conceive, simulate, and validate several alternative solutions.
As demand for tightly integrated control systems increases, experts use test platforms that are designed to guarantee the error-free operation of automation systems, with a particular emphasis on their resilience, safety, and security.
Another research focus is on solutions for improved sustainability – for example, an intelligent system for monitoring and controlling energy demand. With this Demand Response system, energy consumers can automatically reduce their power use at times when rates are high, thus saving money. To this end, consumers receive information concerning power availability, their own electricity requirements, and current prices.
Applications for smart systems include ultrasound examinations, turbines, intelligent buildings, digital factories, and virtual models of the heart.
Generating knowledge from data
Today we live in a datacentric economy where enormous amounts of data are being generated almost continuously.
Experts from the Knowledge & Decision Systems Global Technology Field are focusing their efforts on devising technologies that can harness this data. They are developing methods for extracting information and knowledge in such a way that it can be used to support intelligent decisionmaking.
Countless data sources are processed for this purpose. The list of sources includes sensor and service data, sales information, and publicly available data from blogs and social networking sites. CT researchers are currently developing algorithms and software modules that are capable of processing these large amounts of data. In their work they are using various technologies such as machine learning and pattern recognition. They are also exploiting the potential of automatic natural language processing and semantic search engines.
In addition, researchers are supporting the development of company-wide networked information-sharing platforms, such as a Blogosphere for Siemens-wide communication via weblogs and a Wikisphere for a Siemens-wide glossary.
Organization-specific solutions have also been developed. They include a platform for the Communites4Competencies program at Industrial Solutions, which enables experts to share their knowledge about solutions and their documents with the other members of the community. One successful example of such a collaboration is an application designed for the service sector that can process large quantities of unstructured documents related to future planning activities and the provision of individual services. These applications are designed to significantly improve work processes.
Predictive maintenance
CT experts are also developing a variety of solutions for monitoring conditions on the ground. Whether the specific context is a large power plant, a transmission grid or a healthcare imaging device, these systems can deal with process-related sensor data and service information in order to support improved decision-making. This can lead to a reduction of forced outages and improve the availability and reliability of monitored systems. Examples of such systems include monitoring solutions for power plants that improve the reliability of smart grids through advanced power flow analysis and estimation, and predictive maintenance for molecular imaging systems.
In addition to these activities, the IAT Cluster is driving a strategic cross-GTF initiative that focuses on interdisciplinary innovations for the rapid development of new solutions. These innovations include systems that, thanks to their cognitive abilities, “know” what they are doing and the development of particularly advanced hardware infrastructures.
The IAT Cluster is also acting as an incubator for two Corporate Technology lighthouse projects. In the first of these projects, known as Collective Intelligence, researchers are developing new methods for generating knowledge from available data. In the second lighthouse project, High Performance Building, a team is working with universities and a variety of Siemens business units to develop innovative solutions aimed at increasing the energy efficiency, sustainability, and environmental performance of buildings throughout their entire life cycle.
Last but not least, the researchers are also working hard to continuously perfect the foundation of their own work – the intercontinental exchange of knowledge.