Simulation-based engineering and science
Download 0.9 Mb.
|
- Navigate this page:
- Site: Technical University of Denmark (DTU) Wind Engineering Department of Mechanical Engineering (MEK) Nils Koppels Alle, Building 403
- WTEC Attendees
- Background – Wind Energy
- Research Activities and Simulation Tools
- Site: Technical University of Denmark Center for Biological Sequence Analysis Systems Biology Department (BioCentrum-DTU ) Kemitorvet, Building 208
CONCLUSIONS The Chemical Engineering Department at TU Denmark is a premier research and education center in Denmark. Both CAPEC and IVC-SEP have a long history of collaboration with industry. The base funding for centers comes through membership fees that are used to develop tools, software, and databases of interest to the funding industries. Faculty members in this department are also leading international efforts in developing open standards for chemical engineering software and tools. Site: Technical University of Denmark (DTU) Wind Engineering
http://www.risoe.dk/ Date Visited February 27, 2008 WTEC Attendees: G. Karniadakis (report author), C. Sagui Host: Jens N. Sorensen, Professor of Fluid Mechanics Tel: +45 4525 4314; Fax: +45 4593 0663 E-mail: jns@mek.dtu.dk Jens H. Walther, Associate Professor of Fluid Mechanics E-mail: jhw@mek.dtu.dk
The number of kW of wind power generated in Denmark per 1000 inhabitants exceeds 570 and is by far the highest in Europe and indeed the world. Spain and Germany follow with 340 and 270 (kW/1000 inhabitants), respectively. By the end of January 2007, there were 5,267 turbines installed in Denmark, with a total power of 3,135 MW; the total wind power in European Union countries during that period was 56,535 MW with Germany leading at 22,247 MW. The total installed power and the numbers of turbines in Denmark had a continually increase until 2002. From 2001 to 2003, a replacement agreement was carried out, where smaller and badly placed turbines were replaced with bigger turbines. In 2004 a new replacement agreement running to 2009 was introduced. The agreement involves replacing old 175 MW turbines with 350 MW turbines. Besides the replacement agreement, two offshore wind farms, each for 200 MW, have been approved (Horns Rev II and Rødsand II). These farms are expected to be connected to the grid in 2009–2010. Denmark’s electricity production from wind in 2006 was 6.108 GWh, which corresponded to 16.8 percent of the electricity consumption in Denmark or to the consumption in about 1.73 million Danish households. In the first half of 2007, the turbines produced 3.934 GWh, which corresponds to 21.7 percent of the electricity consumption in Denmark. When offshore turbines at Horns Rev II and Rødsand II are installed, it is expected that wind power will account for 25 percent of Denmark’s total electricity consumption. Danish manufactures of wind turbines have in recent years had a total global market share of 40 percent. The global market for wind power has grown dramatically, and such development is expected to continue. In 2006 Denmark installed about 15,000 MW new capacity, and by the end of 2006 its total installment reached 74,300 MW. The yearly installment of new capacity is expected to increase about 17% to 2011. Systematic research and education in wind turbines in Denmark started more than 30 years ago, and simulation-based design of wind turbines started about 15 years ago. Totally, it is expected that there will be about €150 million per year devoted to research in energy and environmental issues in Denmark (population about 5.5 million).
Many advances in simulation-based design of wind turbines are associated with researchers at DTU and at Risø DTU National Laboratory for Sustainable Energy, which became part of the Technical University of Denmark (DTU) in January 2007. Examples are the development of popular aeroelastic code FLEX and the design of parts of the Nible and Tjaereborg turbines. Researchers at DTU participated in the Danish National Program on Wind Energy (1977–1990). There are approximately 20 researchers at DTU but there are up to 150 such researchers if combined with those of Risø. The annual wind engineering budget at DTU is DKK 20 million (Danish Krone); the sponsors of the program are DTU, EFP (Ministry of Energy), UVE, PSO (Electric Utility), and the EU. Researchers at DTU and Risø work closely with researchers from the University of Aalborg (AaU) and the Danish Hydraulic Institute (DHI) supported by the Danish Research Consortium on Wind Energy. DTU researchers work closely with private companies on wind turbines, and VESTAS (one of the leading companies) has established scholarships for students and five-year full professorships in Wind Energy. Also, Siemens has an office in DTU, and an Indian company SUZLON (fifth-largest wind energy company worldwide) opened an office recently in Copenhagen.
Wind energy involves multiple disciplines, e.g., fluid mechanics, aeroelasticity, and electrical engineering. The work at DTU is focused on the following areas: design of optimum airfoils; dynamic stall, especially 3 D stall; tip flows and yaw; heavily loaded motors; and interference (wake and park) effects. The following are ongoing fully funded projects currently at DTU:
In many of these simulation-based research programs, verification and validation is performed using standard benchmarks for flow but also with experimental data collected at the Department’s scaled-down wind tunnel facilities but also in collaboration with a research institute that is near-by and has full-scale experimental facilities. There are many important simulation tools developed at DTU in collaboration with Risø researchers. In particular, the FLEX code has undergone developments over many generations (currently FLEX5) and is used for designing wind turbines and analyzing loadings. Also, EllipSys is a CFD code for wind turbine aerodynamics simulations in both 2D and 3D. Another program is WPPT/WASP for predicting wind energy; unlike FLEX5, WASP uses the more advanced code EllipSys. Finally, DTU researchers are the keepers of IEA—an international database on wind measurements to determine loadings (including USA). Education DTU offers a unique two-year MSc program open to international students that is taught in English. There is a total of 20 students per year with about 3-4 students per year from the United States. The program entails one semester of required courses (wind turbine aerodynamics, wind aeroelasticity), two semesters of elective courses (aerodynamics and fluid mechanics, structural mechanics, construction and materials, power electronics and grid connection prediction, and optimization), and a last semester for a final project in collaboration with wind energy companies or related institutions. This unique program was featured in the 2005 special issue of Wind Engineering where several MSc theses were published as papers; the WTEC team’s host, Prof Sorensen, was the guest editor of the special issue. Another wind engineering-specific education activity by DTU is the PhD program supported by the Danish Academy on Wind Energy (DAWE) with participating institutions (DTU, AaU, and DHI). About 40–50 PhD students are enrolled in the program. Also supported by the program are summer courses, seminars, and guest researchers.
Research on wind energy in Denmark has been taking place for over 30 years, and DTU researchers in collaboration with those of its partner institution Risø have been the leaders in this field. The research approaches employed at DTU are both of the fundamental type (e.g., large-eddy simulations and sophisticated grid techniques for moving meshes) but also of practical use, e.g., the popular aeroelastic code FLEX that uses lumped modeling for wind energy prediction. DTU also maintains a unique data bank of international meteorological and topographical data. In addition, the education activities at DTU are unique and impressive. It is one of the very few places in the world to offer both MSc and PhD degrees on wind engineering, and its international MSc program attracts students from around the world, including the United States, where there are no such programs currently. Site: Technical University of Denmark Center for Biological Sequence Analysis
Tel: +45-45 25 24 77; Fax: +45-45 93 15 85 Email: brunak@cbs.dtu.dk David Ussery, Microbial Genomics, Visualization Email: dave@cbs.dtu.dk Thomas S. Jensen, Integrative Systems Biology Email: skot@cbs.dtu.duk Petek F. Hallin, Microbial Genomics Email: pfh@chs.dtu.dk
The Center for Biological Sequence Analysis (CBS) at the Technical University of Denmark (DTU) was formed in 1993 by coalescing in one center diverse bioinformatics activities dating back to the mid-1980s. It conducts basic research in the field of bioinformatics and system biology. It employs 100 people (80% research personnel and 20% staff) with 2:1 bio to non-bio backgrounds, and with a highly multidisciplinary profile (molecular biologists, biochemists, medical doctors, physicists, and computer scientists). CBS is one of the largest bioinformatics centers in the European Union.13 It has a strong teaching component, with many courses; some are transmitted in real time over the Internet. CBS showcases a highly popular suite of WWW servers and codes, and it has a very strong publication and citation profile. CBS is funded—in addition to a contribution from the Technical University of Denmark—by the Danish Research Foundation, the Danish Center for Scientific Computing, the Villum Kann Rasmussen Foundation, the Novo Nordisk Foundation, other institutions in the European Union, in industry, and the U.S. National Institutes of Health. CBS Research Information technology has become crucial for research in molecular biology, biotechnology, and pharmacology. Comprehensive public databases of DNA and protein sequences, macromolecular structure, gene and protein expression levels, pathway organization, and cell signaling have been created to facilitate the scientific manipulation of ever-increasing data within biology. Unlike many other groups in the field of biomolecular informatics, the Center for Biological Sequence Analysis directs its research primarily towards topics related to the elucidation of the functional aspects of complex biological mechanisms. CBS currently has 10 research groups:
The CBS publications appear in high-profile journals (Science, Nature, Molecular Cell, etc.) and command high citation records; even fairly recent papers easily surpass the 1,000 citations. Some of these papers appear in the top of the “Hot Biology” ISI list. In the last decade, the Center for Biological Sequence Analysis has produced a large number of computational methods, which are offered to others via WWW servers.
These codes are highly popular, as evidenced by the number of page-views to the CBS WWW pages (over 2 million a month). The popularity of these servers is due to the fact that the computational models have higher fidelity (accuracy) than most of the experiments. For instance in protein sorting, the experimental precision is lower because things depend on where the protein is performing the function, how the cell directs the protein, and so forth. As an example, the SignalP server predicts the presence and location of signal peptide cleavage sites in amino acid sequences from different organisms: Gram-positive prokaryotes, Gram-negative prokaryotes, and eukaryotes. The method incorporates a prediction of cleavage sites and a signal peptide/nonsignal peptide prediction based on a combination of several artificial neural networks and hidden Markov models. The original paper (Nielsen et al. 1997) has over 3,300 citations to date. In general, the WWW sites provide only predictions, no data assessment of the algorithm prediction. Most of the work is carried out on a variety of shared-memory SGI machines. One of the major challenges in the field is the integration of data. The amount of data generated by biology is just exploding. CBS has a relational data warehouse comprising 350+ different databases. In order to handle data integration of 120 terabyte size, CBS has developed its own integration tool (MySQL?). At present there are 15 EU-funded projects dealing with European infrastructure for bioinformatics based on Web services. Parts of the infrastructure needs of Systems Biology are addressed by the European Commission and United States (EC-US) Task Force on Biotechnology Research (http://ec.europa.eu/research/biotechnology/ec-us/). The EC-US Task Force carries out numerous activities designed to bring European and U.S. researchers closer together. Its activities include sponsoring scientific workshops, short training courses, and short-term fellowships. In particular, WTEC sponsored the US-EC Workshop on Infrastructure Needs of Systems Biology (May 2007, Tufts University, Boston) under the auspices of the task force, which brought together 24 scientists from the European Union, the United States, and Canada. In this workshop only one session was dedicated to experiments, while the other three sessions were dedicated to databases, modeling applications, and software infrastructure (http://www.wtec.org/ec-us_sysbio_workshop). An example of “messy” databank is given by the protein-protein interactions (PPI) databanks, which at present include 15 databases that are rather disorganized and chaotic, since there is no agreement on how to represent these interactions, on which format to choose, etc. CBS scientists view this challenge as a scientific window of opportunity, since the databases will simply be “exploding” with information. Data integration requires the analysis of data across different experimental platforms. The results can be used as proxy for experiments that are either impossible to carry out or that will only be feasible in the far future. CBS’s goals go beyond one-gene biology. CBS aims at discovering novel specific functional aspects by exploiting systems-level principles that apply to one or more organisms, and also by finding new components. Ultimately, they aim to develop data integration systems that can talk to one another. This is clearly expressed in the Nature editorial “Let data speak to data” (2005): “Various sorts of data are increasingly being stored in formats that computers can understand and manipulate, allowing databases to talk to one another. This enables their users quickly to adapt technologies to extract and interpret data from different sources, and to create entirely new data products and services.… In biodiversity research, for example, rather than creating centralized monolithic databases, scientists could tap into existing databases wherever the data are held, weaving together all the relevant data on a species, from its taxonomy and genetic sequence to its geographical distribution. Such decentralization also helps to solve the problem that databases are often the fruits of individual or lab research projects that are vulnerable to the vagaries of funding, and to people and labs moving on to pastures new.” As an example of the extremely complex and messy systems that CBS studies, the group studied the temporal aspects of biological networks (as opposed to static topological properties). Their integrative approach combined protein-protein interactions with information on the timing of the transcription of specific genes during the yeast cell cycle, obtained from DNA microarray time series. The resulting time-dependent interaction network places both periodically and constitutively expressed proteins in a temporal cell cycle context, thereby revealing previously unknown components and correlations. The dynamic proteins are generally expressed just before they are needed to carry out their function, generally referred to as just-in-time synthesis. However, they found that most complexes consist of both periodically and constitutively expressed subunits, which suggests that the former control complex activity by a mechanism of just-in-time assembly or activation. Transcriptional regulation influences almost all cell cycle complexes and thereby, indirectly, their static subunits. As a consequence, many cell cycle proteins cannot be identified through the analysis of any single type of experimental data but only through integrative analysis of several data types (de Lichtenberg et al. 2005). Another interesting result the group found is that despite the fact the protein complexes involved in the transcription process are largely the same among all eukaryotes, their regulation has evolved considerably, and thus the identity of the periodically expressed proteins differs significantly between organisms. In addition, these changes in transcriptional regulation have co-evolved with post-translational control independently in several lineages; loss or gain of cell-cycle-regulated transcription of specific genes is often mirrored by changes in phosphorylation of the proteins that they encode (Jensen et al. 2006). The assembly of the same molecular machines at the right time during the cell cycle has therefore evolved very differently, which can have dire consequences when, for instance, drugs designed for human consumption are tested in different organisms. Ultimately, the goal of these and future studies is to achieve cell life predictions and understanding. At present the CBS group is moving towards other frontiers, the “disease interactomes.” This involves the use of text mining to relate which proteins relate to which diseases, or the identification of new disease gene candidates by the phenomic ranking of protein complexes. Briefly, the CBS scientists carried out “…a systematic, large-scale analysis of human protein complexes comprising gene products implicated in many different categories of human disease to create a phenome-interactome network. This was done by integrating quality-controlled interactions of human proteins with a validated, computationally derived phenotype similarity score, permitting identification of previously unknown complexes likely to be associated with disease. Using a phenomic ranking of protein complexes linked to human disease, we developed a Bayesian predictor that in 298 of 669 linkage intervals correctly ranks the known disease-causing protein as the top candidate, and in 870 intervals with no identified disease-causing gene, provides novel candidates implicated in disorders such as retinitis pigmentosa, epithelial ovarian cancer, inflammatory bowel disease, amyotrophic lateral sclerosis, Alzheimer disease, type 2 diabetes and coronary heart disease” (Lage et al. 2007). In this work, they define a “word vector” with the words describing the disease and then look at the relative angles between the vectors in “medical term space” to quantify the phenotypical overlap. The draft of 506 protein complexes associated with pathology is publicly available. This new research leads to new types of “biobanks” with new computational challenges:
Nordic countries with long traditions for keeping medical records and biological samples (detailed from the 1950s, less detailed from the 1850s) provide ideal databases for this task. The combination of medical informatics with bioinformatics and systems biology represents a new trend in disease gene finding and phenotype association, which can also include social and behavioral levels. This also requires linking two 30–40 year old research traditions in systems biology: the data-poor (physics initiated) approach of the Kitano type, and the data-driven, model-fitted approach. Linking medical informatics with bioinformatics and systems biology requires bridging the gap between the molecular level and the phenotypic clinical levels, and linking two exponentially growing types of computer-accessible data: biomolecular databases and their clinical counterparts. The challenge ahead lies not only in linking the clinical level to the “parts list” of the human body (individual genes and proteins) but in understanding defects in biological mechanisms and disease aetiology in a network biology setting. This necessitates the development of a systems biology that includes a changing environment. Finally, a lot of the information on the human genome input-output relation is and will be embedded in biobanks, and complex, messy databases, patient records, and registries in different countries. Text and other types of nontraditional “data types” contain information that can be used to reveal molecular disease mechanisms, finding disease genes and their systemic properties, in particular when a changing environment plays an important role. To achieve this, links to electronic medical records and real-time monitoring of living organisms is needed. This enterprise has over $100 million in Danish funding, from Novo Nordisk and the Rasmussen Foundations. Interestingly, new biotechnology companies are bound to appear, like the recent “23andMe – Genetics Just Got Personal,” a privately held biotechnology company based in California that is developing new ways to help people make sense of their own genetic information. The company offers $1000 tests for select, single-nucleotide polymorphisms. Another big challenge is the visualization of large amounts of data (for instance, a run of a genome sequence generates 7 terabytes of data). CBS has an important effort in this area tackling different biological issues (bacterial genomes, rRNAs, tRNAs, codon usage bias, mRNA, BLAST Atlases for proteome comparisons, etc). CBS has come up with the “DNA Structural Atlas,” which is a method of visualizing structural features within large regions of DNA. It was originally designed for analysis of complete genomes, but it can also be used quite readily for analysis of regions of DNA as small as a few thousand base-pairs in length. The basic idea is to plot the values for six different mechanical-structural properties of the DNA helix in a circle (or arc) representing the complete genome (or chromosome). At the level of whole genomes or chromosomes, large architecturally important regions can be seen. At the level of individual genes (e.g., usually around 10,000 to 20,000 bp for the entire plot), intergenic regions can be examined. The plots are created using the "GeneWiz" program, developed by Hans-Henrik Stærfeldt at CBS.
Download 0.9 Mb. Share with your friends:
|