In soft sensor field, just-in-time learning(JITL) is an effective approach to model nonlinear and time varying processes. However, most similarity criterions in JITL are computed in the input space only while ignoring...In soft sensor field, just-in-time learning(JITL) is an effective approach to model nonlinear and time varying processes. However, most similarity criterions in JITL are computed in the input space only while ignoring important output information, which may lead to inaccurate construction of relevant sample set. To solve this problem, we propose a novel supervised feature extraction method suitable for the regression problem called supervised local and non-local structure preserving projections(SLNSPP), in which both input and output information can be easily and effectively incorporated through a newly defined similarity index. The SLNSPP can not only retain the virtue of locality preserving projections but also prevent faraway points from nearing after projection,which endues SLNSPP with powerful discriminating ability. Such two good properties of SLNSPP are desirable for JITL as they are expected to enhance the accuracy of similar sample selection. Consequently, we present a SLNSPP-JITL framework for developing adaptive soft sensor, including a sparse learning strategy to limit the scale and update the frequency of database. Finally, two case studies are conducted with benchmark datasets to evaluate the performance of the proposed schemes. The results demonstrate the effectiveness of LNSPP and SLNSPP.展开更多
Japanese IMS-VHT project on the Virtual Heat Treatment tool for monitoring and optimising HT process in relation with the international cooperative programs is briefly introduced. This project motivates to develop vir...Japanese IMS-VHT project on the Virtual Heat Treatment tool for monitoring and optimising HT process in relation with the international cooperative programs is briefly introduced. This project motivates to develop virtual tools for computer to optimize the heat treatment condition and to support decision for HT operation by knowledge based database in addition to process simulation. As one of the activities with the cooperation of the Society of Materials Science, Japan and the Japan Society for Heat Treatment, a benchmark project is undergoing. This includes simulation of carburized quenching process of a cylinder, disc, and ring as well as a helical gear by use of common data of materials properties and cooling characteristics by several available simulation programs. A part of the newly obtained results is presented as an interim report.展开更多
During the NERIES Project, an accelerometric database containing European digital information was developed. Besides event and station metadata, ground motion parameters, computed in a homogeneous manner, were assemb...During the NERIES Project, an accelerometric database containing European digital information was developed. Besides event and station metadata, ground motion parameters, computed in a homogeneous manner, were assembled： PGA, PGV, AI, TD, CAV, H1 and PSV（f,5%） （19,961 components, 2629 events, 547 stations）. Merging small and moderate magnitude events produced a unique database capable of providing important information such as： （i） Correlations between several ground motion parameters follow analogous trends as in previous worldwide datasets, with slight corrections. （ii） Although PGA attenuations with distance show great uncertainties, four recent GMPEs recommended for Europe fit quite well the central 50% data interval for the distance range 10 〈 R 〈 200 kin; outside these distances, they do not fit. （iii） Soil amplification ratios indicate that weak motion （low magnitudes and larger distances） shows larger amplification than strong motion （short distances and large magnitudes） as represented in UBC97 for the USA, but not in EC8 for Europe. （iv） Average spectral shapes are smaller than in the EC8. （v） Differences in amplification factors for PGA, PGV and HI for EC8 soil classes B and C, and differences in spectral shapes for these soil classes, indicate that EC8, Type 2 S-coefficient should be frequency dependent, as in UBC97.展开更多
The changing trends in information technology have greatly influenced the role of GIS in spatial data management, analysis, processing and presentation. It has evolved from the conventional cartography and image proce...The changing trends in information technology have greatly influenced the role of GIS in spatial data management, analysis, processing and presentation. It has evolved from the conventional cartography and image processing to advanced 3D visualization and dynamic graphics tools. Due to this evolving nature of GIS, it has found wide applications in a number of diverse fields. Geophysical exploration projects involve data acquisition at hundreds of spatial locations resulting in large number of datasets. It takes a great deal of time to manage all these datasets during data processing and interpretation. This paper presents the use of GIS as an effective project management tool, providing an interactive data access interface in compute intensive geophysical processing applications. A reusable GIS software component is presented which can be used by geophysical applications to manage their datasets. A practical example is included to demonstrate the implementation of this GIS component as an embedded Project Manager in a seismic refraction software.展开更多
基金Supported by the National Natural Science Foundation of China(61273160)the Fundamental Research Funds for the Central Universities(14CX06067A,13CX05021A)
文摘In soft sensor field, just-in-time learning(JITL) is an effective approach to model nonlinear and time varying processes. However, most similarity criterions in JITL are computed in the input space only while ignoring important output information, which may lead to inaccurate construction of relevant sample set. To solve this problem, we propose a novel supervised feature extraction method suitable for the regression problem called supervised local and non-local structure preserving projections(SLNSPP), in which both input and output information can be easily and effectively incorporated through a newly defined similarity index. The SLNSPP can not only retain the virtue of locality preserving projections but also prevent faraway points from nearing after projection,which endues SLNSPP with powerful discriminating ability. Such two good properties of SLNSPP are desirable for JITL as they are expected to enhance the accuracy of similar sample selection. Consequently, we present a SLNSPP-JITL framework for developing adaptive soft sensor, including a sparse learning strategy to limit the scale and update the frequency of database. Finally, two case studies are conducted with benchmark datasets to evaluate the performance of the proposed schemes. The results demonstrate the effectiveness of LNSPP and SLNSPP.
文摘Japanese IMS-VHT project on the Virtual Heat Treatment tool for monitoring and optimising HT process in relation with the international cooperative programs is briefly introduced. This project motivates to develop virtual tools for computer to optimize the heat treatment condition and to support decision for HT operation by knowledge based database in addition to process simulation. As one of the activities with the cooperation of the Society of Materials Science, Japan and the Japan Society for Heat Treatment, a benchmark project is undergoing. This includes simulation of carburized quenching process of a cylinder, disc, and ring as well as a helical gear by use of common data of materials properties and cooling characteristics by several available simulation programs. A part of the newly obtained results is presented as an interim report.
文摘During the NERIES Project, an accelerometric database containing European digital information was developed. Besides event and station metadata, ground motion parameters, computed in a homogeneous manner, were assembled： PGA, PGV, AI, TD, CAV, H1 and PSV（f,5%） （19,961 components, 2629 events, 547 stations）. Merging small and moderate magnitude events produced a unique database capable of providing important information such as： （i） Correlations between several ground motion parameters follow analogous trends as in previous worldwide datasets, with slight corrections. （ii） Although PGA attenuations with distance show great uncertainties, four recent GMPEs recommended for Europe fit quite well the central 50% data interval for the distance range 10 〈 R 〈 200 kin; outside these distances, they do not fit. （iii） Soil amplification ratios indicate that weak motion （low magnitudes and larger distances） shows larger amplification than strong motion （short distances and large magnitudes） as represented in UBC97 for the USA, but not in EC8 for Europe. （iv） Average spectral shapes are smaller than in the EC8. （v） Differences in amplification factors for PGA, PGV and HI for EC8 soil classes B and C, and differences in spectral shapes for these soil classes, indicate that EC8, Type 2 S-coefficient should be frequency dependent, as in UBC97.
文摘The changing trends in information technology have greatly influenced the role of GIS in spatial data management, analysis, processing and presentation. It has evolved from the conventional cartography and image processing to advanced 3D visualization and dynamic graphics tools. Due to this evolving nature of GIS, it has found wide applications in a number of diverse fields. Geophysical exploration projects involve data acquisition at hundreds of spatial locations resulting in large number of datasets. It takes a great deal of time to manage all these datasets during data processing and interpretation. This paper presents the use of GIS as an effective project management tool, providing an interactive data access interface in compute intensive geophysical processing applications. A reusable GIS software component is presented which can be used by geophysical applications to manage their datasets. A practical example is included to demonstrate the implementation of this GIS component as an embedded Project Manager in a seismic refraction software.
