The electromagnetic scattering computation has developed rapidly for many years; some computing problems for complex and coated targets cannot be solved by using the existing theory and computing models. A computing m...The electromagnetic scattering computation has developed rapidly for many years; some computing problems for complex and coated targets cannot be solved by using the existing theory and computing models. A computing model based on data is established for making up the insufficiency of theoretic models. Based on the ＂support vector regression method＂, which is formulated on the principle of minimizing a structural risk, a data model to predicate the unknown radar cross section of some appointed targets is given. Comparison between the actual data and the results of this predicting model based on support vector regression method proved that the support vector regression method is workable and with a comparative precision.展开更多
Polycrystalline Cr2AlC coatings were prepared on M38G superalloy using a two-step method consisting of magnetron sputtering from Cr-Al-C composite targets at room temperature and subsequent annealing at 620 ℃. Partic...Polycrystalline Cr2AlC coatings were prepared on M38G superalloy using a two-step method consisting of magnetron sputtering from Cr-Al-C composite targets at room temperature and subsequent annealing at 620 ℃. Particularly, various targets synthesized by hot pressing mixture of Cr, Al, and C powders at 650-1000 ℃ were used. It was found that regardless of the phase compositions and density of the com- posite targets, when the molar ratio of Cr：Al：C in the starting materials was 2：1：1, phase-pure crystalline Cr2AlC coatings were prepared by magnetron sputtering and post crystallization. The Cr2AIC coatings were dense and crack-free and had a duplex structure. The adhesion strength of the coating deposited on M38G superalloy from the 800 ℃ hot-pressed target and then annealed at 620 ℃ for 20 h in Ar exceeded 82 ± 6 MPa, while its hardness was 12 ± 3 GPa.展开更多
文摘The electromagnetic scattering computation has developed rapidly for many years; some computing problems for complex and coated targets cannot be solved by using the existing theory and computing models. A computing model based on data is established for making up the insufficiency of theoretic models. Based on the ＂support vector regression method＂, which is formulated on the principle of minimizing a structural risk, a data model to predicate the unknown radar cross section of some appointed targets is given. Comparison between the actual data and the results of this predicting model based on support vector regression method proved that the support vector regression method is workable and with a comparative precision.
基金supported by the National Natural Science Foundation of China under Grant Nos.51271191,51571205 and 51401209
文摘Polycrystalline Cr2AlC coatings were prepared on M38G superalloy using a two-step method consisting of magnetron sputtering from Cr-Al-C composite targets at room temperature and subsequent annealing at 620 ℃. Particularly, various targets synthesized by hot pressing mixture of Cr, Al, and C powders at 650-1000 ℃ were used. It was found that regardless of the phase compositions and density of the com- posite targets, when the molar ratio of Cr：Al：C in the starting materials was 2：1：1, phase-pure crystalline Cr2AlC coatings were prepared by magnetron sputtering and post crystallization. The Cr2AIC coatings were dense and crack-free and had a duplex structure. The adhesion strength of the coating deposited on M38G superalloy from the 800 ℃ hot-pressed target and then annealed at 620 ℃ for 20 h in Ar exceeded 82 ± 6 MPa, while its hardness was 12 ± 3 GPa.
