A new process for preparing high-purity iron(HPI)was proposed,and it was investigated by laboratory experiments and pilot tests.The results show that under conditions of a reduced temperature of 1075°C,reduced ti...A new process for preparing high-purity iron(HPI)was proposed,and it was investigated by laboratory experiments and pilot tests.The results show that under conditions of a reduced temperature of 1075°C,reduced time of 5 h,and CaO content of 2.5wt%,a DRI with a metallization rate of 96.5%was obtained through coal-based direct reduction of ultra-high-grade iron concentrate.Then,an HPI with a Fe purity of 99.95%and C,Si,Mn,and P contents as low as 0.0008wt%,0.0006wt%,0.0014wt%,and 0.0015wt%,respectively,was prepared by smelting separation of the DRI using a smelting temperature of 1625°C,smelting time of 45 min,and CaO content of 9.3wt%.The product of the pilot test with a scale of 0.01 Mt/a had a lower impurity content than the Chinese industry standard.An HPI with a Fe purity of 99.98wt%can be produced through the direct reduction?smelting separation of ultra-high-grade iron concentrate at relatively low cost.The proposed process shows a promising prospect for application in the future.展开更多
The p-center problem consists of choosing a subset of vertices in an undirected graph as facilities in order to minimize the maximum distance between a client and its closest facility. This paper presents a greedy ran...The p-center problem consists of choosing a subset of vertices in an undirected graph as facilities in order to minimize the maximum distance between a client and its closest facility. This paper presents a greedy randomized adaptive search procedure with path-relinking （GRASP/PR） algorithm for the p-center problem, which combines both GRASP and path-relinking. Each iteration of GRASP/PR consists of the construction of a randomized greedy solution, followed by a tabu search procedure. The resulting solution is combined with one of the elite solutions by path-relinking, which consists in exploring trajectories that connect high-quality solutions. Experiments show that GRASP/PR is competitive with the state-of-the-art algorithms in the literature in terms of both solution quality and computational efficiency. Specifically, it virtually improves the previous best known results for 10 out of 40 large instances while matching the best known results for others.展开更多
基金the National Natural Science Foundation of China(No.51904063)the Fundamental Research Funds for the Central Universities,China(Nos.N172503016,N172502005,and N172506011)the China Postdoctoral Science Foundation(No.2018M640259).
文摘A new process for preparing high-purity iron(HPI)was proposed,and it was investigated by laboratory experiments and pilot tests.The results show that under conditions of a reduced temperature of 1075°C,reduced time of 5 h,and CaO content of 2.5wt%,a DRI with a metallization rate of 96.5%was obtained through coal-based direct reduction of ultra-high-grade iron concentrate.Then,an HPI with a Fe purity of 99.95%and C,Si,Mn,and P contents as low as 0.0008wt%,0.0006wt%,0.0014wt%,and 0.0015wt%,respectively,was prepared by smelting separation of the DRI using a smelting temperature of 1625°C,smelting time of 45 min,and CaO content of 9.3wt%.The product of the pilot test with a scale of 0.01 Mt/a had a lower impurity content than the Chinese industry standard.An HPI with a Fe purity of 99.98wt%can be produced through the direct reduction?smelting separation of ultra-high-grade iron concentrate at relatively low cost.The proposed process shows a promising prospect for application in the future.
基金The research was supported by the National Natural Science Foundation of China under Grant Nos. 61370183 and 61262011.
文摘The p-center problem consists of choosing a subset of vertices in an undirected graph as facilities in order to minimize the maximum distance between a client and its closest facility. This paper presents a greedy randomized adaptive search procedure with path-relinking （GRASP/PR） algorithm for the p-center problem, which combines both GRASP and path-relinking. Each iteration of GRASP/PR consists of the construction of a randomized greedy solution, followed by a tabu search procedure. The resulting solution is combined with one of the elite solutions by path-relinking, which consists in exploring trajectories that connect high-quality solutions. Experiments show that GRASP/PR is competitive with the state-of-the-art algorithms in the literature in terms of both solution quality and computational efficiency. Specifically, it virtually improves the previous best known results for 10 out of 40 large instances while matching the best known results for others.
