Abstract: The present invention discloses a hybrid-equivalence-based power electronics system partition computing method, and relates to the field of power electronics simulation. In the power electronics system partition computing method, by separating series nodes and parallel nodes from internal nodes, converting the series nodes into an impedance form, and converting the parallel nodes into an admittance form, hybrid equivalent conversion of sub-partitions is implemented. By projecting hybrid equivalent matrices and hybrid excitation matrices of the sub-partitions onto a global coupling coordinate system, there is linear additivity between the hybrid equivalent matrices of the sub-partitions, and a series coupling current and a parallel coupling voltage of a system are efficiently solved, thereby implementing parallel partition decoupling computing of the power electronics system.

Abstract: Disclosed are a testing system and testing method for a magnetic refrigerator. In the system, an adjustable load includes a heating container, and a flow direction control structure. The heating container is provided with an inlet, an outlet, and a heater with controllable heating power. The flow direction control structure is configured to enable a heat exchange fluid flowing out from a cold end of any magnetic regenerator module in magnetic regenerator modules arranged in pairs to flow into a cold end of another magnetic regenerator module through the inlet and outlet of the heating container in sequence, the heater is used to heat the heat exchange fluid inside the heating container, and an on-off state between the adjustable load and the cold end of any magnetic regenerator module is controllable.

Abstract: An adaptive, large graph-oriented unified memory management method is proposed; according to the order of priorities of different types of graph data structure in a graph computing application, whether the current GPU memory is full is sequentially determined by means of GPU memory checking; whether the size of current graph data exceeds the available memory capacity of the GPU is determined by means of data overflow checking; and then a unified memory management policy is configured; the method uses different graph algorithms based on the characteristics of graph data structure and the size of the available GPU memory; the method can significantly improve the performance of processing large graphs, which exceed the video memory capacity under a unified memory architecture, including improving GPU bandwidth utilization, reducing the number of page faults and the overhead of processing page faults, and speeding up the execution of graph computing programs.

Abstract: Provided are a chemical-mechanical polishing composition and a chemical-mechanical polishing method capable of polishing a semiconductor substrate containing a conductive metal such as tungsten or cobalt flatly and at high speed as well as reducing surface defects after polishing. The composition for chemical-mechanical polishing contains (A) silica particles having a functional group represented by general formula (1) and (B) a silane compound. —COO-M+ . . . (1) (M+ represents a monovalent cation.

Abstract: Provided is a node diaphragm scraping and sorting device for a bamboo split, which is configured to scrape node diaphragms of the bamboo splits and to sequentially sort the bamboo inner layers and bamboo outer layers of the bamboo splits towards the same direction. The device includes a scraping device, a sorting device and a clamping and conveying device which are arranged from front to back along an axis. The scraping device includes a conical cylinder, two pressing semi-rings, and a scraping ring. The sorting device includes an e-shaped sleeve composed of an inner cylinder and an outer cylinder, and two shifting forks located at both ends of the sleeve. The clamping and conveying device includes an inner ring moving on the axis, two clamping semi-rings and a conveying power device. The structure, function and technical parameters of each branch device are also introduced in detail.

Abstract: A graph sampling and random walk acceleration method and system based on GPU. The graph data is read from the storage medium through the CPU and converted into CSR format and then output to the GPU. The GPU works based on the defined working mode: Real-time generation of alias table and sampling; or offline judgment whether there is a pre-generated alias table and sampling, which executes the alias method efficiently and in parallel, and can significantly improve the performance of graph data processing on the same hardware platform, including improving the sampling throughput and reducing the overall running time.

Abstract: An adaptive, large graph-oriented unified memory management method is proposed; according to the order of priorities of different types of graph data structure in a graph computing application, whether the current GPU memory is full is sequentially determined by means of GPU memory checking; whether the size of current graph data exceeds the available memory capacity of the GPU is determined by means of data overflow checking; and then a unified memory management policy is configured; the method uses different graph algorithms based on the characteristics of graph data structure and the size of the available GPU memory; the method can significantly improve the performance of processing large graphs, which exceed the video memory capacity under a unified memory architecture, including improving GPU bandwidth utilization, reducing the number of page faults and the overhead of processing page faults, and speeding up the execution of graph computing programs.

Abstract: Provided are abrasive grains and a composition for chemical mechanical polishing which are for selectively polishing a silicon nitride film, and which are applicable not only to silicon oxide films but also to amorphous silicon films and polysilicon films. This method for manufacturing abrasive grains includes: a first step of heating a mixture which contains particles having a sulfanyl group (—SH) fixed to the surface thereof via covalent bonds, and which contains a compound having carbon-carbon unsaturated double bonds; and a second step, which is performed after the first step, of further adding a peroxide and carrying out heating.

Abstract: Provided is a node diaphragm scraping and sorting device for a bamboo split which is configured to scrape node diaphragms of the bamboo splits and to sequentially sort the bamboo inner layers and bamboo outer layers of the bamboo splits towards the same direction. The device includes a node diaphragm scraping device, a sorting device and a clamping and conveying device which are arranged from front to back along an axis. The node diaphragm scraping device includes a conical cylinder, two pressing semi-rings, and a scraping ring. The sorting device includes an e-shaped sleeve composed of an inner cylinder and an outer cylinder, and two shifting forks located at both ends of the sleeve. The clamping and conveying device includes an inner ring moving on the axis, two clamping semi-rings and a conveying power device. The structure, function and technical parameters of each branch device are also introduced in detail.

Abstract: Provided are a composition for chemical mechanical polishing and a method for chemical mechanical polishing, whereby a tungsten film as a wiring material can be polished at high speed, and the occurrence of surface defects in a polished surface can be reduced. A composition for chemical mechanical polishing pertaining contains (A) alumina particles, at least a portion of the surface of which is coated with a coating film of silica alumina, and (B) a liquid medium.

Abstract: Provided are a composition for chemical mechanical polishing and a chemical mechanical polishing method, which enable reduction in occurrence of surface defects in a polished surface and which enable high speed polishing of a tungsten film which is a wiring material. The composition for chemical mechanical polishing contains: (A) particles containing alumina and having a functional group represented by general formula (1); and (B) a liquid medium. (1): —SO3?M+ (In the formula, M+ represents a monovalent cation.

Abstract: Provided are a composition for chemical mechanical polishing and a method for polishing allowing a tungsten film- or silicon nitride film-containing semiconductor substrate to be polished at a high speed, while also enabling a reduction in the occurrence of a surface defect in the polished face after polishing. A composition for chemical mechanical polishing according to the present invention comprises (A) abrasive grains containing titanium nitride and (B) a liquid medium, wherein the absolute value of the zeta-potential of said (A) component in the composition for chemical mechanical polishing is 8 mV or higher.

Abstract: Provided are a chemical-mechanical polishing composition and a chemical-mechanical polishing method capable of polishing a semiconductor substrate containing a conductive metal such as tungsten or cobalt flatly and at high speed as well as reducing surface defects after polishing. The composition for chemical-mechanical polishing contains (A) silica particles having a functional group represented by general formula (1) and (B) a silane compound. —COO-M+ . . . (1) (M+ represents a monovalent cation.

Abstract: Provided are a chemical mechanical polishing composition and a chemical mechanical polishing method that can polish a semiconductor substrate containing an electric conductor metal, such as tungsten or cobalt, flat and at high speed, and reduce post-polishing surface defects. The chemical mechanical polishing composition contains (A) silica particles having the functional group represented by general formula (1), and (B) at least one selected from the group consisting of a carboxylic acid having an unsaturated bond and a salt thereof. (1): —COO-M+ (M+ represents a monovalent cation.

Abstract: A graph sampling and random walk acceleration method and system based on GPU. The graph data is read from the storage medium through the CPU and converted into CSR format and then output to the GPU. The GPU works based on the defined working mode: Real-time generation of alias table and sampling; or offline judgment whether there is a pre-generated alias table and sampling, which executes the alias method efficiently and in parallel, and can significantly improve the performance of graph data processing on the same hardware platform, including improving the sampling throughput and reducing the overall running time.

Abstract: A proton therapy system based on a compact superconducting cyclotron, including: a superconducting cyclotron system, an energy selection system, a beam transport system, a fixed therapy room subsystem and a rotating frame therapy subsystem; a fixed-energy proton beam extracted from a superconducting cyclotron of the superconducting cyclotron system is adjusted into a continuous and adjustable proton beam of 70 MeV to 200 MeV by the energy selection system, thus realizing a longitudinal adjustment for a proton range during treating a tumor, and the continuous and adjustable proton beam is respectively transmitted to the fixed therapy room subsystem and the rotating frame therapy subsystem by the beam transport system. The cooperative control of the superconducting cyclotron system, the energy selection system, the beam transport system and the therapy head realizes the transverse expansion of proton beams, thus realizing intensity modulated radiation therapy for the tumor.

Abstract: A proton therapy system based on a compact superconducting cyclotron, including: a superconducting cyclotron system, an energy selection system, a beam transport system, a fixed therapy room subsystem and a rotating frame therapy subsystem; a fixed-energy proton beam extracted from a superconducting cyclotron of the superconducting cyclotron system is adjusted into a continuous and adjustable proton beam of 70 MeV to 200 MeV by the energy selection system, thus realizing a longitudinal adjustment for a proton range during treating a tumor, and the continuous and adjustable proton beam is respectively transmitted to the fixed therapy room subsystem and the rotating frame therapy subsystem by the beam transport system. The cooperative control of the superconducting cyclotron system, the energy selection system, the beam transport system and the therapy head realizes the transverse expansion of proton beams, thus realizing intensity modulated radiation therapy for the tumor.