7results about How to "Achieve overlap" patented technology

The utility model relates to nanometer silver wire technical field, specifically related to a kind of compression device for improving nanometer silver wire lap joint, the inside one side of bearing frame and the one side of bottom plate are rotatably connected with several rotating rods, and all rotating rods are all sleeved with compression sleeve, and the rotating rod of bearing frame inside and the rotating rod of bottom plate one side are one-to-one corresponding distribution, the top of frame body is fixedly connected with top plate, and the top side of top plate is fixedly connected with hydraulic oil cylinder by bolt, by introducing material surface cleaning function, compression gap hydraulic adjusting mechanism and multi-specification material adaptation structure, the defects of existing compression equipment in cleaning ability insufficient, adjusting flexibility poor and the like are successfully overcome, and the compression quality and product consistency of nanometer silver wire conductive film are significantly improved.

The utility model relates to traffic transportation and emergency rescue mechanical device technical field, concretely relates to a kind of angle-adjustable mobile truss bridge, including bridge main body, linear moving device and lifting device, the first end of bridge main body is slidably hinged on the hull along first direction, and the second end of bridge main body is used to be lapped with shore base;Linear moving device includes two mobile mechanisms oppositely arranged on the hull, one mobile mechanism drives bridge main body to move along first direction, and the other mobile mechanism drives bridge main body to move along the opposite direction of first direction;Lifting device is vertically arranged on the hull, and lifting device can lift or lower the second end of bridge main body.The utility model can drive bridge main body to extend or retract the hull by linear moving device, and lifting device can drive the second end of bridge main body to jack up or lower, to facilitate the angle of bridge main body adjustment, to adapt to different terrain and water level height, realize that bridge main body and different height shore base are lapped.

This invention belongs to the field of sparse matrix solving, specifically involving a SpMV optimization method based on heterogeneous platforms. The method first reads the sparse matrix data and constructs a compressed sparse row storage format for the matrix. Then, using matrix rows as clustering objects, hierarchical clustering and row / column rearrangement are performed based on global and local density features to form dense and sparse sub-blocks suitable for execution on heterogeneous platforms. Dynamic task scheduling is then combined to achieve load balancing between the CPU and DCU. Furthermore, to address the memory access bottleneck of the DCU platform, an LDS double-buffering mechanism is constructed, overlaying data transport and computation processes to reduce global memory access latency. Simultaneously, to address the Bank conflict problem in LDS shared storage, a joint optimization strategy based on padding and transpose access is designed to reduce the probability of conflicts during concurrent access by thread bundles. In addition, in multi-DCU collaborative scenarios, asynchronous communication and computation overlap mechanisms reduce intra-device synchronization waiting and communication overhead.

The utility model provides an edge beam splicing structure of a battery pack box body. The edge beam splicing structure comprises a first edge beam and a second edge beam, the first edge beam comprises at least three first transverse plates and two first longitudinal plates, the two first transverse plates and the two first longitudinal plates define a first frame body with a rectangular section, and the other at least one first transverse plate is arranged in the first frame body; the second edge beam comprises at least three second transverse plates and two second longitudinal plates, the two second transverse plates and the two second longitudinal plates define a second frame body with a rectangular section, and the other at least one second transverse plate is arranged in the second frame body; the first transverse plates and the second transverse plates are matched in a lap joint mode, the end face of at least one first transverse plate located in the first frame abuts against the inner side face of the second longitudinal plate to form a first seam, and the end face of at least one second transverse plate located in the second frame abuts against the inner side face of the first longitudinal plate to form a second seam. And the first seam and the second seam are connected through a welding process. The first transverse plate and the second transverse plate are in staggered lap joint, the strength of the splicing position in the z direction is improved, the connecting strength is improved by increasing welding seams, and the situation that the welding seams crack and fail to finally cause low structural strength and poor airtightness of the battery pack is avoided.

This invention belongs to the field of computer dataflow architecture design technology, and relates to a sub-core-level task collaborative scheduling method and scheduling management unit for heterogeneous dataflow architectures. The method includes: splitting computational tasks into multiple operators, and further refining each operator into multiple sub-cores to construct sub-core waiting queues; clustering operators into different collaborative execution units based on data dependencies and sharing patterns, and constructing sub-core running queues; dynamically monitoring the availability status of the processing unit array during runtime, and dynamically allocating non-conflicting processing unit resources to multiple sub-cores using a resource masking mechanism; updating the status of a sub-core after its execution and releasing the corresponding processing unit resources; iterating the above process until all queues are empty, at which point scheduling ends. This invention, through a sub-core-level task collaborative scheduling mechanism combined with data reuse technology, simultaneously deploys multiple sub-cores to the processing unit array for execution, achieving efficient resource utilization and computational parallelization, and significantly improving the overall system performance.

The utility model provides a kind of foldable handheld fan, including fan unit, handle and rotating support.Handle extends along length direction, and inside has accommodating cavity.Rotating support one end is connected with fan unit, and the other end is rotatably connected with handle through hinging structure.In handheld mode, define a minimum circumscribed cuboid space defined by fan unit and handle together.Rotating support at least a part is overlapped with handle in length direction, to shorten the combined span of fan unit and handle in length direction.At least a part of hinging structure is offset outside accommodating cavity in thickness direction, and simultaneously located in minimum circumscribed cuboid space.The utility model realizes rotating support and handle length overlap and hinging structure convex by rotating support rear shift offset, under the premise of keeping compact appearance, effectively shorten handheld overall length and release handle internal space.

The utility model discloses a foldable baby stroller which comprises a push handle, a first supporting leg frame, a second supporting leg frame and a connecting rod assembly. The push handle, the first supporting leg frame and the second supporting leg frame are all connected with the connecting rod assembly; when one of the push handle, the first supporting leg frame and the second supporting leg frame rotates, the other one of the push handle, the first supporting leg frame and the second supporting leg frame can be driven by the connecting rod assembly to rotate synchronously. And when the push handle, the first supporting leg frame and the second supporting leg frame rotate synchronously, the push handle, the first supporting leg frame and the second supporting leg frame are gradually close to each other or separated from each other. The push rod is rotated, so that the first supporting leg frame and the second supporting leg frame can directly and effectively rotate synchronously, and the first supporting leg frame, the second supporting leg frame and the push rod can synchronously rotate to be close to or away from each other, and therefore, the effect of driving the push rod, the first supporting leg frame and the second supporting leg frame to be overlapped or unfolded can be conveniently and quickly achieved.