Abstract: The present disclosure provides a thermal-stress-pore pressure coupled electromagnetic loading triaxial Hopkinson bar system and test method, the system mainly consists of an electromagnetic pulse generation system, a servo-controlled axial pressure loading system, a servo-controlled confining pressure loading system, a thermal control system, a pore pressure loading system, a bar system, and a data monitoring and acquisition system. Based on the conventional Hopkinson bar, the present disclosure creatively introduces a real-time loading and control system for confining pressure, thermal, and pore pressure, aiming to solve the technical problem that the existing test apparatus cannot be used to study dynamic response of deep rock mass under the coupling effect of thermal-stress-pore pressure and dynamic disturbance during dynamic impact loading.

Abstract: A display panel and a display device; the display panel includes a plurality of bonding pins. At least one bonding pin includes a first bonding electrode disposed on a side of an insulating layer group away from a base substrate, the insulating layer group having a first concave portion, the first bonding electrode having a second concave portion, an orthographic projection of the second concave portion on the base substrate being located within an orthographic projection of the first concave portion on the base substrate. The first bonding electrode includes at least a first conductor layer and a second conductor layer, and the at least one bonding pin further includes a filling portion disposed on a side of the second concave portion away from the base substrate, which is at least partially within the second concave portion.

Abstract: Methods, systems, and assemblies suitable for gas-phase processes are disclosed. An exemplary assembly includes a susceptor ring and at least one injector tube. The injector tube can be disposed within the susceptor ring to provide a gas to a peripheral region of a substrate. Methods, systems, and assemblies can be used to obtain desired (e.g. composition and/or thickness) profiles of material on a substrate surface.

Abstract: A uniaxial bidirectional synchronous control electromagnetic loaded dynamic shear test system and method, a test apparatus thereof including a support platform, a loading bar system, an electromagnetic pulse generation system, a servo-controlled normal pressure loading system, and a data monitoring and acquisition system. The test apparatus can be used to conduct a dynamic shear test research on a rock-like material under a constant normal pressure close to an actual operating condition, and can also be applied to carry out dynamic shear tests on intact rock-like test specimens in various sizes or jointed rock-like test specimens containing a single structural surface to study dynamic shear mechanical property and shear failure behavior under strain rate of 101?103 s?1, thereby providing an important theoretical and technical support for the design, construction, protection, and safety and stability evaluation of geotechnical engineering, structural engineering.

Abstract: Various implementations include in-ear wearable audio devices. In certain implementations, the in-ear audio devices include an eartip with a body having first and second ends, an inner wall extending between the first and second ends defining a hollow passage to conduct acoustic energy, and a deformable outer wall connected to the inner wall of the body at the first end and tapering away from the inner wall toward the second end, where the deformable outer wall is functionally graded from the first end to the second end to comply with an entrance of an ear canal of a user. In additional implementations, the eartip includes a retaining structure, and at least one of the inner wall or the outer wall of the body, or the retaining structure, has an integral electronic component and/or an electronic component signal trace.

Abstract: Embodiments provide anti-frost heaving supports and methods for crude oil pipeline crossing permafrost region. The anti-frost heaving support includes a support frame. An arc-shaped frame is installed on a top of the support frame. Two sets of adjusting devices are disposed on the arc-shaped frame, each set of the adjusting devices includes a pushing device fixedly installed on the arc-shaped frame. In the anti-frost heaving support, a piston plate compresses air inside the sealing frame so that a compressional force of an elastic plate and a stressed plate on the pipeline increases. When the compressional force increases, the elastic plate deforms to make a moving rod to move towards to an interior of the sealing frame. When moving on a tilted surface of a tilting plate, the moving rod pushes the tilting plate to move. The tilting plate pushes the sealing plate to be opened to release the air inside the sealing frame, thereby reducing the compressive force exerted by the stressed plate on the pipeline.

Abstract: A load balancing method, apparatus and device for a parallel model training task, and a computer-readable storage medium, includes: acquiring data traffic and a theoretical computational amount of each network layer in a target model; determining a theoretical computing capability of each computing device, and obtaining an initial computational amount corresponding to each computing device according to the theoretical computing capability and the theoretical computational amount; performing a load balancing operation according to the initial computational amount by using multiple device critical layer position division rule, so as to obtain a plurality of initial balancing schemes; compiling statistics on time performance parameters corresponding to the initial balancing schemes, and determining an intermediate balancing scheme from the initial balancing schemes according to the time performance parameters; and adjusting the intermediate balancing scheme according to the data traffic, so as to obtain a final

Abstract: A wearable device including a body having one or more embedded electronic components, the body further including a thermoset material having a polymeric backbone with at least one urethane linkage and a glass transition temperature. At a first temperature that is lower than the glass transition temperature, the body has an original shape. When the body is heated to a second temperature that is higher than the glass transition temperature, the body is deformable from the original shape to a first shape and when the body is cooled to a third temperature that is lower than the glass transition temperature, the first shape is maintained. The body is further configured to transition from the first shape to the original shape when the body is heated from the third temperature to a fourth temperature that is higher than the glass transition temperature.

Abstract: Disclosed are a method for realizing an nGraph framework supporting an FPGA backend device, and a related apparatus. The method includes: integrating an OpenCL standard API library into an nGraph framework; creating, in the nGraph framework, an FPGA backend device creation module for registering an FPGA rear-end device, initializing an OpenCL environment and acquiring the FPGA backend device; creating, in the nGraph framework, an FPGA buffer space processing module for opening up an FPGA buffer space and for reading and writing an FPGA cache; creating, in the nGraph framework, an OP kernel implementation module for creating an OP kernel and compiling the OP kernel; and creating, in the nGraph framework, an FPGA compiling execution module for registering, scheduling and executing the OP kernel.

Abstract: Disclosed are a method for realizing an nGraph framework supporting an FPGA backend device, and a related apparatus. The method includes: integrating an OpenCL standard API library into an nGraph framework; creating, in the nGraph framework, an FPGA backend device creation module for registering an FPGA rear-end device, initializing an OpenCL environment and acquiring the FPGA backend device; creating, in the nGraph framework, an FPGA buffer space processing module for opening up an FPGA buffer space and for reading and writing an FPGA cache; creating, in the nGraph framework, an OP kernel implementation module for creating an OP kernel and compiling the OP kernel; and creating, in the nGraph framework, an FPGA compiling execution module for registering, scheduling and executing the OP kernel.

Abstract: A method of enhancing session continuity for system interworking is proposed. Upon a UE performs inter-system change from A/Gb mode or Iu mode to S1 mode, the UE identifies that some PDN connection(s) are not associated with a PDU session ID (PSI). In one novel aspect, the UE deactivates all EPS bearer contexts for such PDN connection(s) locally, includes the EPS bearer context status IE in a TRACKING AREA UPDATE REQUEST message of a tracking area updating (TAU) procedure upon inter-system change from A/Gb mode or Iu mode to S1 mode, and then initiates UE requested PDN connectivity procedure(s) for such PDN connection(s). Furthermore, if no EPS bearer exists after local deactivation and EMM-REGISTERED without PDN connection is not supported by the UE, then the UE initiates a re-attach procedure followed by the UE requested PDN connectivity procedure(s) for such PDN connection (s) .

Abstract: Disclosed is a coated rubber particle, preparation thereof and its application as a plugging material in drilling fluids. The coated rubber particle comprises a rubber core and a coating layer, wherein the core has a fluorine content (at %) of 0-15%, the coating layer has a fluorine content (at %) of 30-80%, and the coated rubber particle has a 24-hour oil absorption rate of 4% or less. The coated rubber particle has good oil resistance, and is particularly suitable for use as a plugging material in drilling fluids, particularly oil-based drilling fluids.

Abstract: A load balancing method, apparatus and device for a parallel model training task, and a computer-readable storage medium, includes: acquiring data traffic and a theoretical computational amount of each network layer in a target model; determining a theoretical computing capability of each computing device, and obtaining an initial computational amount corresponding to each computing device according to the theoretical computing capability and the theoretical computational amount; performing a load balancing operation according to the initial computational amount by using multiple device critical layer position division rule, so as to obtain a plurality of initial balancing schemes; compiling statistics on time performance parameters corresponding to the initial balancing schemes, and determining an intermediate balancing scheme from the initial balancing schemes according to the time performance parameters; and adjusting the intermediate balancing scheme according to the data traffic, so as to obtain a final

Abstract: An ear tip includes a body configured to be mounted onto an earbud. The body includes a first end, a second end opposite the first end, and a first wall extending between the first and second ends. The first wall defines and surrounds a hollow passage that is configured to conduct sound waves. The first wall is configured to engage a nozzle on the earbud. The first wall includes a ring that is formed of a rigid material and engages the nozzle. The ring includes at least one C-shaped member with at least one gap and a compliant material is molded around the ring and fills the gap.

Abstract: An image processing method is provided, which is applied to a deep learning model. A cache queue is provided in front of each layer of the deep learning model; a plurality of computation tasks are preset for each layer of the deep learning model in advance, and are configured for computing weight parameters and corresponding to-be-processed data in a plurality of channels in each corresponding layer in parallel, and storing a computation result into a cache queue behind the corresponding layer thereof; in addition, as long as the cache queue in front of the layer includes the computation result stored in the previous layer, the layer can obtain the to-be-processed data from the computation result, subsequent computation is performed, and a parallel pipeline computation mode is also formed between the layers.

Abstract: An ear tip includes a body configured to be mounted onto an earbud. The body includes a first end, a second end opposite the first end, and an inner wall extending between the first and second ends. The inner wall defines and surrounds a hollow passage that is configured to conduct sound waves. The body also includes an outer wall that is connected to the inner wall at the first end and extends away from the inner wall toward the second end. The inner wall has an oblong cross-sectional shape that is configured to accommodate a corresponding nozzle on the earbud. The inner wall includes a ring that is formed of a rigid material and engages and conforms to the oblong shape of the nozzle, which inhibits improper mounting and rotation of the ear tip relative to the nozzle.

Abstract: An ear tip includes a body configured to be mounted onto an earbud. The body includes a first end, a second end opposite the first end, and an inner wall extending between the first and second ends. The inner wall defines and surrounds a hollow passage that is configured to conduct sound waves. The body also includes an outer wall that is connected to the inner wall at the first end and extends away from the inner wall toward the second end. The inner wall has an oblong cross-sectional shape that is configured to accommodate a corresponding nozzle on the earbud. The inner wall includes a ring that is formed of a rigid material and engages and conforms to the oblong shape of the nozzle, which inhibits improper mounting and rotation of the ear tip relative to the nozzle.

Abstract: The present disclosure discloses a method for preparing rare earth oxide by recycling ammonia and carbon, comprising the steps of: (1) heating raw materials containing a first rare earth carbonate and a first rare earth oxide with microwave and calcining at 500-1000° C. for 20-120 min to obtain a second rare earth oxide and carbon dioxide; (2) reacting carbon dioxide with a first ammonia water to obtain a precipitant; (3) reacting the precipitant with rare earth chloride to obtain a second rare earth carbonate and ammonium chloride wastewater. In the method, calcination time is short, rare earth recovery rate, utilization rate of ammonia and carbon resources are high. The present disclosure also provides a use of a rare earth oxide in shortening calcination time and/or increasing rare earth yield.

Abstract: A computing method and apparatus for a convolutional neural network model. The method comprises: acquiring a computing model of a training task of a convolutional neural network model (S101); then splitting multiply-accumulate operation in a computing model of a training task of the convolutional neural network model into a plurality of multiply-add operation tasks (S102); confirming a computing device corresponding to each multiply-add operation task according to the correlation between a preset computing model and the computing device (S103); and finally, respectively computing each multiply-add operation task by utilizing the computing device corresponding to each multiply-add operation task (S104). The purposes of improving the flexibility of migration of a CNN model training task on different computing devices or cooperative computing of different processors and improving the computing speed are achieved.

Abstract: An image processing method is provided, which is applied to a deep learning model. A cache queue is provided in front of each layer of the deep learning model; a plurality of computation tasks are preset for each layer of the deep learning model in advance, and are configured for computing weight parameters and corresponding to-be-processed data in a plurality of channels in each corresponding layer in parallel, and storing a computation result into a cache queue behind the corresponding layer thereof; in addition, as long as the cache queue in front of the layer includes the computation result stored in the previous layer, the layer can obtain the to-be-processed data from the computation result, subsequent computation is performed, and a parallel pipeline computation mode is also formed between the layers.