192results about How to "Improve reuse rate" patented technology

The invention discloses a calculation method for a nuclear reactor neutron transport equation, and belongs to the field of nuclear reactor core design and safety. According to the method, the problem that the convergence speed of an inner iteration process of a transport equation and a low-order equation in a traditional PI-CMFD method is low is solved. The method specifically comprises the following steps: step 1, reading a coarse mesh model, a fine mesh model, material information and boundary conditions of a nuclear reactor; step 2, establishing a numerical discrete model of a steady-state multi-group transport equation according to the fine grid model read in the step 1; 3, after boundary constraint and material constraint are applied to the established numerical discrete model of the steady-state multi-group transport equation, a discrete equation set is constructed by transforming a test function in the model; and 4, solving the constructed discrete equation set based on the grid mapping relation between the coarse grid model and the fine grid model by adopting a PI method and a CMFD method to obtain a neutron angular flux value and a characteristic value of the neutron transport equation. The method can be applied to calculation of the neutron transport equation of the nuclear reactor.

The invention provides a data retrieval system and method based on a large language model and a readable storage medium. The system comprises a data analysis module, a multi-mode integration module, a semantic retrieval module and a post-processing module which are arranged in sequence. The data analysis module analyzes the original engineering data to generate structured data; the multi-modal integration module maps the structured data to the same semantic vector space to construct a vector database, and establishes a cross-modal association index to construct a multi-modal engineering knowledge graph; the semantic retrieval module is based on a large language model and a retrieval enhancement technology, responds to a natural language query of a user, retrieves related information from a vector database and a multi-modal engineering knowledge graph and generates answers; the post-processing module performs abstract generation on the natural language answers based on a large language model and generates associated recommendation content. According to the method, the query accuracy, the retrieval efficiency and the user query interaction experience can be remarkably improved; and the scheme reuse rate can be improved based on the recommended content.

The invention provides a conversation system interaction method and device based on KV cache multiplexing, and the method comprises the steps: determining a reference cue word when a target cue word is received, and then determining an editing operation set corresponding to the minimum number of editing times needed for converting the reference cue word into the target cue word, and for each decoder layer, determining a unified attention map of the reference cue word in each decoder layer, and for each decoder layer, determining a target lexical element of the target cue word needing to recalculate the KV cache in the decoder layer according to the editing operation set and the unified attention map of the reference cue word in the decoder layer. According to the method, the target lexical elements are determined by editing the operation set and the unified attention graph, so that the target lexical elements, needing to be recalculated in the KV cache, of the target cue word in each decoder layer can be accurately recognized, and the KV cache of the reference cue word can be reused for other lexical elements, in each decoder layer, of the target cue word; the reuse rate of the KV cache and the response speed of the dialogue system are improved as much as possible.

The invention discloses a glass curtain wall construction system with a flexible cable driving parallel robot, and the system comprises a server which is in wireless connection with a terminal installed on a robot motion platform and a wireless communication module, and is used for receiving and transmitting data, obtaining a BIM model of a building needing to be installed, and constructing an overall installation module, and a process two-dimensional code is generated and pasted to the position where the glass curtain wall needs to be installed. The invention relates to the technical field of glass curtain wall installation. According to the glass curtain wall construction system and method of the flexible cable driven parallel robot, a camera and a marking two-dimensional code are used for cooperatively marking and installing each piece of glass, the rapidity and accuracy of installation are guaranteed, meanwhile, all data are recorded in a server, the installation process can be rapidly verified and modified, the reworking responsible degree is reduced, and the construction efficiency is improved. The problem that in the existing glass curtain wall installation process, the installation position is prone to occurring, reworking is caused, and the construction period is delayed is solved, meanwhile, the manual intervention process is added, and it is guaranteed that the whole installation process can be safely carried out.

The neural network forward prediction hardware accelerator based on the programmable logic device can be applied to the technical field of neural network hardware acceleration. The accelerator comprises a multi-layer hardware accelerator module unit, and the hardware accelerator module unit comprises an input feature map cache module and a data processing module; the accelerator module units are connected in an assembly line structure mode; multi-channel feature maps can be processed in parallel; and the output data of the hardware accelerator module unit of each layer is stored in the input feature map cache module of the hardware accelerator module unit of the next layer for the operation of the next layer. By adopting an assembly line structure, the throughput rate of the hardware accelerator is improved; a multi-channel parallel processing mode is adopted, and the running speed of a hardware accelerator is increased; and on-chip storage is adopted, so that the off-chip access times are reduced, and the power consumption of the hardware accelerator is reduced.

The invention discloses a data read-write method, device and equipment and a computer readable storage medium, which are applied to the field of data caching, and comprise the following steps: determining respective corresponding target sub-cache pools according to the capacity of each message in cached data, determining respective corresponding target cache blocks according to the capacities of the cache blocks in the respective corresponding target sub-cache pools, and writing the messages into the respective corresponding target cache blocks; and determining the released cache block according to the reading condition of the message, and placing the released cache block in the sub-cache pool of the corresponding level. According to the invention, the problems of low utilization rate and reuse rate of the cache space are solved.

The invention discloses a spacecraft electric propulsion system-level simulation model, which comprises an industrial personal computer (including test equipment), an EPCU simulator, a ground remote control and telemetering system, a storage and supply subsystem and a PPUamp, a TSU subsystem; the system comprises a ground remote control and telemetering system, a storage and supply subsystem and a PPUamp. The TSU subsystems are all real systems; the EPCU simulator comprises a ground remote control and telemetering system data interaction module, a storage and supply subsystem data communication module and a PPUamp; the TSU subsystem comprises a data communication module and a satellite-borne EPCU application module; the industrial personal computer is used for realizing task scheduling of the satellite-borne EPCU application module, calling of the ground remote control and telemetering system data interaction module, calling of the storage and supply subsystem data communication module and PPUamp; and calling a data communication module of the TSU subsystem. According to the method, the design scheme of the satellite electric propulsion system can be quickly verified on the premise that no satellite-borne electric propulsion control unit (EPCU) exists, and the method is suitable for verifying the feasibility of scheme design and the conformity condition of functional performance indexes of products selected by the system in the early stage of system design.

The invention discloses a neural network accelerator based on eDRAM in-memory calculation and a control method thereof, the neural network accelerator comprises a top layer controller and an extensible and reconfigurable SR-CIM macro array, and the extensible and reconfigurable SR-CIM macro array is operated in a global system mode. The method comprises the following steps: acquiring weight data and input activation data; setting a working mode of the extensible and reconfigurable SR-CIM macro array; and according to the working mode of the extensible and reconfigurable SR-CIM macro array, performing operation on the weight data and the input activation data, and transmitting an operation result to the off-chip neural network. According to the embodiment of the invention, the computing efficiency of the neural network can be improved by improving the reuse rate of the data and reducing the access times of the off-chip data in different data flow modes. The method can be widely applied to the technical field of neural network accelerators.

The utility model relates to the technical field of activated carbon production, in particular to a wastewater treatment system for activated carbon production, which comprises a rinsing tank, a phosphoric acid recovery tank, a pH regulating tank, an alkali liquor storage tank, a sedimentation tank, a precipitate recovery tank and a deep dephosphorization tank, the rinsing tank, the pH adjusting tank and the sedimentation tank are connected in sequence; the phosphoric acid recovery pool is connected with the rinsing pool; the alkali liquor storage tank is connected with the pH adjusting tank; the sedimentation tank is respectively connected with the sediment recovery tank and the deep dephosphorization tank. According to the wastewater treatment system for activated carbon production, provided by the utility model, the wastewater treatment cost is reduced while the recycling rate of activated carbon rinsing wastewater is increased, and the wastewater treatment system conforms to the green and environment-friendly recyclable development concept.

The invention provides a query page generation method and device based on data configuration, equipment and a medium, and relates to the field of data processing. According to the method, by calling the initial query component, it is ensured that a unified starting point exists in the development process, so that the consistency and maintainability of codes are kept, and meanwhile the development time is saved. The target query component is generated by configuring the query page attribute of the initial query component, the query interface can be quickly customized and adjusted according to different query requirements, consistency and reusability between different query pages are ensured, and the workload of repeated development is reduced. The query page of the target data is generated by utilizing the target query component, so that a new query function can be quickly deployed, manual layout and coding do not need to be performed every time, repeated work is reduced, the code reuse rate is improved, the development process is simplified, and the development efficiency of the query page is improved.

The invention provides a storage and calculation integrated structure and a chip, and belongs to the technical field of integrated circuits, and the storage and calculation integrated structure comprises at least one near storage core grain which is used for providing large-capacity storage; the in-memory computing unit is used for providing computing processing, and the near-memory core grains provide cache space for data and / or instructions processed by the in-memory computing unit; the at least one logic processing core particle is used for loading the data and / or instructions stored in the near memory core particle into the in-memory computing unit, and updating and storing the data and / or instructions processed by the in-memory computing unit into the near memory core particle; and the logic processing core particle is integrated with the near memory core particle and the in-memory computing unit through hybrid bonding and / or TSV (Through Silicon Via). The storage and calculation integrated structure is applied to high-performance calculation, the high-performance calculation requirement and the large-capacity storage requirement can be met at the same time, the high-performance calculation of the system is maximized, and support is provided for the development of the calculation power technology from the hardware structure.

The utility model relates to the technical field of rubber anti-aging agent production, in particular to chemical powder three-roller grinding equipment which comprises a three-roller grinding machine body, two threaded holes are formed in the upper side of the three-roller grinding machine body, one side of a scraping plate is arranged to be an arc-shaped face, one side of the scraping plate is attached to the surface of a discharging roller on the three-roller grinding machine body, and the other side of the scraping plate is arranged on the arc-shaped face. A reciprocating lead screw is arranged at the output end of a driving motor to rotate, the reciprocating lead screw drives a driving block to move, the reciprocating lead screw drives the driving block to move, the two threaded holes are fixedly provided with cleaning mechanisms, the two sides of a scraping plate are provided with scraping assemblies in a sliding mode, the inner wall of a moving groove is provided with a driving hole, and a driving unit is rotatably installed in the driving hole. The driving block drives the scraping plate to move, the scraping plate is attached to the discharging roller on the three-roller grinding machine body, redundant rubber anti-aging agents accumulated at the discharging roller are automatically scraped, manual scraping is not needed, manpower resources are saved, and the practicability of the device is improved.

The invention provides a rapid prototyping algorithm verification platform and method, and the method comprises the steps: receiving a configuration parameter inputted by a user through a configuration interface, and generating a configuration file based on the configuration parameter and a script written according to an AUTOSAR standard; building an algorithm model based on a function module which is selected by a user and meets the AUTOSAR standard; generating a model code corresponding to the algorithm model; and compiling and linking the configuration file and the model code in the underlying project to obtain a hex file. In the scheme, the configuration file and the model code conforming to the AUTOSAR standard are obtained by using the script written based on the AUTOSAR standard and each functional module obtained by packaging the ECU abstraction layer interface in the AUTOSAR standard, and are automatically integrated to the underlying project for compiling and linking, so that the code reuse rate is improved, the manpower is saved, and the efficiency is improved. And the purpose of shortening algorithm verification time is achieved.

The invention relates to a sea area management and control early warning evidence obtaining method based on video intelligent analysis, and the method comprises the steps: collecting equipment parameters: collecting internal and external parameters of equipment and environment parameters; equipment screening control is carried out, controllable and controllable equipment with a good detection effect is selected through equipment state analysis and target detection effectiveness analysis, and follow-up monitoring of a target area is realized through pan-tilt control; performing video intelligent analysis, determining a follow-up concerned target through mapping matching of a video intelligent analysis model and different coordinate system positioning information, and performing accurate linkage and video snapshot by fine tuning equipment; according to early warning and evidence obtaining association, structured attribute analysis is performed on a target and an area detected in a video capturing process, and an evidence obtaining result is added to early warning information with an association relationship, so that repeated evidence obtaining is avoided. According to the invention, the accuracy and efficiency of early warning evidence obtaining can be effectively improved, the utilization efficiency of data and resources is improved, and powerful technical support is provided for sea area management and control.

The utility model provides an auxiliary clamping jaw applied to a hydraulic chuck, and belongs to the technical field of machine manufacturing. The clamping jaw solves the problem that when the size of a clamped product is changed, the moving range of the clamping jaw is not good to correspond to the product needing to be clamped, and consequently the product is not stably installed. The auxiliary clamping jaw applied to the hydraulic chuck comprises a chuck body and a plurality of main base jaws, a plurality of standard oil pressure raw jaws are arranged on the chuck body and the corresponding main base jaws, and a driving assembly for driving the base jaws to move is arranged in the chuck body. A plurality of standard oil pressure raw claws are arranged on the chuck body, clamping spaces for clamping workpieces are formed among the standard oil pressure raw claws, auxiliary claws connected with the standard oil pressure raw claws are arranged between the standard oil pressure raw claws and the chuck body, sliding parts are arranged between the auxiliary claws and the chuck body, and positioning parts are arranged between the auxiliary claws and the standard oil pressure raw claws. The utility model has the advantage of being convenient to install and dismantle.

The invention discloses a masonry mortar joint control system and method. The system comprises a sand paving and trowelling unit used for controlling the thickness and uniformity of masonry mortar joints, a detachable guide rail shaping unit used for shaping the side faces of the masonry mortar joints and enabling the sand paving and trowelling unit to horizontally move, and a movable telescopic supporting unit used for controlling the working height of the sand paving and trowelling unit. The sand paving and trowelling unit and the detachable guide rail shaping unit are fixed through the movable telescopic supporting unit; the working positions and heights of the detachable guide rail shaping unit and the sand paving and trowelling unit are adjusted through the movable telescopic supporting unit, the sand paving and trowelling unit is driven to move on the detachable guide rail shaping unit, and control over masonry mortar joints is achieved through the control box jointly restrained by mortar and the detachable guide rail shaping unit.

The present invention provides an interface switching method, device and electronic device. If an interface switching controller receives an interface switching instruction, it first calls the switching interface in the interface set corresponding to the switched interface to play the switching animation corresponding to the switched interface on the current interface. When the target interface to be switched to does not exist, the creation interface in the interface set corresponding to the target interface is called to create the target interface, and the entry interface in the interface set corresponding to the target interface is called to play the entry animation corresponding to the target interface on the current interface, so that when switching to the target interface, the animation of the switched interface is played first, and then the animation of the target interface is played, thereby achieving seamless connection of the animation playback. In the present invention, it is only necessary to configure the interface sets of different interfaces to realize interface switching of different interfaces, reduce the amount of repeated code writing, and improve the code reuse rate.

The present disclosure relates to a firmware upgrading method and device, electronic equipment and storage medium. The method comprises: in response to an upgrading instruction of new media for firmware data to be upgraded, receiving filled-in structure data corresponding to the new media; the filled-in structure data comprises a preset callback function corresponding to the new media; the preset callback function is used to obtain preset upgrading data corresponding to the new media; in response to a first calling instruction, calling the preset callback function based on the filled-in structure data to obtain a preset check value in the preset upgrading data; comparing the preset check value with a current check value of the firmware data to be upgraded to determine current upgrading data corresponding to the firmware data to be upgraded from the preset upgrading data; and updating the firmware data to be upgraded based on the current upgrading data. According to the embodiments of the present disclosure, the code of the upgrading core layer does not need to be modified, the whole set of upgrading logic can be reused, the code reuse rate is improved, and the maintenance cost is reduced.

Embodiments of the present application provide a somatic intelligent robot control method, device and equipment based on skill tree, belonging to the technical field of intelligent robots. The method comprises: in response to receiving a task instruction, acquiring a current skill node and a target state indicated by the task instruction; matching the target state with a pre-constructed skill tree to determine a target skill node corresponding to the target state; searching for a feasible path from the current skill node to the target skill node in the skill tree, and generating a task plan composed of at least one skill node according to the feasible path; and calling skill operations encapsulated by each skill node in sequence according to the task plan to drive the somatic intelligent robot to complete the task instruction. In the above manner, the target state is mapped to the skill tree node at one time, and then the feasible path is searched, so that the task plan is automatically generated, and the reuse rate of the skill of the somatic intelligent robot is improved.

The invention discloses a single-chip microcomputer multi-engineering algorithm optimization method and system and a related device, which are used for improving the reuse rate of codes. The method comprises the steps that an algorithm space is divided in link files of at least two projects of the single-chip microcomputer, the algorithm space is used for storing and sharing an algorithm library in an RAM memory area, and the algorithm library comprises all algorithm codes in the at least two projects; compiling the algorithm library to obtain a library file and storing the library file in a Flash storage area; when the first project runs, loading the library file into the algorithm space from the Flash storage area so as to execute the required algorithm code; and when the subsequent project runs, calling the library file from the algorithm space to execute the required algorithm code.

The invention relates to a communication control method and device, computer equipment, a readable storage medium and a program product, which are applied to a control unit, are arranged in a system in which a ground terminal and a low-altitude terminal coexist, and are used for dynamically combining a cellular unit and a cellular-free unit in the system. The control unit is in communication connection with all cellular base stations on a flight path of the low-altitude terminal; the method comprises the following steps: determining a target cellular base station meeting requirements from adjacent cellular base stations according to network performance index values of the adjacent cellular base stations located around a low-altitude terminal and sent by the low-altitude terminal; adjusting the type of the target cellular base station to a mixed base station comprising a cellular unit and a non-cellular unit; wherein the ground terminal is served by the cellular unit, the low-altitude terminal is served by the non-cellular unit, and wireless resources are dynamically shared between the cellular unit and the non-cellular unit in the hybrid base station. By adopting the method, the same-frequency interference when the ground network and the low-altitude network coexist can be reduced.

The invention provides an image processing method and device based on element-by-element operation, equipment and a medium, and relates to the technical field of GPUs (Graphics Processing Unit), and the method comprises the following steps: according to the number of a plurality of elements contained in a plurality of input tensors needing to participate in operation and hardware resource information of GPU equipment, determining the number of the plurality of elements contained in the plurality of input tensors needing to participate in operation; determining the number of thread blocks required to be used by the plurality of input tensors in the element-by-element operation process; according to the number of the thread blocks, the number of the elements and the memory layout of the multiple elements, kernel functions of element-by-element operation class operators corresponding to the multiple input tensors needing to participate in operation are determined on the basis of a predefined template structure body, and parameters of the template structure body comprise element types and the number of the elements; and based on the number of the thread blocks, initiating a plurality of threads to enable the plurality of threads to obtain the plurality of elements according to the kernel function, and executing an element-by-element operation process of the plurality of input tensors to obtain an image processing result.

The utility model discloses a kind of polysilicon hydrogenation device silicon powder high-efficiency recovery system, including fluidized bed reactor, further including inner cyclone separator, double cyclone separator, metal filter and silicon powder collecting tank;Multiple inner cyclone separators are fixedly arranged in the top inside fluidized bed reactor by fixed support;The air outlet of multiple inner cyclone separators is communicated with the air inlet of gas collector, and the air outlet of gas collector is communicated with the reaction gas outlet in the top of fluidized bed reactor;The reaction gas outlet of fluidized bed reactor is communicated with the air inlet of double cyclone separator by pipeline, and the air outlet of double cyclone separator is communicated with the air inlet of metal filter by pipeline.The advantage is that: by three-stage gradient separation, silicon powder recovery efficiency is greatly improved;Silicon powder separated by double cyclone separator can be returned to fluidized bed reactor dense phase region through return chute pipe, realize raw material recycling, reduce silicon powder waste.

The present invention belongs to the technical field of computer technology server-side data processing, and discloses a templated business cache processing method and system. For a relational database using structured data, this method completes data storage in the relational database by defining a fixed data model as a cache data structure; the data stored in the cache is converted into a usable binary format in a serialized manner; an abstract interface for the data structure and an abstract interface for data operations are defined, and cache commands are performed on the detailed problem data used by the template class. The present invention uses a template to encapsulate the specific implementation of the operation steps into the template class, so that a general processing that supports multiple data types can be used to improve the code reuse rate and reduce redundancy. At the same time, the template class can be easily expanded to support new data types, and the unified maintenance method in the template class makes the code easier to maintain and expand.

The utility model discloses a hand restraining and protecting structure for a critical patient, and relates to the technical field of medical care instruments. The protective sleeve is used for isolating and protecting the hand of a patient, the restraint strap is used for fixing the protective sleeve to the wrist of the patient, the protective sleeve is of a trumpet-shaped structure, and the small opening end of the protective sleeve is fixedly sewn or detachably connected with the restraint strap; a fixing band for binding the restraint strap on the wrist of a patient is arranged on the outer side of the restraint strap, and a hand limiting rope is fixedly connected to the outer side of the restraint strap. By arranging the protective sleeve, the fingers and the palm of a patient are covered and isolated, the situation that the patient scratches all around in a fuzzy consciousness or dysphoria state, and a monitoring data line and a medical instrument catheter are damaged is avoided, meanwhile, the trumpet-shaped protective sleeve also guarantees relative freedom of the fingers and the palm, and the problem of unsmooth ventilation does not exist; the protective sleeve is arranged to be of a specification-adjustable structure, so that the constraint protective structure can be adaptively adjusted according to the size of the hand of the patient, and the scene adaptability of the constraint protective structure is enhanced.

The utility model belongs to mould technical field especially is a kind of multi-product common die structure, including lower bolster plate, the upper surface both sides of lower bolster plate are all fixedly connected with lower foot pad, the top of lower foot pad is fixedly connected with lower die holder, the upper surface of lower die holder is installed with lower backing plate, the upper surface of lower backing plate is fixedly connected with guide pillar.The utility model can be punched to product one by the setting of drawing punch, punching punch, chamfering punch, ejector pin one and ejector pin two, and it can be replaced design, can be replaced as drawing punch two, punching punch two, chamfering punch two, ejector pin five and ejector pin six, so as to be able to punch to product two, so as to realize when switching product without overall replacement mould, greatly shorten product switching time, improve production flexibility, also greatly improve the reuse rate of mould, reduce equipment investment and cost.

The invention provides a vehicle charging system, a vehicle charging control method and a vehicle. The system comprises an energy conversion device, a direct current output device and a controller. Wherein the energy conversion device and the direct current output device are connected with each other, and the energy conversion device and the direct current output device are both connected to a battery of the vehicle; the energy conversion device comprises a multi-phase inverter, a multi-phase winding and a first capacitor, wire inlet ends of the multi-phase winding are correspondingly connected to multi-phase bridge arms of the multi-phase inverter respectively, and the first capacitor is connected to the multi-phase inverter in parallel; the controller is connected to the multi-phase inverter and controls multi-phase bridge arms of the multi-phase inverter to act so as to form direct current with alternating current components to charge a battery of the vehicle. When the vehicle battery is charged through direct current, the alternating current component is superposed in the direct current, so that the service life of the vehicle battery can be effectively prolonged while the charging speed of the vehicle battery is ensured.

The present invention discloses a method for generating target sample annotations, which belongs to the field of deep learning technology and solves the problem of multiplexing in a neural network of target samples with different image sizes. The method includes: cutting a large image according to a preset overlap ratio, a pre-cut small image size, and a preset external expansion ratio; marking the target samples in the cut small images to obtain the relative coordinates of the annotation points on each small image; determining the absolute coordinates of each annotation point on the large image according to the relative coordinates corresponding to the cutting position of the small image relative to the large image, the relative coordinates on each small image, the preset overlap ratio, the pre-cut small image size, and the preset external expansion ratio, and obtaining a large image containing the absolute coordinates of the annotation points.

A digital project task management cooperation platform system based on cloud deployment comprises a data layer, a service layer, an interface layer and an access layer. The service layer comprises service configuration, an intelligent task engine, basic service and task service; the service layer provides diversified contract service interfaces for the access layer; the intelligent task engine provides standard task management service; each basic service is integrated, and a universal intelligent task service system is realized; the intelligent task engine processes wide task types; the state and progress of a task are managed through a task service, document data related to the task are stored and managed through a file service, and smooth communication between task participants is ensured through a message service; and finally, key information in the document is automatically extracted through an intelligent identification technology. The platform system can greatly improve the conditions of information asymmetry, unsmooth communication and low management efficiency in the existing project management.

The invention provides a work content and progress plan association method, system and device based on a BIM model. The method comprises the steps that the BIM model of a project is acquired; obtaining a progress plan library of the project; performing dynamic code association based on the component identification code of each component in the BIM model and the progress plan code in the progress plan library to obtain an associated BIM model; acquiring a post work content library; the post work content code is consistent with the progress plan code; and based on the progress node of the associated BIM model, triggering a post work task corresponding to a post work content code which is the same as the progress plan code in the post work content library, and sending the post work task to a corresponding post worker. According to the invention, the post work content and the progress plan can be associated by using the component identification code, the progress plan code and the post work content code based on the BIM model, so that the project construction management efficiency is improved.