62results about How to "Flexible allocation" patented technology

The invention discloses a power efficiency optimization method based on intelligent load management. The method comprises the following steps: S1, generating an initial load characteristic matrix; s2, decomposing the initial load characteristic matrix into a short-term behavior characteristic matrix and a long-term trend characteristic matrix through pseudo-symmetry decoupling; s3, multi-level load demand prediction is carried out through a fusion structure of the multi-scale convolution and the recurrent neural network; s4, coding a cognitive gene sequence, triggering gene mutation and recombination operation when significant change occurs, and generating a load distribution strategy; s5, a load distribution strategy is transmitted into a causal reasoning and self-supervision feedback control module, and if the deviation between feedback data and a prediction result exceeds a threshold value, a self-learning mechanism is triggered; s6, dynamically distributing the load priority, and generating a decision path of load management; and S7, performing adaptive optimization through a multi-layer feature feedback closed-loop optimization system. The intelligent load management method is adopted, and the power efficiency is optimized through multi-layer feature decoupling and closed loop.

The invention relates to the technical field of communication networks, and discloses a service channel time slot allocation method, which comprises the following steps of: 1, acquiring network data in real time from network equipment, the network data comprising bandwidth utilization rate, delay, packet loss rate and service flow type; 2, predicting future network flow fluctuation through a long-short-term memory network model based on historical flow data and real-time network data; and step 3, dynamically adjusting the allocation of time slot resources according to the network flow fluctuation prediction result, and preferentially allocating the time slot resources for high-priority services. According to the method, the dynamic adjustment of the time slot resources is realized by combining the flow prediction of the LSTM model and the Q-learning reinforcement learning algorithm, so that the system can flexibly allocate the resources according to the actual network load condition, the resource waste is reduced, the utilization efficiency of the bandwidth is improved, and the utilization rate of the overall network resources is remarkably improved especially under the condition of high network load.

The invention provides a multi-agent cooperation system, which comprises a multi-agent cooperation engine module, a task decomposition and distribution module and a tool calling normal form module, and is characterized in that the task decomposition and distribution module is used for decomposing tasks and distributing decomposed sub-tasks to agents; the multi-agent cooperation engine module is used for managing cooperation and communication between agents; and the tool calling module enables the intelligent agent to dynamically call different tools. According to the multi-agent collaboration system, the collaboration efficiency among the agents, the flexibility of task decomposition, the adaptivity of tool calling and the intelligent level of user interaction are remarkably improved, and remarkable technical advantages and effects are shown in complex task processing and dynamic environment processing.

The invention provides a resource adjustment method and device for edge computing nodes, equipment and a medium, and belongs to the technical field of computers.The method comprises the steps that the use condition and service scene data of target resources associated with all the edge computing nodes are monitored, the target resource comprises at least one of a central processing unit (CPU), a memory and a network bandwidth; according to the monitored data, a target scheduling strategy is selected from a pre-established scheduling strategy library, and the scheduling strategy library comprises resource scheduling strategies for different service scenes and resource use conditions; and according to the target scheduling strategy and the current use condition of the target resource by each edge computing node, adjusting the resource currently allocated to each edge computing node.

The invention provides a three-dimensional scene rendering method and device, equipment and a medium, relates to the field of computers, in particular to the technical field of page development, and can be used for application scenes such as generative search, intelligent document editing, intelligent assistants, virtual assistants and intelligent e-commerce. The method comprises the following steps: determining a plurality of elements in a to-be-rendered three-dimensional scene; based on a preset rule, putting the plurality of elements into a plurality of independent rendering layers; determining respective rendering priorities of the plurality of rendering layers; and based on the corresponding rendering priorities, sequentially rendering the plurality of rendering layers.

The invention relates to a task processing method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring data processing tasks from an edge processing node, and acquiring task demand information corresponding to each data processing task; acquiring node state information of each processing node in the computing power network; according to the node state information and task demand information carried by the data processing task, determining a target processing node by taking resource utilization rate maximization of each processing node in the computing power network as a target; determining a target transmission path and a target transmission protocol by taking maximization of transmission bandwidth and minimization of transmission delay as targets according to data size information and time delay requirements carried by the data processing task, and a network state and a transmission protocol of a target processing node; and transmitting the data processing task to the target processing node through the target transmission path and the target transmission protocol. By adopting the method, the task processing efficiency can be improved.

The invention discloses an industrial internet trusted data communication method and system, and relates to the technical field of data communication, and the method comprises the steps: taking an equipment end and an employee end as data collection points, building an initial trust value for each data collection node, and obtaining a dynamically adjusted trust value for each data collection node according to the historical performance and real-time state of the data collection node; classifying and aggregating the collected data according to the data type and the trust value; dynamically allocating bandwidth resources according to the trust value and the data type; after data transmission is completed, a bidirectional confirmation mechanism is started, a sender and a receiver confirm whether the data is successfully sent and received through a feedback mechanism, and if the sender does not receive a confirmation message within a specified time, a timeout mechanism is triggered, and a data packet is sent again until the data packet is confirmed to be received; and calculating an average transmission success rate, and dynamically adjusting strategies of data acquisition, compression and bandwidth allocation.

The invention relates to the technical field of embedded systems, in particular to a data backup and recovery method applied to an embedded device, an electronic device and a storage medium, comprising: in response to creation of a user partition and a backup partition, storing data in the user partition, and backing up the data stored in the user partition in the backup partition; wherein the user partition comprises a plurality of user sub-partitions, the backup partition comprises a plurality of backup sub-partitions, and each user sub-partition and each backup sub-partition store verification files; when it is confirmed that data consistency verification is started, whether data stored in the user sub-partition is consistent with data stored in the corresponding backup sub-partition or not is verified through the verification file; and if the user sub-partitions and the backup sub-partitions are not consistent, data recovery operation is performed, so that the data stored in the corresponding user sub-partitions and the backup sub-partitions are kept consistent, and the defects of an existing data backup and recovery mechanism of the embedded system are effectively overcome.

The invention relates to the technical field of electric power communication, in particular to an electric power communication planning method, system, equipment and medium based on cross-layer slicing, and the method comprises the steps: classifying all services in an electric power communication network, and obtaining the service type of the electric power communication network; dividing the power communication network resource into a plurality of slices among a plurality of network layers according to the service communication key index of each service type and the available state of the current network resource, and forming a cross-layer slice model comprising the plurality of network layers; determining cross-layer slicing resources of the cross-layer slicing model by using a reinforcement learning algorithm according to the service communication key indexes of the service types; and searching an optimal power communication path by using a path optimization algorithm according to the current power communication network state and the cross-layer slice resources. Through the cross-layer slicing model and the reinforcement learning algorithm, dynamic scheduling and efficient management of power communication network resources are realized, and the utilization rate of the network resources is improved.

The invention belongs to the technical field of medical detection, and provides an automatic specific protein analyzer for comprehensive multi-item detection, which mainly comprises an online conveying track, an urgent treatment assembly, a double-sample-arm sampling mechanism, a sample disc, a reagent disc, a reaction disc, an ISE electrolyte module and a double-light-path detection module. The online conveying track is used for automatic detection and transmission of samples and online connection of adjacent detection instruments, the urgent treatment assembly is used for urgent detection of the samples, the double-sample-arm sampling mechanism is used for selective sampling of different samples, the ISE electrolyte module is used for electrolyte detection, and the double-sample-arm sampling mechanism is matched to achieve electrolyte module detection. The detection sequence and speed of other samples are not delayed; the double-light-path detection module is used for detecting different samples, the photosensitive sensitivity is improved, and the detection efficiency is improved. By adopting the structure, the detection requirements and sampling conditions of a current detection instrument on different samples are met, and the compatibility and detection efficiency of the detection instrument are improved.

The invention discloses a hybrid energy storage power distribution method based on a power grid frequency modulation demand, and relates to the field of hybrid energy storage frequency modulation. The method comprises the steps that when a primary frequency modulation signal instruction and a secondary frequency modulation signal instruction are consistent in direction, a first total response instruction of the hybrid energy storage system and the working condition of the hybrid energy storage system are determined; based on the working condition of the hybrid energy storage system, aiming at prolonging the continuous regulation capability of the hybrid energy storage system, distributing a power instruction for each energy storage unit according to a first total response instruction of the hybrid energy storage system; when the directions of the primary frequency modulation signal instruction and the secondary frequency modulation signal instruction are opposite, taking the primary frequency modulation signal instruction as a second total response instruction, and distributing an initial power instruction for each energy storage unit; dynamically adjusting an initial power instruction of each energy storage unit by using a fuzzy controller; and determining a final power instruction of each energy storage unit according to the adjusted power instruction and rated power of each energy storage unit. The frequency modulation accuracy and stability can be improved, and the energy loss can be effectively reduced.

The invention provides a photovoltaic power generation hydrogen production control method and system, a program product and a storage medium, and relates to the technical field of photovoltaic power generation. The system determines whether adjustment is needed or not by judging the relation between the current illumination parameter and the preset threshold value. In the adjusting state, the system calculates the total energy receiving power, and dynamic adjustment of the target power point is achieved. When the target power point is larger than the total energy receiving power, the system analyzes according to the temperature of the electrolytic cell and calculates needed energy and excess energy, so that whether the temperature of the electrolytic cell is increased or the working voltage point is adjusted is determined. According to the dynamic adjustment mechanism based on multiple parameters, the system can flexibly distribute energy according to real-time working conditions, the situation that electric energy is forced to be connected to a grid is avoided, meanwhile, the performance of the system is improved through temperature adjustment when the hydrogen production power is not high, and finally the energy utilization efficiency is improved.

The invention discloses a virtuality and reality combined method and system for dynamically reconstructing diversified network topology, and belongs to the technical field of novel power systems, and the method comprises the following steps: virtualizing physical equipment in a power system business scene network by using NFV to obtain virtual equipment; constructing a network architecture for realizing a service scene network of the power system by adopting a virtualization mode; physical equipment and virtual equipment in the service scene network of the power system are controlled in a centralized manner based on the SDN; and the SDN controller dynamically reconstructs the service scene network of the power system according to a set strategy. According to the invention, dynamic allocation of network resources and real-time reconstruction of topology are realized through a virtuality and reality combination technology, a software defined network technology and a network function virtualization technology, and virtualized resources and resources of physical equipment can be effectively integrated so as to quickly respond to changes of application requirements.

The invention provides a power management system and a control method, which are applied to the technical field of power electronics. The power management system comprises a power switching module, a control module and a driving module, the power supply switching module comprises at least two input paths, and each input path is used for receiving different input voltages; wherein the at least two input channels comprise a main input channel. The control module is used for acquiring input parameters of the main input path; and the control module is used for receiving the input parameters of the main input channel, judging whether the current input state of the power management system is abnormal according to the input parameters of the main input channel, and if so, sending a control signal to the power switching module to cut off the main input channel and conduct any other channel except the main input channel. On the basis, the condition of multi-power-supply input can be maintained, the electrical system can more flexibly distribute electric power resources, meanwhile, after one power supply breaks down, input adjustment of the power supplies is achieved, normal operation of the electrical system is guaranteed, and the overall efficiency of the system is improved.

The present application discloses a detachable mobile energy storage device management system, method, equipment and medium, which mainly relates to the field of energy storage device application technology, and is used to solve the existing mobile energy storage system that contains only one energy storage module, which affects efficiency, disconnects the node at the fault, has poor power supply continuity and low distributed power supply utilization efficiency, and the mobile route of the existing mobile energy storage system has an unbalanced power supply problem. It includes: an island division component to obtain a number of power grid islands; a route division component to obtain the calculated difference of distributed power sources; determine the priority of distributed power sources according to weights and calculated differences; complete the connection between the detachable energy storage component and the distributed power source; obtain the calculated difference corresponding to the distributed power source in real time, and when it is less than the preset difference threshold, the distributed power source is disconnected from the detachable energy storage component until an extreme fault elimination signal is received, or the calculated differences of the distributed power sources are all less than the preset difference threshold.

The invention discloses a large model reasoning method and device based on cross-domain edge device collaboration, and the method comprises the steps: dynamically sensing the available resource condition of a graphics processor in a plurality of working nodes based on a distributed hash table, and distributing a large model block to the working nodes according to the available resource condition; the available resource condition represents an unallocated working node set; receiving a user reasoning request, wherein the reasoning request at least comprises one or more reasoning tasks; based on the large model blocks allocated by the working nodes, allocating the reasoning tasks to one or more management nodes; an area where each management node is located comprises a working node set; the management node distributes the reasoning task to the working node set of the region for reasoning; the working node set comprises a plurality of working nodes. By means of the method and device, the efficiency of the distributed reasoning task can be improved, and flexible allocation of resources and high reliability of the system can be achieved.

This utility model discloses a regional energy supply system for peak shaving of renewable energy power generation, belonging to the field of energy technology. It includes an energy supply system configured as a multi-heat source heat pump unit, which is connected to a load system, a supplementary energy system, and an energy exchange system. The multi-heat source heat pump unit includes a condenser, a water tank, a compressor, and a heat pump controller. The refrigerant inlet of the condenser is connected to the water tank and the load system, and the refrigerant outlet of the condenser is connected to the compressor, the load system, and the water tank. This utility model efficiently integrates multiple renewable energy sources, achieving stable and flexible energy supply, promoting peak shaving and absorption of renewable energy, reducing power curtailment, and demonstrating good economic and environmental benefits.

The application discloses a staggered cross type airing quilt rod unfolding structure, an unfolding method and an airing machine. The staggered cross type airing quilt rod unfolding structure comprises: an airing rod assembly having an airing area extending along a first direction; and an airing quilt rod assembly movably connected to the airing rod assembly and having a storage state and an unfolding state. In the storage state, the airing quilt rod assembly is folded above the airing rod assembly. In the unfolding state, the airing quilt rod assembly is located below the airing rod assembly, and the extending direction of the airing quilt rod assembly intersects with the first direction, so as to form a staggered cross type airing space. The application has the advantages that the staggered cross type layout fundamentally solves the interference problem of clothes and quilts; the height difference is properly designed, so that the compact structure and long clothes airing are considered; and the cross structure breaks the fixed space distribution mode.

The present invention discloses a hybrid accelerator for convolutional neural networks and a load balancing method thereof, which belongs to the technical field of deep learning hardware acceleration, and is used to solve the technical problem that the two current dedicated accelerator architectures for convolutional neural networks cannot realize the flexible allocation and efficient utilization of computing resources of the computing platform, and the improvement of the computing performance of the convolutional neural network is limited. The hybrid accelerator includes: a single-layer single-engine architecture acceleration component and a single-layer multi-engine architecture acceleration component; the single-layer single-engine architecture acceleration component and the single-layer multi-engine architecture acceleration component are connected through a double buffer; the single-layer single-engine architecture acceleration component and the single-layer multi-engine architecture acceleration component are used to process different batches of input data at the same time; wherein the input data is the data to be calculated of each convolutional layer of the convolutional neural network; the double buffer is used to store the output data of the single-layer single-engine architecture acceleration component and the single-layer multi-engine architecture acceleration component.

The present invention discloses a scheduling method for a charging drone to charge a mission drone in the air, which is used to schedule the charging drone to charge the mission drone that is performing a mission in the air. The method determines the remote charging or near-field charging mode according to the charging demand of the mission drone, and then performs classification modeling and calculation according to the number of participating charging drones, including establishing a model for the charging scheduling problem of a single charging drone and establishing a model for solving the charging scheduling problem of multiple charging drones; finally, the charging scheduling strategy of the charging drone is optimized based on the algorithm of deep reinforcement learning or multi-agent reinforcement learning, and the charging drone makes a decision based on the optimized strategy and the state of the current environment, and completes the charging of the mission drone according to the charging scheduling instruction. The present invention realizes the shortest task completion time of the mission drone, the fair charging of the mission drone, and the optimization of the charging scheduling strategy of the charging drone.

The invention provides a service access method and equipment. The method is applied to a first device, and the method comprises the following steps: obtaining a first corresponding relationship between a plurality of service names and a plurality of service identities (ID); a first service ID corresponding to the first service name is determined according to the first corresponding relation, and the first service name comprises the service name of the source service and / or the service name of the target service; and sending a first message to a second device where the target service is located, wherein the first message comprises the first service ID. In the technical scheme, when service access is carried out, service information can be directly associated by carrying the service ID in the first message, so that a manager can configure a network security policy based on the service information, and a source service or a target service of service access can be limited to improve network security.

The invention relates to the technical field of battery containers, in particular to a container type control room and energy storage battery cabin combined heat management system and method.The system comprises a refrigerant loop and a secondary refrigerant loop which can conduct heat transfer, and the refrigerant loop is connected to the inner side and the outer side of a control room in series to adjust the temperature and humidity of the control room; the secondary refrigerant loop is connected with the battery cluster in the battery cabin and is used for adjusting the temperature of the battery cluster; the refrigerant loop comprises an indoor unit and an outdoor unit which are connected in series, the indoor unit is arranged in the control room, the outdoor unit is arranged outside the control room, and an outdoor unit assembly can exchange heat with the secondary refrigerant loop; the battery compartment and the control chamber are arranged in the container. The mechanism is compact, the battery cabin heat management loop and the control room heat management loop are fused, energy cross-region transfer can be achieved, and cold or heat can be flexibly distributed according to the real-time temperature control requirements of the battery and the control room.

The invention provides an access method of an AMP multi-core heterogeneous system and the AMP multi-core heterogeneous system, the AMP multi-core heterogeneous system comprises a first core, a second core, a shared memory and a mailbox, the first core is internally provided with a physical serial port and a first virtual serial port which are communicated with each other, the second core is internally provided with a second virtual serial port, and based on the virtual serial ports, efficient and real-time tty data exchange is realized. A basic serial communication function is realized between the master core and the slave core, and the slave core can access the physical serial port of the master core through the virtual serial port equipment, so that the peripheral serial port equipment can be accessed through the physical serial port of the master core, and the configuration and data communication of the slave core to the serial port equipment are realized; therefore, not only is the resource competition of the hardware interface between the master core and the plurality of slave cores reduced, but also the flexible distribution and scheduling of the UART serial data flow are achieved, so that the requirements of various communication modes of the master core, the slave cores and the UART physical serial port are met.

The utility model relates to a ventilation equipment field provides a kind of air duct assembly, air conditioning system and vehicle.Wind channel component includes air supply device, air supply device includes wind box, and wind box is opened with air inlet and air outlet, air inlet is provided with filter piece, and the position corresponding with air inlet in wind box is provided with air guide piece;Distribution box, fluid communication is in air outlet, and distribution box is provided with valve group, and distribution box includes first air duct, second air duct and third air duct, first air duct is fluid communication with the driver's room of vehicle, second air duct is fluid communication with the traction motor cabin of vehicle, third air duct is fluid communication with the passenger room of vehicle, and valve group is used to distribute the air intake of first air duct, second air duct and third air duct.The air duct assembly can make the airflow output of each air duct stable, avoid the air supply fluctuation caused by external wind pressure variation, ensure the comfort of driver's room, passenger room and the heat dissipation effect of traction motor cabin, improve the reliability of vehicle operation, simplify the structure of air duct assembly.

An embodiment of the present application discloses a storage space management method, apparatus, device and storage medium, wherein the storage space includes at least two storage areas, and each storage area can respond to a different range of space request amounts. The method includes: receiving a space allocation request sent by a thread, and determining the space request amount of the space allocation request for the storage space; based on the space request amount and the request amount range corresponding to each storage area, determining a legal application area in the storage space; the legal application area includes at least one storage area; determining a target storage domain segment in the legal application area, and allocating the target storage domain segment to the thread.

The invention relates to the field of virtual network mapping, and discloses a virtual network mapping method and a related system.According to the virtual network mapping method and system, virtual network requests are divided into a detachable type and a non-detachable type, and different types of requests are processed in a classified mode, so that optimization processing can be better conducted according to the characteristics of each type of request. Therefore, the resource utilization rate of the detachable requests can be improved, and the service quality of the non-detachable requests can be ensured. According to the method, the virtual network request can be split for recombination, and similar requests can be merged, so that the waste of resources is reduced, and the utilization efficiency of the resources is improved. According to the virtual network request received in real time and the change condition of the physical network resource, the mapping scheme is dynamically adjusted, so that the network resource allocation is more flexible. And the occupation condition of physical network resources is calculated in real time, the bottleneck of resource utilization can be found in time, and the problem is solved through subsequent adjustment and optimization.

The invention discloses a bandwidth management method and device based on a multi-wave WDM-PON system, equipment and a medium, and relates to the technical field of optical communication. The method is applied to an optical line terminal OLT in a multi-wave WDM-PON system, the single-port MAC bandwidth of the OLT is divided into a plurality of bandwidth time slots SLOTs, and the method comprises the following steps: distributing each SLOT to a plurality of supported wavelength channels; and when valid data are written into the SLOT, performing time slot allocation on the SLOT based on the valid data. By implementing the technical scheme provided by the invention, the overall bandwidth utilization rate of the multi-wave WDM-PON system can be improved, and the system can utilize available bandwidth resources more efficiently, so that the problem of low utilization rate caused by unbalanced bandwidth resource allocation in the prior art is solved.

The invention provides a satellite frequency plan arrangement method. The method comprises the following steps: acquiring beam target parameters; respectively calculating an uplink bandwidth demand and a downlink bandwidth demand corresponding to each beam; calculating a beam uplink allocation bandwidth combination and a beam downlink allocation bandwidth combination; determining an uplink allocation bandwidth and a downlink allocation bandwidth corresponding to each beam; performing letter coding on the uplink and downlink selectable frequency bands of each wave beam according to a preset coding rule; constructing a beam uplink topological matrix and a beam downlink topological matrix; wherein elements in the matrix are respectively used for representing an uplink topological relation and a downlink topological relation between beams; sequentially allocating corresponding uplink frequency bands and downlink frequency bands to the beams; and performing reverse decoding on the frequency band codes of the allocated uplink frequency band and the downlink frequency band based on a preset coding rule to obtain a corresponding uplink real frequency band and a corresponding downlink real frequency band. Thus, the user demand matching rate is improved, the user service quality is guaranteed, and meanwhile the better satellite resource utilization rate is achieved.

The invention relates to the technical field of petrochemical engineering, in particular to a pre-flash tower bottom oil diversion and emergency oil return system which comprises a switching and shunting module, a pressure mixing module and a cooperative control module. Boosting the pressure of the slurry oil and then dividing the slurry oil into a first branch flow and a second branch flow; the first branch is communicated with a first conveying pipeline, the first conveying pipeline is communicated with a wax oil mixing part, and slurry oil and product wax oil from the vacuum tower are mixed at the wax oil mixing part; the second branch is communicated with a second conveying pipeline, and the second conveying pipeline is communicated with an emergency connector. According to the invention, integrated hardware multiplexing and cooperative control are carried out on the load optimization process and the blockage emergency oil return function of the vacuum tower, and embedded active safety protection is constructed while load reduction and efficiency improvement of the vacuum tower are realized, so that the processing load and operation continuity of the device are systematically improved.

The present invention provides a digital intelligent power control system, which relates to the technical field of data processing. The method includes: collecting load data of equipment to obtain an initial load data set, preprocessing the initial load data set, and sorting it according to a time series, respectively extracting the load data therein by the time length of seconds and minutes, obtaining short-term load data and long-term load data, analyzing the short-term load data to obtain a transient load change rate and a high-frequency load fluctuation event, analyzing the long-term load data set to obtain a load fluctuation range and a periodic load fluctuation, and merging them to obtain a priority evaluation data set, allocating the load priority of the equipment according to the priority evaluation data set, obtaining a load priority table, and adjusting the allocation of power resources according to the priority evaluation data set to obtain a power allocation strategy, and allocating power resources to the equipment. The present invention can intelligently control electricity.