388results about How to "Reduce transfer time" patented technology

The invention relates to the technical field of data storage management, in particular to a data storage method based on a large model. The method comprises the following steps: performing input class cluster fragmentation optimization on a target pre-training large model to obtain storage modulus fragmentation data; constructing an intelligent fragmentation storage strategy by using the storage modulus fragmentation data; obtaining target model training data; performing intelligent fragmentation storage on the target model training data by using an intelligent fragmentation storage strategy, and performing redundancy compression optimization to obtain dynamic redundancy storage index data; and training data storage space optimization is carried out according to the dynamic redundancy storage index data, and an optimized data storage strategy is obtained. According to the method, intelligent fragmentation storage and refined data redundancy compression are performed on the data, so that efficient and stable model training data access is realized, the problem of low data storage and access efficiency in a large model training process is solved, and the model training efficiency and reasoning real-time performance are remarkably improved.

The invention provides an electric control box remote data acquisition system based on the Internet of Things, and relates to the technical field of data processing, and the method comprises the steps: obtaining the operation state data of an electric control box through a sensor, transmitting the operation state data through a wireless signal network, obtaining a data frame set, analyzing the data identifier of each data frame, verifying the integrity of the data, and obtaining the data frame set; determining performance parameters of the wireless signal network, detecting whether an abnormal change occurs in the transmission process, generating a solution when the abnormal change occurs, generating a retransmission signal when packet loss occurs in the transmission process, determining a retransmission queue according to the data identifier, and transmitting the retransmission signal according to the retransmission queue when delay occurs in the transmission process. The priority of a retransmission queue and the distribution of network bandwidth are dynamically adjusted, and when signal attenuation occurs in the transmission process, the execution sequence is adjusted according to the operation record and result of a solution by switching a transmission channel and adjusting transmission power. The transmission process of the operation data of the electric control box can be detected.

The invention relates to the technical field of music sharing, and particularly discloses a music sharing method and system based on an intelligent Bluetooth headset and the Bluetooth headset. Comparing the working state signal interference evaluation parameters of each Bluetooth earphone with a working state signal interference evaluation parameter threshold value of the Bluetooth earphone, and if the working state signal interference evaluation parameter of a certain Bluetooth earphone is greater than the working state signal interference evaluation parameter threshold value of the Bluetooth earphone, switching a working channel of the Bluetooth earphone; according to the method, the quality of each channel is evaluated, the to-be-adjusted Bluetooth headset is subjected to channel switching according to each channel switching reference index of the to-be-adjusted Bluetooth headset, and the working channels of the Bluetooth headset are dynamically switched, so that the Bluetooth headset can be helped to maintain stable connection, disconnection or unstable signal conditions are reduced, and the service life of the Bluetooth headset is prolonged. And the data transmission rate can be improved, and congestion is reduced.

The invention discloses a shield tunnel risk early warning method and system based on digital twinning and federated learning, and the method comprises the steps: a shield tunnel construction side client serves as an edge calculation terminal, and obtains basic parameters; a first shield digital twinning system, a second shield digital twinning system and a third shield digital twinning system are correspondingly constructed by the shield machine producer client side, the shield tunnel construction side client side and the third-party client side according to the basic parameters; a shield construction risk early warning model is arranged in each shield digital twin system; the shield machine production side client, the shield tunnel construction side client and the third side client are all participants, and each participant sends model parameters of the respective shield construction risk early warning model to the central server; the central server is used for averaging the parameters of the three models and issuing the average value of the parameters to each participant; and each participant updates the model of the participant according to the average value of the parameters, and realizes shield construction risk early warning based on the updated model.

The invention discloses a high-speed internet access system and method suitable for a low-orbit satellite, and relates to the field of satellite internet communication. The system comprises a low-orbit satellite group, a ground receiving station, user terminal equipment, a central control center and the like. The low earth orbit satellite group forms a dense satellite network on a near earth orbit so as to realize seamless network coverage at any corner of the earth. The ground receiving station supports simultaneous establishment of connection with a plurality of satellites, and ensures stability and high speed of data transmission. The user terminal equipment enables a user to directly communicate with the nearest low-orbit satellite, and achieves the quick access to the Internet. The central control center adopts a dynamic resource allocation algorithm to dynamically adjust spectrum resources and channel allocation according to real-time communication requirements and the state of the satellite network, and network performance is optimized. The invention not only provides an unprecedented network access solution for special environments such as remote areas and offshore platforms, but also has great significance for improving the popularity rate and quality of global internet access.

The invention relates to the technical field of fault diagnosis, and discloses a vehicle fault diagnosis system and method, a vehicle terminal and a storage medium, the system comprises the vehicle terminal and an edge calculation center, the vehicle terminal is used for collecting diagnosis fault codes and vehicle operation data of the vehicle terminal, and if a first fault code is found in the diagnosis fault codes, the vehicle operation data is sent to the edge calculation center; if yes, determining a first fault result according to the first fault code; if a second fault code is found in the diagnosis fault code, an edge node is determined according to the position information of the vehicle terminal, the edge node is connected, the second fault code and second operation data corresponding to the second fault code are sent to the edge node, the edge node is at least one of edge calculation centers, and the edge calculation centers are connected to the edge node; the edge node is used for performing fault diagnosis according to a third fault code and third operation data corresponding to the third fault code if the third fault code is found in the received second fault code to obtain a second diagnosis result; through the method, a more practical and more accurate vehicle fault diagnosis system is provided.

The invention relates to the technical field of computers, and provides a high-speed data transmission method for micro-computing equipment, which comprises the following steps of: acquiring to-be-transmitted data of the micro-computing equipment, identifying the characteristics of the to-be-transmitted data, analyzing a transmission mode, determining a data priority, analyzing hardware configuration information based on the data priority, and detecting the performance of a data transmission interface. The method comprises the following steps: calculating an interface transmission rate and evaluating transmission quality to formulate a data transmission strategy, partitioning data, analyzing a transmission sequence, constructing a transmission time window, monitoring a window state to construct a high-speed transmission path, identifying a transmission node, analyzing a transmission index of the transmission node, and finally generating a data transmission scheme based on the transmission index. According to the invention, the data transmission efficiency of the micro-computing device can be improved.

The invention discloses an electric propulsion GEO satellite orbit transfer and point fixing method, and belongs to the technical field of satellite orbit dynamics, and the method comprises the following steps: S1, building a low-thrust satellite orbit dynamics model based on multi-perturbation and earth shadow constraints; s2, according to the low-thrust satellite orbit dynamic model, a Lyyunov function is constructed, and an RQ-Law feedback control model is constructed by using the Lyyunov function; s3, weight parameters of the RQ-Law feedback control model are optimized; and S4, using the optimized RQ-Law feedback control model to complete orbit transfer and point fixing. Aiming at the problem of long time of current satellite orbit transfer, the invention provides an innovative method, the traditional orbit transfer and fixed point staged optimization are integrated into one stage, and the integration of orbit transfer and fixed point is realized.

The invention discloses a holographic image information processing system, which comprises an edge calculation and data preprocessing module, a data compression and coding module, a holographic image information processing module, a holographic image information processing module, a holographic image information processing module, a holographic image information processing module, a holographic image information processing module and a holographic image information processing module, wherein the edge calculation and data preprocessing module is responsible for acquiring original holographic image data from holographic image acquisition equipment or a storage system; according to the holographic image compression method, lossless compression or lossy compression is decided to be adopted by judging image quality requirements, image data needing to keep high quality of a holographic image are compressed by adopting a Huffman coding lossless compression algorithm, and image data receiving quality loss are compressed by adopting a JPEG lossy compression algorithm. The system can intelligently select a lossless or lossy compression algorithm according to the image quality requirement, effectively reduces the data volume of the holographic image, and guarantees the image quality. The parallel transmission and streaming media processing module can transmit a plurality of data blocks at the same time by using a multi-path parallel transmission technology, so that the overall transmission time is remarkably shortened.

The invention discloses a firmware upgrading method and device, electronic equipment and a readable storage medium. The method comprises the steps that a differential upgrading package sent by a cloud end is received; the differential upgrade package is obtained by performing differential operation on a first firmware upgrade package and a second firmware upgrade package with different versions through a target differential algorithm; performing reduction operation based on the differential upgrade package and the existing firmware upgrade package through a target reduction algorithm corresponding to the target differential algorithm to obtain a target firmware upgrade package; the existing firmware upgrade package is obtained after the electronic equipment stores the first firmware upgrade package; and performing firmware upgrading through the target firmware upgrading package. In the application, the cloud only needs to send the differential upgrade package and does not need to send the complete second firmware upgrade package, so that the data volume sent by the cloud is reduced, the data transmission time between the cloud and the electronic equipment is greatly reduced, the upgrade package transmission time is further reduced, and the firmware upgrade efficiency is improved.

The invention discloses a LoRa multi-hop network adaptive transmission system and method for link quality, and the system comprises a layering module which determines the hop count of each node in a network from a gateway; the routing exploration module is used for determining a father node of each node and selecting according to the signal intensity in a packet in the layering module; according to the parameter optimization module, the gateway calculates SF and BW which can be optimized according to the link quality from the node to the gateway, and if the link quality is good, the node is switched into optimized parameters; and the node management module is used for periodically sending layered messages by each node and periodically updating a neighbor table based on signal intensity so as to manage deletion and addition of the nodes. According to the method, the transmission time is further shortened while the node is ensured to meet the delivery rate through the method for adaptively adjusting the communication parameters based on the link quality, and the reliability of data transmission is ensured.

According to the model training-oriented gradient data transmission method and the related equipment provided by the invention, the compression ratio is dynamically determined according to the change of the transmission rate by monitoring the transmission rate in real time, so that the gradient data can be intelligently compressed according to the compression ratio, the data transmission quantity is remarkably reduced, and the network burden is relieved; and meanwhile, the transmission rate of each working end is adjusted through congestion control according to the bandwidth of a transmission link between the switch and the parameter server so as to relieve congestion. Under the combined action of the two, the transmission time of the gradient data is remarkably shortened, and the data transmission efficiency is improved, so that the training process of the distributed artificial intelligence model is accelerated, and the overall training efficiency is improved.

The invention discloses a power distribution network automatic communication data packet forwarding method based on an improved EPON technology, and the method comprises the steps: 1, classifying and scheduling data packets, and determining the transmission sequence of data packets with different priorities through weighted polling scheduling; step 2, an optical line terminal (OLT) receives the data packet and records sending address information and forwarding rules of the data packet; step 3, the ONU extracts a target address and a forwarding rule, and obtains a position and a connection state to calculate an optimal data transmission path; and step 4, monitoring and managing the operation state, and ensuring the forwarding stability and reliability of the data packet. According to the method for realizing artificial intelligence by improving the EPON technology, automatic communication data packet forwarding is realized, and the transmission efficiency is improved.

The invention provides a low-delay communication method, system and device for a distributed safety production monitoring platform and a medium, and belongs to the technical field of data communication, and the method comprises the following steps: identifying a network topology structure of distributed nodes in the safety production monitoring platform and transmission requirements among the distributed nodes, and determining a starting end and a destination end; determining a routing path of each transmission demand according to the network topology structure, and determining a sending end and a receiving end of each sub-path according to the starting end, the destination end and the routing path; fragmenting the data at the starting end of each transmission demand, and transmitting the data packets of each fragment to the destination end in a parallel mode; and verifying the data packet of each fragment at the destination end of each transmission demand, combining the data packets of each fragment to obtain original data after the verification is passed, and sending a retransmission request to the starting end after the verification is not passed. According to the invention, the communication delay is reduced, and the fault-tolerant capability and reliability of data communication are improved.

The invention relates to the technical field of non-standard automation, in particular to full-automatic lock catch feeding, injection molding and transferring equipment which comprises a feeding mechanism, a plate placing mechanism, a material taking mechanism, an injection molding mechanism and a discharging and transferring mechanism. The feeding mechanism comprises a fixed shaft feeding assembly and a shaft sleeve feeding assembly, the fixed shaft feeding assembly is used for locking fixed shaft feeding, and the shaft sleeve feeding assembly is used for locking shaft sleeve feeding; the wobble plate mechanism comprises a wobble plate mechanical arm, a wobble plate driving assembly and a wobble plate material frame, the wobble plate mechanical arm is used for grabbing the locking fixed shaft and the locking shaft sleeve on the fixed shaft feeding assembly and the shaft sleeve feeding assembly respectively and placing the locking fixed shaft and the locking shaft sleeve on the wobble plate material frame for positioning and placing, and the wobble plate driving assembly is used for driving the wobble plate material frame to rotate. The placed locking fixing shaft and the locking shaft sleeve face the injection molding mechanism; full-automatic lock catch feeding, injection molding, taking and transferring are achieved, manpower and material resources are saved, and the production efficiency is improved.

The invention relates to the technical field of AI big data processing, and discloses an AI big data real-time processing and analysis method, which comprises the following steps: step 1, data acquisition and preprocessing; step 2, feature extraction and intelligent adaptation; step 3, feature association and data screening; step 4, data analysis; 5, system architecture and technical support; and step 6, edge calculation support. For different types of data, such as images, texts, sensor data and the like, the edge device firstly performs standardization processing, converts the data into a general format, facilitates subsequent unified processing, converts the images into a specific resolution or color format, and converts the texts into a unified coding format; and the sensor data is converted into a specific data unit or range, so that the beneficial effect of uniformly processing different data types to improve the data processing efficiency is achieved.

The embodiment of the invention provides a log backup method and device, a database system and computing device.The log backup method is applied to the database system and comprises the steps that a logic log, generated through triggering, of a first data operation instruction in a target transaction is obtained, and the target transaction comprises multiple data operation instructions; the first data operation instruction is any one of a plurality of data operation instructions, and the logic log is generated by executing the first data operation instruction to complete modification operation on main library data; copying the logic log in real time and sending the logic log to the standby library, and returning to execute the step of obtaining the logic log of the first data operation instruction generated by triggering in the target transaction; and under the condition that the standby library receives the logic logs of the data operation instructions of the target transaction, submitting the logic logs of the data operation instructions. Through real-time transmission of the logic log, the risk of degradation to asynchronous replication caused by timeout is reduced, and the stability, availability and response speed of a database system are enhanced.

The invention belongs to the field of semiconductor processing and manufacturing, and provides a temporary bonding and de-bonding method, a composite substrate preparation method and a wafer manufacturing method.The temporary bonding and de-bonding method comprises the steps that silicon oxide layers are formed on two silicon substrates; the silicon oxide layer is subjected to ion implantation, hydrofluoric acid treatment and plasma activation, temporary bonding of the two silicon substrates is realized through hydrophilic bonding, and a temporary bonding interface is cracked through de-bonding annealing, so that de-bonding is realized. According to the method, a bonding medium does not need to be used, the thermal properties of the two substrates can be kept consistent, and warping and breaking are effectively avoided; temporary bonding is high in strength and good in stability, the functional substrate can be greatly thinned through the method, and therefore the composite substrate is prepared or the method is further used for manufacturing a multi-layer wafer bonding structure.

The invention provides a reasoning calculation method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a target text to be reasoned through a large language model; the target text is preprocessed, and all first tokens in the target text are obtained; through a target data loading strategy obtained by limiting parameters of each second token in N second tokens which need to be loaded and are processed recently and uniformly storing row and column data of a feedforward neural network layer of the large language model, target parameters used for reasoning are loaded from a persistent storage medium storing the large language model; wherein the second token comprises the first token and / or a token obtained by reasoning based on the first token; and performing reasoning calculation on the integer ID through the target parameter to complete reasoning of the new second token. According to the invention, the problems of high data transmission cost and low reasoning speed in the prior art are solved.

The invention discloses a relay protection device testing method and system, electronic equipment and a storage medium. The method comprises the following steps: acquiring operation data of the relay protection device under different test conditions through test equipment, and packaging the operation data according to a preset packaging format; sending the packaged operation data to the test host in the form of a data packet through the target communication channel by the test equipment; wherein the target communication channel is constructed based on a peripheral component interconnection rapid bus PCIE interface of the test host and a PCIE communication module of the test equipment; verifying the received data packet through the test host, and storing the verified data packet in a data buffer area of the test host; and analyzing the packaged operation data in the data packet in the data buffer area through the test host, and determining whether the relay protection device is in a normal operation state or not according to the operation data obtained after analysis. According to the scheme, the data transmission time can be shortened, and the test efficiency is improved.

The utility model provides a stacked multi-wafer storage and transmission mechanism, which comprises a driving device, a mechanical arm, a transmission chamber, process cavities, a LoadLock module and wafers, and is characterized in that the driving device is in control connection with the mechanical arm, the mechanical arm is arranged in the transmission chamber, the transmission chamber is connected with at least one process cavity through a transmission channel, and the transmission chamber is connected with the LoadLock module; and the mechanical arm controls the wafer to be conveyed between the process cavity and the LoadLock module. According to the invention, the LoadLock modules which are subjected to stacking design are re-arranged, and the original LoadLock modules which are arranged in a single layer are changed into the double-layer stacked LoadLock modules, so that the waiting time of the mechanical arm during wafer conveying work can be effectively saved, the overall wafer conveying time is shortened, and the wafer production efficiency is improved.

The invention relates to the field of gas pipe network leakage early warning, in particular to a gas pipe network leakage early warning method based on edge calculation, which is applied to an equipment terminal, and comprises the following steps: responding to a leakage monitoring request, and receiving pipe network current data acquired by an acquisition end, the pipe network current data comprising pressure current data and leakage current data; calling a preset screening model to process the pipe network current data to obtain abnormal deviation values, and obtaining effective pipe network data based on the abnormal deviation values and the corresponding data fluctuation thresholds; receiving pipe network annotation data, obtaining hidden layer output values of the pipe network annotation data according to the initial evaluation model, and determining a state output weight based on each hidden layer output value to obtain a pipe network state evaluation model; and inputting the effective pipe network data into the pipe network state evaluation model for evaluation, generating a state evaluation result, processing an abnormal state evaluation result based on a preset early warning strategy, generating leakage evaluation data, and sending the leakage evaluation data to a data center.

The invention discloses a multi-channel signal transmission method and system based on big data, and aims to improve the signal transmission efficiency and stability, the method comprises the following steps: capturing and preprocessing each channel signal data through a big data platform, and calculating each channel signal purity index; inputting the signal purity index of each channel into a machine learning model, predicting a future state value of each channel, and calculating an expected bandwidth utilization rate of each channel according to the future state value of each channel; obtaining the expected data transmission rate of each channel according to the expected bandwidth utilization rate of each channel; calculating the transmission performance of each channel according to the expected data transmission rate of each channel and the stability of each channel; acquiring signal data to be transmitted, determining transmission requirements, selecting a plurality of channels with transmission performance meeting the transmission requirements, and transmitting the signal data to be transmitted through the selected channels; according to the invention, key technical problems can be effectively solved, and signal transmission requirements in different scenes can be met.

The invention discloses a dual-mode communication unit frequency modulation method based on an adaptive algorithm, relates to the technical field of communication, and aims to solve the problems that the traditional communication method is prone to signal instability in a special working environment, so that the communication is interrupted repeatedly, the working continuity and safety are influenced, the transmission efficiency is low, and the communication efficiency is low. The method comprises the following steps: acquiring and analyzing environment parameters, and acquiring special working environment parameters; the temperature (T), humidity (H) and electromagnetic interference intensity (E) parameters in a special working environment are acquired through various high-precision sensors, and the sensors are calibrated, so that the communication mode can be automatically adjusted according to the change of the special working environment by accurately acquiring the environmental parameters, comprehensively evaluating the communication quality and flexibly adjusting the communication frequency; the problems of unstable signal and low transmission efficiency of traditional communication in a special environment are solved.

The invention relates to the technical field of transfer paper coatings, in particular to a thermal dye sublimation transfer paper coating and a preparation method thereof. The preparation method comprises the following steps: coating substrate paper with a photo-thermal dual-curing polyurethane emulsion, and curing to form a hydrophobic layer; the photo-thermal dual-curing polyurethane emulsion is formed by mixing epoxy group end-capped silicon-containing polyurethane and double-bond end-capped polyurethane, wherein the epoxy group end-capped silicon-containing polyurethane is obtained by modifying hydroxyl-terminated polydimethylsiloxane and capping propylene glycol diglycidyl ester, and the double-bond end-capped polyurethane is obtained by carrying out chain extension on glycerol monoglycidyl ether and capping hydroxyethyl acrylate; coating the hydrophobic layer with an ink-absorbing layer coating containing sulfhydrylated silicon dioxide, an acrylamide monomer and sodium carboxymethyl cellulose, and performing crosslinking to form an ink-absorbing layer; and finally, coating a polyvinyl alcohol solution on the ink absorption layer to form a surface layer. The thermal dye sublimation transfer paper designed by the invention is high in coating adhesive force and good in transfer effect.

The invention discloses a CT detector chip, a pixel arrangement method of the chip and a CT scanning method. The CT detector chip comprises a substrate, the substrate is provided with M rows of pixel areas, and each row of pixel areas comprises N pixel positions; two rows of detector modules are respectively arranged in the first row of pixel areas and the Mth row of pixel areas, and the other rows of detector modules are uniformly arranged in the second row to the (M-1) th row of pixel areas at intervals of one row of pixel areas; the circuit transmission grid electrode is arranged on the substrate, is electrically connected with the plurality of rows of detector modules and is used for controlling the on-off of a circuit of each row of detector modules; and the data reader is arranged on the substrate, is electrically connected with the plurality of rows of detector modules and is used for reading the electric signals. According to the method, the detector modules of the CT detector chip are sparsely arranged, so that the projection data volume is reduced, and the transmission time of the projection data is shortened; moreover, a more complex circuit can be designed at the blank pixel position of the chip, so that the performance of the chip is improved.

The present invention discloses a battery charging rack, a charging compartment, and a battery swap station or energy storage station having the charging compartment. The battery charging rack includes a plurality of vertically arranged battery compartments, each battery compartment having a battery inlet and outlet, the battery inlet and outlet are used for a battery transfer device to store a battery pack on the battery charging rack or to take the battery pack out of the battery charging rack, the battery compartment includes a first times speed chain displacement mechanism, the first times speed chain displacement mechanism is extended in the battery compartment along a first direction of the battery pack inlet and outlet direction, and is used to drive the battery pack to move along the first direction. On the one hand, the position of the battery pack in the first direction can be accurately adjusted, and since the electrical connector is usually arranged on the side of the battery compartment, it is beneficial for the battery pack to be aligned with the electrical connector in the first direction; on the other hand, it can drive the battery pack from the battery transfer device to the battery compartment, or drive the battery pack from the battery compartment to the battery transfer device, thereby improving the efficiency of the battery swap electrical connector.

The invention discloses antibacterial gynecological gel and a preparation method thereof, and relates to the field of gynecological products, arginine powder and sorbitol are dissolved in water, a hydrothermal reaction is performed at 130-160 DEG C, cooling is performed to obtain an arginine carbon quantum dot solution, then modified hydrogel and a plant extract are added, vacuum stirring and uniform mixing are performed, ultraviolet irradiation is started while stirring, standing is performed, and the antibacterial gynecological gel is obtained. The bacteriostatic gynecological gel is obtained. The adopted gel matrix releases arginine carbon quantum dots in a weak acid environment, and is compounded with active substances, especially flavonoid substances, in the plant extracting solution, so that a better synergistic antibacterial effect is achieved. Sensitive response of the carbon quantum dots to illumination is combined, so that the carbon quantum dots can be attached to a gel matrix more easily; meanwhile, polar groups, especially oxygen-containing groups, are introduced to the surface of the gel, so that the viscosity and the adhesiveness of the gel material are greatly improved, and the affected part can be better treated.

The utility model relates to the technical field of biscuit production, and discloses an automatic biscuit dough feeding device which comprises a base, two frame bodies are fixedly connected to the bottom side of the base, a plurality of rotating rollers are rotationally connected to the close sides of the frame bodies, a conveying belt is arranged on the outer walls of the rotating rollers in a sleeved mode, a stirring barrel is fixedly connected to the top side of the base, and a conveying belt is arranged on the conveying belt. Two augers are rotationally connected to the inner wall of the stirring barrel, a first rotating shaft is rotationally connected to the rear side of the stirring barrel, and the augers are connected with the first rotating shaft through a first bevel gear set. According to the dough cutting and conveying device, an integrated structural design is adopted, dough cutting and conveying are integrated, intermediate steps and transfer time in the production process are reduced, so that production efficiency is improved, meanwhile, a motor which is singly used as a driving source is adopted for driving, high cost caused by a large amount of power consumption brought by a plurality of driving sources is reduced, and the dough cutting and conveying device is suitable for large-scale popularization and application. Therefore, the utilization efficiency of the motor is improved.

The invention discloses a Micro-LED display-oriented driving method and chip, and belongs to the technical field of LED display, and the method comprises the steps: obtaining the total number of pixel points of a to-be-displayed image; according to the total number of the pixel points, the image to be displayed is divided into a plurality of sub-areas, and each sub-area comprises a pixel array composed of a plurality of driving units; transmitting the gray data of each pixel point to the driving unit of the corresponding sub-region through a preset protocol; the driving unit receives a row signal, gray data, reference current and a clock signal as input signals; when a row signal is received, the driving unit converts gray data into a PWM signal, and the duty ratio of the PWM signal is determined by the gray data; and controlling the output time of the reference current according to the PWM signal so as to generate a driving current corresponding to the gray data to drive the pixel point to emit light. Aiming at the problem that a Micro-LED display driving chip in the prior art is high in power consumption, the display uniformity is improved while the power consumption is reduced.