283results about How to "Scale up" patented technology

The invention relates to the field of mechanical detection, in particular to a fault diagnosis method for a wind driven generator based on a multi-modal large language model. According to the research, a GPT-3-based data set construction scheme is innovatively provided, and PDF-format mechanical fault related literatures are effectively converted into a structured high-quality question and answer pair data set. Then, fine adjustment is carried out on the base large language model by utilizing a data set, and the performance of the model in a professional task of mechanical fault diagnosis is successfully improved through fine adjustment parameter adjustment and exploration of a fine adjustment method. Besides, according to an actual use scene, a signal-to-text module is expanded, so that the signal-to-text module can receive multi-modal data of mechanical faults in real time, feature information contained in digital signals is changed into a text form and transmitted to a large language model module, and finally a diagnosis result is output by a large language model.

The invention relates to the technical field of lithium batteries, and discloses a lithium battery life decline quantitative prediction and health state intelligent evaluation method, which comprises the following steps: S1, multi-source data acquisition: arranging strain gauge sensors at battery electrode plates and a shell to acquire stress-strain data, integrating a gas sensor in a battery module to acquire gas data, and calculating the data of the stress-strain data; battery data and environment data are acquired by a high-precision synchronous acquisition system through nanosecond clock signals and are processed and stored, and the data quality is monitored at the same time; s2, data preprocessing: generating an extended data set through data transformation and combination and GAN-based data, and processing abnormal data by using a method based on statistics and machine learning; according to the invention, through multi-source data fusion acquisition, a stress-strain sensor and a plurality of gas sensors are added to monitor the internal mechanical change and gas component concentration of the battery, so that the problem that the accuracy and practicability of the model are greatly reduced due to a traditional method based on an electrochemical model is improved.

The invention discloses a small sample learning method and system based on data enhancement. Collecting electric power operation and maintenance text data from a plurality of data sources, preprocessing and labeling the data, and constructing an electric power field corpus; performing fine tuning on a predefined pre-training language model by using the power field corpus, training power field word vectors and entity vectors from the power field corpus, and fusing the power field word vectors and entity vectors into the pre-training language model; performing named entity recognition and multi-task learning of power operation and maintenance knowledge based on a pre-training language model, constructing a multi-task adversarial learning model in combination with an adversarial learning algorithm, training the multi-task adversarial learning model by adopting multiple data enhancement strategies, and performing data prediction through integrated learning and self-training by utilizing the multi-task adversarial learning model; and evaluating a model prediction result, and dynamically adjusting the weight of each data enhancement strategy based on an evaluation result. The small sample learning performance is improved.

The invention discloses a cabinet type server and a communication method, and relates to the field of computers. The cabinet type server comprises a plurality of fusion nodes and a plurality of switching nodes, a first fusion node in the plurality of fusion nodes comprises a plurality of computing chips and at least one first switching chip, the at least one first switching chip is used for realizing communication connection among the plurality of computing chips, and a second fusion node in the plurality of fusion nodes is used for realizing communication connection among the plurality of computing chips. The first switching node in the plurality of switching nodes is coupled with the first fusion node through the connector, and the first switching node is used for realizing communication connection between the first fusion node and other fusion nodes in the plurality of fusion nodes. A switching chip and a computing chip are deployed nearby, a plurality of fusion nodes and a switching node adopt a mode of orthogonal connection and data two-stage switching, and a high-density system with high performance, large bandwidth and low time delay is constructed.

A compressed air energy storage underground cavern model test system designed according to a preset joint disclosed by the present invention comprises a simulation cavern, a loading system, a measurement system and a protection device, the measurement system comprises a plurality of temperature sensors, a plurality of pressure sensors, a plurality of crack meters and a plurality of optical fiber gas sensors, and the simulation cavern comprises a steel cylinder sealed by a sealing door. A deformation simulation layer, a lining layer, a leveling layer, a geotechnical cloth layer and a flexible sealing layer are arranged on the inner wall of the steel cylinder, a plurality of longitudinal exhaust pipelines are embedded in the joint of the deformation simulation layer and the lining layer, a plurality of exhaust pipes are inserted into the cylinder wall of the steel cylinder, each exhaust pipe is communicated with one longitudinal exhaust pipeline, a plurality of longitudinal prefabricated cracks are distributed in the lining layer, and the longitudinal prefabricated cracks are communicated with the geotechnical cloth layer. A crack meter is embedded in the middle of each longitudinal prefabricated crack, and the simulation cavern is communicated with the loading system through an air inlet in the sealing door. The test system is low in cost, convenient to operate, large in scale, comprehensive in test item, high in test precision and high in reference value for actual engineering.

The invention provides a graphics processor interconnection structure and a data transmission method, and relates to the technical field of intelligent computing, the graphics processor interconnection structure comprises a plurality of graphics processors and at least one switching device group, each switching device group comprises a plurality of switching devices, the plurality of switching devices in each switching device group comprise switching devices of a first level and switching devices of a second level; wherein each graphics processor is connected with any switching device in each switching device group, and the switching device of the first level in each switching device group is connected with the switching device of the second level; the sum of the number of the switching devices connected with any switching device in each switching device group and the number of the graphics processors is smaller than or equal to the number of ports of the corresponding switching device. According to the structure, interconnection between large-scale graphics processors is realized.

The invention belongs to the field of software security and vulnerability detection, and particularly discloses a method and system for automatically generating and labeling a vulnerability program, and the method comprises the steps: determining a vulnerability repairing program of obtained vulnerability data, extracting a code mode containing a specific type from the vulnerability repairing program, and obtaining a screened vulnerability mode; the step of extracting the code mode of the specific type comprises the substeps of extracting a repair statement of a single modification block, generating a symbolized repair statement, removing repeated repair statements, identifying a key variable and describing a vulnerability triggering statement; according to the vulnerability mode, accurately injecting the vulnerability mode into a normal program in combination with a code analysis tool, and generating a vulnerability program containing a specific vulnerability type; and performing vulnerability type labeling on the vulnerability program in combination with static analysis and a large language model. According to the invention, the detection accuracy of the vulnerability detection model can be improved.

The invention provides a bus-based inertial measurement unit batch calibration method and system, and the method comprises the steps: placing inertial measurement units under different temperature conditions, executing a calibration test of preset attitude change and preset angular velocity change under each temperature condition, and obtaining test data; transmitting test data of all the inertial measurement units to a calibration system in batches through a bus; in the calibration system, obtaining test data of each inertial measurement unit according to the storage path, and calculating calibration parameters of each inertial measurement unit based on the test data; and downloading the generated calibration parameters to each inertial measurement unit in batches through a bus. According to the invention, the bus is adopted to calibrate the inertial measurement units in batches, so that the limitation of the traditional serial communication interface on the data transmission rate and the connection number is overcome, the calibration period is remarkably shortened and the calibration cost is reduced by improving the calibration efficiency and scale, and powerful support is provided for the large-scale production and application of the IMU.

The invention belongs to the technical field of medical data processing, and discloses a temporal-mandibular joint disease automatic diagnosis method and system based on artificial intelligence, and the method comprises the steps: collecting and preprocessing data; expanding the patient data, dynamically adjusting the structure and parameters of a data generator, and generating the patient data; performing feature extraction on the patient data, and automatically adjusting an evolution rule of a feature extraction network to obtain patient data features; performing dimensionality reduction on the data features to obtain low-dimensional patient data features; inputting the low-dimensional patient data features into a classifier, and dynamically adjusting the configuration and weight of hidden layer nodes of the classifier according to the low-dimensional patient data features to obtain a classification result; and defining a loss function, and training the model. Parameters can be automatically adjusted in the data feature extraction process, the data dimension reduction process is optimized by adopting quantum state mapping and an adaptive adjustment mechanism, the classifier can dynamically adjust the number of hidden layer nodes and an activation function, and the accuracy and robustness of a diagnosis result are improved.

The invention belongs to the technical field of computer-aided diagnosis, and particularly relates to a dynamic brain network feature learning method based on Mamba, which comprises the following steps of: performing data processing, including data collection, data preprocessing, dynamic function network sequence construction and state conversion sequence construction; performing sample enhancement by taking the processed data as sample data; constructing a Mama model comprising a linear embedding layer, a Mama layer, a linear mapping layer I, an activation function and a loss function as a classification model; training the Mama model by using the state conversion sequence, and optimizing hyper-parameters of the Mama model; and based on the trained Mama model, predicting the enhanced sample data, and determining a classification result of the sample data through a label output by the Mama model. According to the method, the relation hidden in the dynamic brain network state transition sequence can be efficiently captured, and high-precision classification of each brain network sequence is realized.

An integrated device for electrocatalytic reduction of carbon dioxide / carbon monoxide relates to the technical field of energy catalysis and comprises a reactor module, a gas-liquid passage module, an electrolyte passage module, a temperature control liquid passage module, a reaction circuit module and a control module. The reactor module comprises a cathode temperature control plate, a cathode plate, a membrane electrode, an anode plate and an anode temperature control plate; the gas-liquid passage module is used for providing raw material gas with adjustable temperature and humidity for a cathode reaction runner of the reactor module and outputting a reaction product; the electrolyte path module enables electrolyte to circulate in the anode reaction flow channel; the temperature control liquid passage module is used for controlling the temperature of the reactor module and the electrolyte storage tank; the reaction circuit module provides electric energy for the reactor module; and the control module is used for integrally controlling each module, monitoring the state of the device and recording operation data. The system has the advantages of being large in scale, capable of achieving integrated control, more convenient to use, more in controllable parameters, high in safety and the like.

The invention relates to the field of epistaxis identification, in particular to an epistaxis identification method based on multi-scale feature fusion and a computer device. According to the scheme, the method comprises the steps of extracting local features of an epistaxis image by adopting a full convolutional network, extracting global features by adopting Transform, capturing global context information, performing different scale transformation on a local feature map and a global feature map, generating feature maps of different scales, performing multi-scale feature fusion, performing context fusion, and calculating an attention weight. And finally, outputting final features, then carrying out cross entropy loss calculation and model training, and finally identifying epistaxis through the trained model. Aiming at the problems that a data set is unbalanced and a small bleeding area is easy to ignore, the attention weight is dynamically adjusted in the training process, and the method focuses on difficult negative samples and small sample areas, so that the accuracy is improved. And finally, the accuracy of epistaxis image recognition is improved. The method is suitable for epistaxis identification.

The method for predicting the milling position degree of the aero-engine blade based on the geometrical characteristics comprises the steps that key machining parameters are determined, and current data in the machining process are collected; importing the blade model into CAD software, parameterizing the curved surface of the blade, establishing a UV coordinate system, uniformly arranging detection points, detecting the location degree of each detection point, and calculating the curvature of each detection point along a V-direction streamline through the CAD software; calculating the residual height h of each detection point, and forming a data set by the current data corresponding to different key processing parameters, the curvature of the detection point along the V-direction streamline, the residual height of the detection point and the location degree of the corresponding detection point; constructing a neural network model; and training a neural network model by using the data set to predict the milling position degree of the aero-engine blade. According to the method, the complex nonlinear relation can be learned from the processing data, and the generalization ability and the prediction precision of the model can be improved through the geometric features.

The invention provides a railway internal data circulation sharing method based on semi-supervised federated learning, and relates to the technical field of semi-supervising, and the method comprises the steps: obtaining a plurality of initial local models from a plurality of clients; obtaining an initial global model; allocating a pseudo tag to each piece of no-tag data in the no-tag data set; obtaining a plurality of local models; distributing a corresponding weight parameter to each local model; and obtaining a data sharing model. According to the method, the model obtained through training based on the label data set is used for distributing the pseudo labels for the non-label data, so that the non-label data can participate in subsequent model training, the scale of the training data is further expanded, and the generalization ability of the model is improved; and distributing corresponding weight parameters to each local model based on the clustering distribution of the data on each client, thereby reflecting the importance of each local model in the data sharing model, so that the finally obtained data sharing model not only can consider the global generality, but also can reflect the personalities of different clients.

The invention relates to the technical field of airport special vehicle scheduling, in particular to an airport special vehicle scheduling method and system based on deep reinforcement learning, and the method comprises the following steps: taking the shortest flight service time as a target, modeling an airport special vehicle scheduling problem as a vehicle path planning problem with a time window, and solving a vehicle path planning problem with a time window; decomposing a vehicle path planning problem according to preset service priorities of various types of special vehicles; performing Markov decision process modeling on each sub-problem, building a neural network model based on an attention mechanism, generating a path solution of vehicle planning by using the neural network model, calculating a total return according to the path solution of each sub-problem, training the neural network model by using a deep reinforcement learning algorithm based on the reward of the total return, and performing vehicle planning. And obtaining a trained neural network model, and solving the airport special vehicle scheduling problem by using the trained neural network model. According to the method, scheduling solving of various types of special vehicles and various types of airport special vehicles can be realized, and the calculation efficiency is higher.

The invention discloses a single-face array low-altitude radar target recognition system based on deep learning, and the system comprises a data collection module which obtains the radar echo data of different targets in various environments, a receiver has a specific sampling frequency and a dynamic range, the collection time is set according to the characteristics of the targets, and the data is compressed and stored; the data preprocessing module is used for denoising, clutter suppression and normalization; the feature extraction module extracts features by using CNN; the target recognition module classifies and recognizes a target; the model training and optimizing module is used for training a model and optimizing parameters; and the visual display module displays the result. According to the invention, data can be accurately collected and efficiently preprocessed, target features are accurately extracted and recognized by using deep learning, the model is optimized to improve the accuracy, the result can be visually displayed, the accuracy, real-time performance and reliability of low-altitude radar target recognition are effectively improved, and the monitoring requirements are met.

The invention belongs to the technical field of education, and discloses a method for retrieving answers to questions from an educational knowledge graph, which comprises the following steps: acquiring a first educational question set # imgabs0 #, and analyzing cognitive representation and language style in the first educational question set; unifying language styles and cognitive representations in the first education question set through an instruction of a large language model; extracting subject subjects and combining questions to obtain complete and simplified questions; and performing multi-hop answer reasoning to deduce the answer of the recombined question. The method is improved in the aspects of multi-hop, educational cognition expression and different types of language styles, can be applied to different models of educational cognition expression problems, and has very high universality. According to the education cognition expression question answering method supported by the large language model, different cognition expression modes of learners in the real world are considered, and the effect is better in practical application compared with an existing model.

The invention relates to a deep-sea mining system which comprises a mining AUV (autonomous underwater vehicle), a seabed wireless charging base station, a nodule cage, a lifting device, a laying and recycling system and a production support mother ship. The multiple mining AUVs, the nodule cage and the lifting device are in communication connection and positioning with the mother ship through underwater sound, the seabed wireless charging base station is connected with the mother ship through a tape cable, and an electric power and information transmission line is arranged in the tape cable. The mining AUV is provided with bionic fins, mechanical legs and a picking manipulator, the mining AUV moves on the seabed in a floating mode or a mechanical leg crawling mode, and the picking manipulator is used for collecting mineral products on the seabed. The deep-sea mining system has good environmental benefits and economic benefits. The method mainly comprises the steps that contact with a deposition layer is small, and disturbance to sediments is small; the mining AUV can identify and avoid nodules containing life; and the cost can be reduced without stand pipes.

The invention discloses a two-ship meeting state intelligent identification method based on machine learning, and relates to the technical field of machine learning. According to the method, feature extraction and analysis are performed on AIS data, an intelligent recognition model and a training method based on machine learning are constructed, the trained model can intelligently recognize navigation behaviors of ships in different meeting states such as tracking, crossing and meeting, recognition precision and decision-making efficiency can be improved, follow-up ship collision avoidance is assisted, and the ship collision avoidance efficiency is improved. And the navigation safety is enhanced. According to the invention, a new technical means is provided for ship navigation safety and marine traffic management, and the development of intelligent shipping is promoted.

The invention discloses a large-scale tea lesser leafhopper cultivation and preservation method suitable for cicada tea production. The method comprises the following steps: step 1, preferably selecting Jinxuan, jade or Fuding white tea as a host plant variety, and determining the specification of the host plant; step 2, establishing a tea lesser leafhopper adult maintenance area, a nymph maintenance area and an egg incubation area, performing partition maintenance, and preferably selecting optimal maintenance conditions and culture density of different maintenance areas; step 3, preferably selecting a collection mode of tea lesser leafhopper eggs: preferably selecting an adult male-female ratio, and preferably selecting egg laying population density, egg collection duration and the like; 4, a tea lesser leafhopper egg storage chamber is established, and storage conditions are optimized; and step 5, preferably selecting incubation conditions and incubation time of the tea lesser leafhopper eggs after low-temperature storage. According to the large-scale production method, the tea branches to be trimmed in the tea garden are utilized, and the cost is low. The tea branches are used for culture, and the death rate caused by the fact that the leafhoppers need to adapt to host plants again during release is reduced.

The invention relates to an artificial intelligence image processing method. The method comprises the following steps: S1, data reading: reading historical image data and a to-be-processed image from a storage medium; s2, data screening: according to statistical analysis of historical image data, determining an image feature needing to be screened as a first image feature, obtaining an interval corresponding to the first image feature, if a first feature value of the preprocessed to-be-processed image data accords with the interval, carrying out screening, and otherwise, transmitting to the next step; s3, data enhancement and synthesis: performing data enhancement operation on the remaining images, and synthesizing new image data by using a generative adversarial network; and S4, brightness level division and label mapping: converting the color image into a gray level image, calculating an average gray level value of the gray level image, and dividing the image into low brightness, medium brightness and high brightness levels according to comparison between the gray level value of each pixel point and the average gray level value.

The invention provides an SAR image generation method based on style migration, and belongs to the technical field of synthetic aperture radar imaging and artificial intelligence. According to the method, firstly, an SAR image data set is used for training a convolutional neural network, so that the network can accurately extract the features of an SAR image, then, a style migration method is utilized, the neural network obtained through training is combined, an optical image is taken as a content image, the SAR image is taken as a style image, and the generation of a migration image is realized by optimizing the total migration loss. According to the method, the convolutional neural network and the style migration technology are utilized, the visual features are extracted from the optical image and converted into the features conforming to the SAR image style, and automatic generation from the optical image to the SAR image is achieved. The SAR image generated by the method can be used for data set expansion, data diversity is enhanced, and the method has a wide application prospect in the field of remote sensing image analysis and data processing.

The invention discloses a dynamic cost sensitive fault diagnosis system based on an AC-DCC model. The real-time performance and the accuracy of fault diagnosis are improved in an auxiliary mode. The system can automatically position fault equipment by analyzing multi-dimensional data such as time difference, equipment utilization rate, historical maintenance record, output deviation and alarm record, and can quickly adjust a production plan according to a diagnosis result, thereby reducing the negative influence of faults on production. In order to solve the problem of data imbalance in the industrial process, a dynamic cost sensitive classifier (AC-DCC) based on an Actor-Critic model is introduced into the system; according to the model, through strategy gradient optimization, sample weights are learned in a self-adaptive mode, and particularly, weight distribution of minority samples in the pseudo label generation and classification process is dynamically adjusted for the minority samples. According to the AC-DCC model, a new reward function is designed, attention to minority class fault samples is enhanced, the classification performance of minority class faults is improved, interference of the faults on production can be reduced through intelligent scheduling, and the efficiency and reliability of industrial production are optimized.

The invention relates to the technical field of phosphorus chemical new materials, in particular to a preparation method for synthesizing a supported Sn24P19.3I8 catalyst in one step. According to the method, the phosphorus source, the tin powder, the iodine and the carbon additive are used as raw materials, and the supported Sn24P19.3I8 catalyst is prepared through one-step solvothermal reaction, so that the preparation process of the catalyst is simplified, the preparation efficiency and scale of the supported Sn24P19.3I8 catalyst are remarkably improved, meanwhile, the energy consumption in the production process is reduced, the catalytic activity and stability of the catalyst are improved, and the method is suitable for industrial production. The method has important significance for promoting large-scale efficient preparation of black phosphorus, lays a catalyst foundation for developing more novel elemental phosphorus, and promotes market promotion and industrial development of phosphorus chemical new materials.

The present invention relates to the technical field of hospitals, clinics or similar buildings, and discloses a modular, easily expandable, large-scale intensive operating area. The problem that the module is difficult to design due to the complex connection relationship of the operating room is solved by a grouped multi-core method, ensuring that the basic module can complete the operation in a short time without relying on other parts with a smaller area, and using the configuration to achieve sharing of different modules by departments. The problem that the size of the operating area cannot be increased by increasing the number of modules due to the differences between the building structure and the operating rooms is solved by a modular hierarchical method, adding two new module levels, and the splicing rules of the building modules ensure that all basic modules can play a role while increasing the number of basic modules in different buildings. The splicing rules of the partition modules ensure that the differentiated needs of different operating rooms are met while increasing the number of basic modules. The combination of the above points realizes the modular design of the operating area and the expansion of the size of the operating area by increasing the number of modules.

The present invention relates to a power data enhancement analysis method based on a large language model empowering Text2SQL, comprising the following steps: S1: collecting natural language texts related to power and corresponding SQL query pairs; S2: preprocessing the data collected in step S1; S3: using a pre-trained large language model, fine-tuning the model based on a training data set, combining natural language understanding and information extraction to enhance the generalization ability of the model, and obtaining a large power language model; S4: expanding the training data set and the test data set; S5: evaluating the performance of the large power language model; S6: using adversarial training to enhance the robustness of the model, and using automatic parameter adjustment technology to optimize the model parameters, to obtain an optimized large power language model; S7: packaging the optimized large power language model into a container for deployment, setting a real-time monitoring mechanism after deployment, and tracking the conversion effect of natural language queries to SQL in the power system. The present invention can more effectively process natural language queries in the power system.

The invention relates to an asphalt mixture gradation identification method and system based on Vision Transform, and the method comprises the steps: shooting a slice map of an asphalt mixture test piece, and recording the gradation of the test piece; an OpenCV library is used, automatic cutting is carried out according to the aspect ratio of the slice map, and a data set is constructed in combination with grading of test pieces; preprocessing the data set sample, wherein the preprocessing comprises size adjustment, brightness matching, image enhancement and data amplification; and a Vision Transform classification network is used, training and testing are completed on the established data set, and asphalt mixture gradation identification is realized. Compared with the prior art, the method has the advantages that the accuracy rate is guaranteed, the cost of grading identification of the asphalt mixture can be remarkably reduced, and the tediousness caused by tests in the identification process is reduced.

The invention discloses a contract calling and delivery time simulation scheduling method and system, and relates to the technical field of information and control, and the method comprises the steps: obtaining the initial data of a production line, and carrying out the preprocessing of the initial data, and obtaining a to-be-scheduled resource amount; the preprocessing comprises data processing, beforehand statistics and parameter configuration; according to the to-be-ranked resource quantity and the base table data, multiple sets of galvanizing production line roller period schedules are obtained through a multi-process scheduling algorithm of rolling type roller period pre-ranking, and monthly scheduling results of all production lines are output and visually displayed in combination with a material ranking algorithm; and generating a material calling plan and a delivery time plan according to the monthly scheduling result of each production line. The material calling and delivery time plan finally formed by the method not only pays attention to several processes involved in monthly scheduling and resources in a library, but also takes a time base table as a core, and the scheduling of upstream and downstream resources in a month is comprehensively managed. Through the process design and implementation, the production efficiency and the on-time delivery rate can be effectively improved.

The invention relates to the technical field of cloud computing, in particular to a cloud tenant bearing scale optimization method and device, equipment, a medium and a product, and the method comprises the steps: recognizing a sender and an access purpose of an access request, and determining a data flow direction corresponding to the access request; in response to the condition that the sender is the virtual machine, address conversion and message encapsulation are performed on the access request through the virtual switch, and the encapsulated access request is sent to the cloud gateway; and in response to the fact that the sender is the Underlay network or the public network, the cloud gateway is determined based on the sender, the cloud gateway confirms the virtual private cloud based on the access purpose, the access request is subjected to message packaging, and the access request is sent to the virtual switch. According to the technical scheme of the invention, the original address translation is sunk to the virtual switches corresponding to the virtual private clouds from the cloud gateway for processing according to the data in the cloud-out direction and the cloud-in direction, so that the memory space occupied by the NAT table items can be released, the detailed routing capacity of the virtual machines is improved, and the expansion of the cloud tenant scale borne by the cloud data center is facilitated.

The invention discloses a PCIe device remote control method and device, an electronic device and a medium. The device comprises a local server and a remote server. A PCIe hardware equipment control component, a virtual PCIe bus and a plurality of local RDMA (Remote Direct Memory Access) network cards are deployed on the local server; pCIe hardware equipment, a PCIe hardware equipment proxy component and a plurality of remote RDMA (Remote Direct Memory Access) network cards are deployed on the remote server; the PCIe hardware equipment control component is used for realizing virtual PCIe equipment presentation of PCIe hardware equipment deployed on a remote server on a local CPU (Central Processing Unit); the virtual PCIe bus is used for managing the virtual PCIe bus and virtual PCIe equipment on the bus for the local CPU; and the PCIe hardware equipment proxy components are paired with PCIe hardware equipment deployed on a remote server one by one and are used for finishing proxy forwarding of PCIe requests between the remote RDMA network card and the PCIe hardware equipment.