200results about How to "Overcome deficiencies" patented technology

The invention belongs to the technical field of polymer plastic materials, and particularly relates to a high-temperature-resistant anti-corrosion polytetrafluoroethylene material and a preparation method thereof. The polytetrafluoroethylene is prepared from the following raw materials in parts by weight: 50 to 60 parts of polytetrafluoroethylene, 5 to 10 parts of fluorinated modified graphene oxide, 3 to 5 parts of aluminum powder, 5 to 8 parts of modified silicon carbide fiber, 3 to 5 parts of silane modified nano silicon dioxide, 2 to 3 parts of polyphenylene sulfide, 2 to 3 parts of polyimide, 0.5 to 0.8 part of dispersing agent and 0.5 to 1 part of antioxidant. The heat conduction performance and the interface bonding force are improved through the fluorinated modified graphene oxide, the impact resistance and the wear resistance are enhanced through the silicon carbide fibers, and the corrosion resistance and the wear resistance are improved through the aluminum powder; silane modified silicon dioxide enhances the mechanical property of the composite material by improving the dispersity of the filler, and PPS and PI interact with the surface of the filler through polar groups in molecular chains, so that the interface strength is improved, and the toughness and the thermal stability are improved.

The invention belongs to the technical field of building materials, and particularly relates to a preparation method of a high-performance rubber concrete material, which comprises the following steps: (1) carrying out ultrasonic treatment on rubber powder, rubber particles, water and tetraethoxysilane, adding a silane coupling agent, continuously carrying out ultrasonic treatment for 30 minutes, drying and sieving to obtain modified rubber powder and modified rubber particles; (2) uniformly mixing calcium fluosilicate, aluminum oxide and magnesium oxide, calcining at high temperature, adding blast furnace slag and silicon powder, and grinding and activating under pressure to obtain a reinforcing agent; and (3) uniformly mixing cement, fly ash and gypsum, adding water, stirring to obtain slurry, adding the modified rubber powder, the modified rubber particles and the reinforcing agent, then adding the functional additive, the short glass fiber and the medium sand, and uniformly mixing to obtain the rubber concrete. The compressive strength of the prepared rubber concrete is 42 MPa or above, the tensile strength ranges from 3.0 MPa to 4.0 MPa, the breaking strength ranges from 5.0 MPa to 6.0 MPa, and application requirements are met.

The invention belongs to the technical field of building energy conservation and carbon reduction, discloses a building energy conservation and carbon reduction method, device and equipment and a storage medium, and constructs a high-precision building BIM model. A digital twinborn model of a Xian mansion is constructed by applying a BIM technology, and the model not only depicts the physical structure of the building in detail, but also simulates key operation details such as personnel flow and equipment operation. By strategically simplifying the model, it is ensured that the model is efficient and accurate. The comparative analysis of the corrected model and actually measured data shows that the prediction error of the model is only 0.10%, the extremely high consistency of the prediction result and the actual situation is highlighted, the high accuracy of model correction is verified, and the high standard requirement of in-depth research is met; according to the invention, balance optimization of carbon emission minimization and comfort maximization is realized. By intelligently adjusting the operation strategy of the air conditioning equipment, the method obviously reduces the carbon emission of the building while ensuring the living comfort.

The invention discloses a DC bus interference anomaly detection method based on a dynamic local outlier factor algorithm. The DC bus interference anomaly detection method is especially suitable for real-time operation state monitoring of a complex power system. Aiming at the limitation that a traditional anomaly detection method is high in calculation complexity and limited in detection precision in high-dimensional time series data processing, optimization is carried out on the basis of an outlier detection theory, a dynamic mode updating algorithm is provided, the algorithm evaluates outlier characteristics of time series data through local density, and the detection precision of the time series data is improved. The global calculation overhead is obviously reduced, and the real-time performance and the detection efficiency of the system are improved. Compared with a traditional method, the method consumes manual intervention, is high in detection precision and strong in early warning, can be widely applied to the field of health assessment and intelligent operation maintenance of power equipment, and shows a good engineering application prospect.

The invention provides a hybrid power device and an aircraft, and relates to the field of aircraft power systems. The hybrid power device comprises a power supply system and a control system. The power supply system comprises a solid oxide fuel cell module, a storage battery module and a super capacitor module. The solid oxide fuel cell module comprises an integrated plate type metal support solid oxide fuel cell or electric pile and has the advantages that the fuel gas sealing performance is good, the number and the weight of structural parts such as metal connecting bodies, end plates and sealing materials are reduced, the light weight can be realized, and the power-to-weight ratio is favorably improved. The control system comprises an operation control module and an energy management module; the energy management module is connected with the power input end of the aircraft to obtain the power requirement of the aircraft and calculates the output power of the solid oxide fuel cell module, the storage battery module and the super-capacitor module according to the requirement, the weight of a power system is reduced, the flight endurance time is prolonged, and the flight safety is improved. And the output performance and the service life of the hybrid power system are greatly improved.

The invention relates to the technical field of radar weak target detection, in particular to a radar weak target detection method and device based on deep learning. According to the technical scheme, the method comprises the following steps: data preprocessing and enhancement: receiving an original echo signal of a radar, generating three-dimensional time frequency-wavelet coefficient spatial data by combining time frequency analysis with wavelet transform, and expanding a data set in multiple modes; and feature extraction network construction: constructing a 3D-CNN and LSTM fused network, wherein the 3D-CNN has a convolution operation with a variable convolution kernel size. According to the method, the data are richer and more comprehensive through time-frequency analysis in combination with wavelet transform and data enhancement, and multi-dimensional features can be accurately captured through 3D-CNN and LSTM fusion and an adaptive convolution kernel; and the training effect is ensured through the adaptive learning rate and the early stop method, and the accuracy and adaptability of weak target detection are jointly improved.

The invention is applicable to the technical field of knowledge migration, and provides a graph domain adaptive node classification method based on instance learning, which comprises the following steps of: acquiring a source domain data set with a mark and a target domain data set without a mark; the representation of the source domain data set and the representation of the target domain data set are learned through a GNN model, modeling is conducted on the source domain data set and the target domain data set, and embedded vectors of the source domain data set and the target domain data set are obtained; the embedded vectors of the target domain data set are classified through KNN, and target domain category pseudo labels are obtained; performing fine-grained instance contrast learning on embedded vectors of the source domain graph data set and the target domain graph data set; and after classification is finished, displaying a final classification result of the target domain graph data set. According to the method, the defects of an existing graph domain adaptation technology in cross-domain tasks are overcome, and the method has wide applicability and can be applied to graph data analysis tasks in various cross-domain scenes, such as social network analysis, recommendation systems, knowledge graphs and biological networks.

The invention provides a farmland irrigation water body suspended matter capturing device and a using method, the farmland irrigation water body suspended matter capturing device comprises a hydroelectric generator and a capturing device which are arranged in a ditch in the water flow direction, the bottom of the capturing device is provided with a low-position water passing channel, the end of the low-position water passing channel is provided with a valve plate moving along with rising of the water level, and the valve plate is used for blocking the low-position water passing channel; a plurality of rotating capturing frames are arranged at the top of the capturing device, and the rotating unfolding direction of the capturing frames is the same as the water flow direction; a plurality of electrode plates are arranged on the two sides of the capturing device, and each electrode plate comprises a positive electrode and a negative electrode which are arranged on the two sides of the capturing device respectively and used for adsorbing suspended solids in water. Through comprehensive utilization of hydroelectric generation, electrode plate electrostatic adsorption, mechanical filtration and automatic adjustment technologies, the problem that existing irrigation water is high in sand content is effectively solved, cleanliness of the irrigation water is ensured, and the efficiency and quality of farmland irrigation are improved. The technical scheme is simple and convenient to operate, low in maintenance cost, environment-friendly, energy-saving and wide in application prospect.

The invention relates to an unsupervised speech recognition modeling method based on phoneme segment-level characterization discretization, and belongs to the field of speech recognition. According to the method, phoneme features of original audio are extracted by using an IFMF model through a voice feature discretization module, then a K-Means clustering algorithm is trained to cluster audio representation, and a clustering index is acquired as a discrete mark; training a generator network and a discriminator through an adversarial learning module; and decoding the phoneme discretization-based model output obtained by unsupervised training by using a language model through a voice discrete representation decoder module. According to the method, multiple speech features are fused, and phoneme segment-level characterization discretization is adopted, so that the influence of tone information on Lao is considered, and the influence of information redundancy caused by fine-grained features on cross-modal modeling is reduced as much as possible. Compared with a traditional speech recognition method, the method provided by the invention obtains a competitive result.

The invention relates to the technical field of image processing, in particular to a marine metal corrosion degree detection method based on image processing, and the method comprises the steps: obtaining a plurality of grey-scale maps of a marine metal sample, taking the probability that a pixel point belongs to a background pixel point as a first probability, dividing a target pixel point and the background pixel point based on the first probability, taking the probability that the target pixel point belongs to the analysis pixel point as a second probability, determining an analysis area according to the second probability, and calculating the probability that the edge pixel point belongs to an inner edge pixel point as a third probability according to the fluctuation characteristics of the gray values of the edge pixel point of the analysis area and the adjacent pixel point in the gradient direction of the edge pixel point, inner edge pixel points are determined according to the third probability, the inner edge pixel points are traversed in the main direction, shadow areas are jointly determined according to the traversed pixel points and preset conditions, the corrosion degree is determined according to the area ratio of all the analysis areas to all the shadow areas, and the accuracy of marine metal corrosion degree detection is improved.

The invention belongs to the technical field of multi-scene patrol analysis management, and particularly discloses a multi-scene intelligent shooting patrol analysis management system for regional security and fire protection. Comprising a regional information recording module, a regional security monitoring module, a regional security analysis module, a regional hidden danger monitoring module, a regional hidden danger analysis module, a regional fire-fighting monitoring module, a regional fire-fighting analysis module, a regional early warning confirmation module and a regional early warning management terminal. According to the invention, the pertinence of regional monitoring analysis is guaranteed by analyzing the security and protection qualified index, the fire hazard qualified index and the fire hazard qualified index of each monitoring region, so that the discovery timeliness of potential problems is improved, and meanwhile, according to the security and protection qualified index, the fire hazard qualified index and the fire hazard qualified index of each monitoring region, the monitoring accuracy is improved. And the early warning item of each monitoring area is confirmed, so that the accuracy and effectiveness of early warning are guaranteed, and the management efficiency of area management is improved.

The invention relates to the technical field of lithium metal battery health monitoring, and particularly discloses a lithium metal battery pack health monitoring system and method based on thin film sensors, the system comprises a plurality of thin film sensors distributed among lithium metal battery pack cells, and each thin film sensor comprises a temperature measuring unit, a pressure measuring unit, a flexible circuit board and silica gel foam; the adapter plate is electrically connected with the plurality of thin film sensors; the signal acquisition unit is electrically connected with the adapter plate and comprises a constant current source circuit, a signal conditioning circuit, a filter circuit and an analog-to-digital conversion circuit; the input end of the data transmission unit is electrically connected with the output of the analog-to-digital conversion circuit; the input end of the digital signal processor is electrically connected with the output end of the data transmission unit; and the input end of the battery management unit is electrically connected with the output end of the digital signal processor. The mechanical information and the temperature information between the battery monomers can be monitored at the same time, the battery management can evaluate the health state of the battery more comprehensively based on the information, and the abnormal problem can be warned in advance.

The invention relates to the technical field of new energy, in particular to a new energy equipment fault monitoring and analyzing method. According to the invention, the integrity and real-time performance of data are improved through data fusion and real-time state estimation, and the reliability and real-time performance of fault detection are enhanced; a nonlinear estimation method is combined with a physical model and a data driving model, so that the processing capacity of a complex nonlinear system is improved, and the defects in nonlinear system processing are overcome; multi-scale data fusion and comprehensive fault analysis are adopted, comprehensive analysis of different levels of data of the energy storage equipment is achieved, comprehensiveness and accuracy of fault detection are improved, the problems that in the prior art, accuracy and integrity of sensor data are seriously relied on, and an electric power system is often affected by noise, faults, loss and the like are solved, and the system reliability is improved. And meanwhile, the problems that a state estimation method is difficult to meet the requirement of real-time monitoring of a new energy power system and a traditional linearization processing method is poor in effect when a strong nonlinear system is processed are solved.

A blink monitoring method and system in a strong light environment, a medium and a product, the method comprising: a blink monitoring system determines a user's eyelid contour and a user's pupil visible portion from a user's eye image, predicts a user's pupil invisible portion based on user's complete pupil information, and determines a user's eye image; then calculating a pupil visible proportion, an effective pixel proportion and an eyelid height proportion; when the difference between the pupil visible proportion and the effective pixel proportion is within a preset error range, the blink monitoring system calculates the user eye closing degree corresponding to the user eye image according to the first weight of the eyelid height-width proportion and the second weight of the pupil visible proportion. In a preset time period, the blink monitoring system obtains a plurality of user eye images with the same time interval, and the user eye closing degree of each user eye image is determined in sequence, so that the blink condition of the user in the preset time period is evaluated. By implementing the method, the blinking index of the user can be accurately obtained, so that the fatigue state of the user is predicted.

The invention provides a multi-level encryption traffic classification system based on deep learning. The data collection module is used for providing original traffic data and providing input for subsequent data preprocessing and model training; the data preprocessing module is used for carrying out standardization processing on the original traffic, ensuring the consistency and stability of input data and providing high-quality input data for subsequent model training; a Globe-Local ResNet backbone network architecture is used for carrying out fusion extraction of global and local features for network traffic, and the detection capability of an original ResNet on the network traffic is effectively improved. And the multi-level classification head module is used for classifying the traffic step by step in a multi-level mode and reducing the problem of sharp increase of the number of categories and the problems of increase of the number of categories and instability of the model. According to the method, efficient multi-dimensional encrypted traffic classification is realized, and through reasonable data processing and model design, the complexity improvement and the accuracy reduction caused by sharp increase of the number of categories are avoided while the classification precision is ensured.

The invention discloses a training method for a self-supervised microscopic image processing neural network, and the method comprises the steps: obtaining a fluorescent microscopic image data set through employing an optical microscopic imaging system with a time-dependent single photon counting function, and enabling the fluorescent microscopic image data set to comprise a photon counting distribution set {Gmn (t)} MN for M * N pixel points, m and N are respectively integers greater than 1, m is from 1 to M, n is from 1 to N, and t represents time; and constructing a training data set for a neural network comprising a fluorescence lifetime prediction convolution module and / or a denoising convolution module from the photon counting distribution set of the M * N pixel points in a photon flow redistribution mode.

The invention relates to an indoor positioning method and device based on transfer learning, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: constructing a channel state information image according to channel state information of a position to be positioned in a target domain; inputting the channel state information image into a target domain positioning model to obtain a positioning result of the to-be-positioned position output by the target domain positioning model; the target domain positioning model is obtained by training the following steps: pre-training an initial model by using a source domain sample set to obtain a source domain positioning model; carrying out migration training on the source domain positioning model by using the target domain sample set to obtain a target domain positioning model; the target domain sample set comprises a first target domain sample constructed according to the first channel state information, and a second target domain sample constructed according to second channel state information obtained by expanding similar samples in the source domain sample set and the first target domain sample. By adopting the method, the positioning accuracy in the target domain can be improved.

The invention discloses a mesenchymal stem cell membrane coated kaempferol bionic nano-drug as well as a preparation method and application thereof, and relates to the technical field of targeted drug preparation, the bionic nano-drug is composed of nano-particles containing kaempferol and a mesenchymal stem cell membrane coated outside the nano-particles, the mesenchymal stem cell membrane is derived from umbilical cord mesenchymal stem cells; the particle size of the bionic nano-drug is 167.67 + / -1.84 nm, the PDI is 0.181 + / -0.027, the potential is-5.41 + / -0.74 mV, the encapsulation efficiency is 79.97 + / -0.19%, and the entrapment rate is 7.27 + / -0.02%. According to the invention, the water solubility of the kaempferol is enhanced, the specific enrichment of the kaempferol at the diseased region of the AS can be enhanced, the utilization efficiency of Kae is improved, and the defects of small molecular drugs are overcome; in addition, through nanocrystallization of Kae, direct contact of cells and tissues with Kae can be reduced, and cytotoxicity is reduced.

The invention provides a motor health assessment method based on multi-dimensional parameters. The motor health assessment method comprises the following steps of S1, collecting the multi-dimensional parameters of a motor during operation through a sensor; s2, cleaning, denoising and normalizing the collected data to obtain fusion state parameters; s3, the fault samples are marked according to fault types, degrees and position information, and healthy samples and fault samples are divided; s4, constructing a health evaluation function based on the health sample; s5, solving the evaluation function to obtain a health evaluation value; and S6, evaluating the health state of the motor according to the evaluation value. According to the method, data input is optimized by fusing multi-dimensional parameter information and weighted calculation, and the health state is quantified in combination with the Euclidean distance function, so that the evaluation result is more accurate. According to the method, the health level and the fault risk of the motor can be effectively identified, and an intelligent and scientific health assessment means is provided for maintenance and operation of a motor system.

The invention belongs to the field of electrochemistry, and discloses a nitro functionalized Zn-Mo bimetal organic framework solid electrolyte and a preparation method thereof. The preparation method comprises the following steps: firstly, synthesizing Zn-Mo bimetal organic frameworks MOFs with nitro functional groups through a reflux method, blending the Zn-Mo bimetal organic frameworks MOFs with a lithium salt and a polymer to obtain viscous slurry, pouring the viscous slurry on a glass plate to cure the viscous slurry, standing, and carrying out vacuum drying at 60 DEG C to obtain the nitro functional Zn-Mo bimetal organic framework solid electrolyte. The preparation method is simple and easy to operate, and the electrochemical stability and thermal stability of the battery can be effectively enhanced; the lithium ion transference number of the prepared diaphragm is up to 0.70, the assembled lithium iron phosphate battery is 120 mAh g <-1 > under the rated capacitance of 1C, the capacity retention ratio is close to 95%, and the lithium iron phosphate battery shows good cycling stability and multiplying power.

The invention relates to a vehicle regenerative braking control method, device and system, a vehicle and a storage medium. The vehicle regenerative braking control method comprises the following steps that in response to starting of an anti-lock braking system, a wheel speed signal of the vehicle is obtained; judging whether the vehicle vibrates or not based on the obtained wheel speed signal; under the condition that the vehicle does not generate vibration, the braking of the vehicle is kept in a common ABS hydraulic braking mode; in a case where vibration is generated in a vehicle, it is determined whether to execute electric regenerative braking based on a maximum recoverable torque currently executable by a motor and a target electric braking torque, and in a case where the maximum recoverable torque is greater than a predetermined value and the target electric braking torque is less than or equal to the maximum recoverable torque, the maximum recoverable torque is greater than a predetermined value and the target electric braking torque is greater than or equal to a predetermined value. Gradually converting braking of the vehicle into an electric regenerative braking mode; otherwise, the braking of the vehicle is kept in a common ABS hydraulic braking mode. According to the invention, vibration generated during low-load braking of the vehicle can be reduced or completely eliminated.

The invention relates to the technical field of artificial intelligence, in particular to an efficiency evaluation method and device for intelligent operation and maintenance of a cluster system. According to the method, the efficiency of the cluster system in different operation stages can be scientifically and accurately evaluated by constructing the reasonable efficiency index system, the network average efficiency calculation model, the task execution balance degree evaluation model and the elasticity evaluation model. According to the method, the deep reinforcement learning algorithm is introduced, the multi-dimensional situation information is collected, analyzed and predicted in real time, and the intelligent operation and maintenance decision-making efficiency of the cluster system can be effectively improved.

A soilless culture incubator for forestry seedling raising comprises an incubator body, incubator supporting legs, a nutrient solution discharging pipe, a valve, a culture partition plate, a filter screen outer frame, a U-shaped handle, a rectangular baffle, a centralizing hopper, a nutrient solution box and a water inlet pipe, the lower portion of the incubator body is connected with the incubator supporting legs, the four incubator supporting legs are distributed at the four corners of the incubator body, and the nutrient solution discharging pipe is arranged on the left side of the incubator body. The lower portion of the nutrient solution discharge pipe is connected with a valve, the cultivation partition plate is matched with the cultivation box and inserted into the cultivation box, the cultivation partition plate is provided with a plurality of rectangular flow holes and a plurality of vertical insertion holes, the rectangular flow holes are communicated with the vertical insertion holes, the outer side of the filter screen is connected with a filter screen outer frame, and the filter screen outer frame is matched with the vertical insertion holes. The filter screen outer frames are inserted into the vertical insertion holes, the filter screen outer frames slide in the vertical insertion holes, the filter screens are opposite to the rectangular flow holes, and the multiple filter screen outer frames are distributed on the cultivation partition plate.

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 utility model discloses a glass curtain wall with a fireproof structure, which relates to the technical field of glass curtain walls and comprises four groups of curtain wall mounting frames, a connecting block is arranged between every two groups of curtain wall mounting frames, and connectors are arranged between two ends of each curtain wall mounting frame and the inside of the corresponding connecting block. A fixing rod is arranged between the connecting block and the connector, two sets of sealing rings are arranged between the outer wall of the fixing rod and the connecting block, mounting bases are arranged on the inner side wall of the curtain wall mounting frame, double-layer vacuum glass is arranged among the four sets of mounting bases, and mounting grooves are formed between the double-layer vacuum glass and the mounting bases. A water storage box is arranged in the connecting block, and a water storage pipe is arranged in the middle of the curtain wall mounting frame. According to the glass curtain wall with the fireproof structure, the temperature in the curtain wall mounting frame can be reduced, the mounting frame is prevented from deforming in a fire disaster, and the curtain wall structure is better protected.

The invention discloses a honeycomb sandwich structure infrared image defect detection method and system based on improved YOLOv5. The method comprises the following steps: collecting defect images of a honeycomb sandwich structure to form a sample data set; training defect images in the data set based on an improved YOLOv5 model, predicting positions and categories of honeycomb sandwich structure defects, training an optimization model by using a loss function until convergence, and obtaining and storing optimal model weight data; the improved YOLOv5 model is based on a YOLOv5 architecture, a Swin Transform module is embedded in a backbone network, and the detection performance of the model on honeycomb sandwich structure defects is improved in combination with a C3 convolutional layer, a CBS module and an SPPF module; and acquiring a defect image of the honeycomb sandwich structure to be detected, detecting and identifying the defect image by using the improved YOLOv5 model based on the optimal model weight data, and generating and outputting a detection result. According to the method, small-size defect identification in multi-scale defect identification can be effectively realized, and the defects of a traditional convolutional neural network in processing complex backgrounds and small target detection are overcome.

This invention discloses a multi-vehicle cooperative control method, system, medium, and program product. The method includes constructing a lateral and longitudinal coupled dynamic model for the multi-vehicle cooperative transportation system and extracting the model matrix of the lateral and longitudinal coupled dynamic model; constructing system constraint equations for the multi-vehicle cooperative transportation system with the cargo as the leader, including motion constraints and articulation constraints, and extracting the constraint matrix of the system constraint equations; obtaining the real-time state of the multi-vehicle cooperative transportation system; calculating the constraint forces of the multi-vehicle cooperative transportation system based on the model matrix of the lateral and longitudinal coupled dynamic model, the constraint matrix of the system constraint equations, and the real-time state; and calculating the actual control quantities of each vehicle based on the constraint forces of the multi-vehicle cooperative transportation system. This invention aims to solve the challenges of strong coupling, nonlinearity, and underactuation in existing multi-vehicle cooperative transportation systems, and improve the tracking performance of multi-vehicle cooperative transportation systems under topologically stable configurations.

The server acquires a deterioration degree, a charging rate, and an initial full charge capacity of the battery pack mounted on the electric moving body at a lease start time, calculates an amount of power of the battery pack at the lease start time on the basis of the deterioration degree, the charging rate, and the initial full charge capacity, and determines a lease state of the battery pack on the basis of the amount of power. The rental fee of the electric moving body or the rental fee of the battery pack with the electric power less than the current full charge capacity is calculated, and the rental fee is output by calculating the rental fee of the electric moving body with the built-in battery pack with the electric power less than the current full charge capacity.

The invention discloses a network traffic classification method and device, equipment, a storage medium and a product, relates to the technical field of artificial intelligence, and discloses a network traffic classification method comprising the following steps: obtaining a network traffic data packet collected by a network shunt; the network flow data packet is pre-processed, real network flow data samples are obtained, and the real network flow data samples comprise marked flow data samples and unmarked flow data samples; generating analog traffic data samples through a generator, and inputting the analog traffic data samples, the marked traffic data samples and the unmarked traffic data samples into a discriminator to obtain the probability that each traffic data sample belongs to each traffic type; and inputting the probability that each traffic data sample belongs to each traffic type into a classifier to obtain the traffic type corresponding to each traffic data sample, thereby improving the network traffic classification efficiency.

The invention provides a document-level relation extraction method integrating a self-encoding language model and a generative language model, which comprises the following steps of: firstly, extracting a data set by using a document-level relation, finely adjusting a document-level relation dichotomy model which takes a small self-encoding pre-training language model as an encoder, and carrying out dichotomy prediction on whether a relation exists or not on all entity pairs in a document; filling the cue word template with the document and the entity pairs with the relationship, finely adjusting the generative large language model, and generating the relationship category of the entity pairs; and finally, aligning a generation result of the large language model with a preset relation label by utilizing a natural language reasoning model so as to realize final document-level relation extraction. The invention provides an efficient and accurate document-level relationship extraction method, overcomes the defects of a traditional method in the aspects of long text information extraction and logical reasoning, and remarkably improves the performance and reliability of document-level relationship extraction.