172results about How to "Achieve capture" patented technology

The invention discloses an e-commerce platform commodity recommendation method and system based on user preference analysis, and relates to the technical field of e-commerce platforms, and the method comprises the steps: constructing a user multi-modal interest feature vector; the multi-modal interest feature vector fuses the text search history, the image browsing preference, the audio consumption habit and the video interaction behavior of the user; utilizing a cross-modal attention mechanism to match the multi-modal interest feature vector of the user with multi-modal attribute information of commodities to generate a candidate commodity set; and applying a dynamic context-aware reinforcement learning model to reorder the candidate commodity set according to real-time interaction behaviors of the user and environmental factors, and outputting a final recommendation result. According to the method, multi-modal interest feature extraction, dynamic context perception and reinforcement learning technologies are combined, so that comprehensive capture and accurate modeling of user interests are realized.

The invention discloses a construction site digital-to-analog conversion and collaborative management method based on a distributed computing architecture. The method comprises the following steps: S1, collecting microscopic level data of a construction site; s2, obtaining mid-level data of the construction site; s3, acquiring macroscopic hierarchical data, and constructing a multi-scale spatio-temporal data set; s4, decomposing and processing the multi-scale spatio-temporal data set by using a spatio-temporal diagram convolutional network, and constructing a dynamic digital twin model; s5, designing a competition and cooperation mechanism between agents by using a game theory model, solving resource allocation conflicts through Nash equilibrium, and dynamically optimizing resource scheduling; s6, optimizing a scheduling strategy of the intelligent agent through a distributed algorithm based on near-end strategy optimization; s7, detecting errors in digital-to-analog conversion and scheduling processes by using a belief propagation algorithm; and S8, optimizing a scheduling strategy through a distributed adaptive feedback control mechanism. According to the invention, a distributed computing architecture, space-time diagram convolution, an intelligent scheduling technology and the like are adopted, and digital-to-analog conversion and resource management of the construction site are optimized.

The invention provides an LED packaging material defect detection method, which comprises the following steps: performing image enhancement processing on each layer of separated image to obtain each layer of clear material image; carrying out local fine scanning on the identified cross defect area to obtain micron-sized three-dimensional structure data, and reconstructing fine morphology characteristics of the cross defect; according to the reconstructed cross flaw structure, combining various material performance parameters pre-stored in a material database to simulate and calculate the influence of flaws on the LED performance, including optical, thermal and mechanical properties; and judging the position, the size, the type and the influence degree on the performance of the flaws, and carrying out quantitative scoring on the overall quality of LED packaging to obtain a final detection result and quality grade judgment.

The invention provides a lottery store violation abnormity identification method and system, and relates to the technical field of image identification, and the method comprises the steps: collecting multi-source heterogeneous image data in real time through an edge computing device; preprocessing the collected multi-source heterogeneous image data to form a stereo image data set containing space, depth and thermodynamic information; constructing a multi-task deep learning model, wherein the multi-task deep learning model comprises a face recognition module, a behavior analysis module and an environment detection module; inputting the three-dimensional image data set into the trained multi-task deep learning model, and executing face recognition, behavior analysis and environment detection tasks at the same time; based on an execution result of the multi-task deep learning model, generating a violation score by applying a fuzzy comprehensive evaluation method, and weighting by considering historical violation records; and generating a lottery store site investigation plan based on the violation scoring result.

The invention relates to an earthquake fracture inversion method based on multi-source observation data. The method comprises the following steps: acquiring a seismic source mechanism solution of a specified earthquake event; according to the seismic source mechanism solution, geometric parameters of a fault of the seismic event are obtained; generating a three-dimensional grid model of a fault plane to which the fault belongs according to a pre-constructed geometric shape model of the fault and the geometric parameters of the fault; according to the limiting condition of each sub-fault in the three-dimensional grid model, obtaining the slippage of the sub-fault to which the seismic fracture belongs; and according to the observation data to which the seismic event belongs and the slippage of the sub-faults, performing inversion verification by means of an inversion model, and if the verification is passed, taking the slippage of all the sub-faults as a seismic fracture inversion result when the verification is passed. According to the method, the stability and reliability of an inversion result are improved, the sliding vector at each sub-fault can be better estimated, and the sliding process of the whole seismic fault can be more comprehensively understood.

The invention relates to the technical field of offshore wind power operation risk analysis, and discloses an offshore wind power operation risk analysis system based on deep learning, and the system comprises a data collection and preprocessing module which is used for collecting wind power equipment operation state data and environment data of multiple types of sensors, and carrying out the cleaning and preprocessing of the data; the multi-modal data fusion and feature extraction module is used for extracting and fusing multi-modal features from different data sources; and the real-time risk prediction and intelligent decision module is used for analyzing the operation state of the wind power equipment in real time based on a deep learning model, and performing prediction and alarm when a fault trend is detected. By introducing a self-attention mechanism and a Transform architecture, deep fusion analysis is performed on multi-modal data from wind power equipment, and information of different data sources can be weighted and integrated in a unified feature space, so that the system can adaptively focus on important features, and the analysis precision and fault detection capability of the multi-modal data are remarkably improved.

The invention relates to the technical field of polaroid performance optimization, and discloses a performance optimization design method and system for a multi-layer composite polaroid, and the method comprises the steps: carrying out the polarization spectrum analysis of a use environment of a display screen, obtaining incident polarization spectrum data, and calculating the initial stretch ratio data of each polarization layer; measuring to obtain molecular orientation degree data and polarization performance data of each polarization layer; performing feature extraction and fusion to obtain a fusion feature vector of each polarization layer; calculating the influence of the molecular orientation degree change of each polarized light layer on the optical performance of different polarization states to obtain the importance weight of the polarization state of each polarized light layer; reinforcement learning optimization is carried out, and a stretch ratio adjustment strategy of each polarization layer is generated; according to the technical scheme, multi-objective optimization and sequential quadratic programming are carried out on the processing parameters of each polarizing layer, and the stretching temperature and speed control sequence of each polarizing layer is generated, so that the precise control of the stretching temperature and speed is realized, and the performance index and the production efficiency of the polaroid are improved.

The invention relates to the technical field of data mining, and provides a spatial correlation interest point recommendation method, system and device, a product and a medium. The method comprises the following steps: constructing a position vector pre-training network, and training a position vector by using the position vector pre-training network; a position vector pre-training network is utilized to generate position vectors for describing position information of the interest points, a track coding network is constructed, the position vectors are fused, sign-in data of each user are converted into track slice vectors, and interest point preference scores of the users are generated; and combining a multi-target learning function of a user access target and a space association target set, and generating loss by utilizing interest point preference scores to realize training of a track coding network. According to the method, the interest point recommendation requirement is met from the two perspectives of geographic feature representation and geographic association target learning of the user sign-in data, the spatial clustering mode in the user moving behavior is effectively captured, and the interest point recommendation effect is improved.

The invention provides an integrated containment pressure suppression, filtration and discharge system. The integrated containment pressure suppression, filtration and discharge system comprises a containment, a discharge valve, a discharge pipeline, a spray pipe, a first-stage pressure suppression water tank, a second-stage pressure suppression water tank, a baffle plate steam-water separator, an expansion tank, a rupture disk structure, a combined metal fiber filter layer, a radioactivity monitor, a first stop valve and a second stop valve. The two-stage pressure suppression water tank is arranged, the high removal efficiency of airborne radioactive substances is achieved through two-stage bubbling water washing, then the combined metal fiber filter is arranged by means of an existing cabin, and the requirements for rapid pressure reduction and radioactive substance removal of the containment under the LOCA accident working condition can be met; meanwhile, the problem that the containment filtering and discharging system of the existing nuclear power station occupies too many resources is solved, and the economical efficiency is high.

The invention relates to the technical field of organic chemistry and environmental science, in particular to a fluorine-philic and hydrophobic functional monomer, a synthesis method and application thereof, a fluorine-containing compound detection electrode based on the functional monomer and a preparation method of the fluorine-containing compound detection electrode. The chemical structure of the fluorine-philic and hydrophobic functional monomer comprises pyrrole-3-carboxylic acid and fluorine telomer alcohol bonded with the pyrrole-3-carboxylic acid. A long fluorocarbon chain is introduced into a basic functional monomer, so that the fluorine affinity and hydrophobicity of the monomer to a fluorine-containing compound are remarkably improved. Therefore, a derivative pyrrole-3-carboxylic acid of pyrrole is used as a basic functional monomer, and the fluorine telomeric alcohol is bonded on the pyrrole-3-carboxylic acid, so that a novel functional monomer with fluorine-philic and hydrophobic functions is synthesized. The functional monomer realizes specific capture of fluorine-containing organic matter molecules by simultaneously utilizing amido with affinity to perfluorinated compounds in a pyrrole ring and unique fluorine-fluorine interaction force. A perfluorinated compound detection method established based on the functional monomer can avoid interference caused by a sample matrix.

The invention relates to the field of image processing, and discloses a visual question-answering method, system and device based on a remote sensing tampered image, and a medium, and the method comprises the steps: carrying out the visual feature extraction of the tampered image through a main branch network, and carrying out the feature extraction of the input features of each layer of encoder through an edge prior guide block; performing edge detection on the image through a prior branch to obtain an edge prior feature, fusing the edge prior feature with an input feature in each edge prior guide block, and outputting the fused edge prior feature as an input feature of a next edge prior guide block to perform visual feature extraction; carrying out feature extraction on an input text through edge prior feature guidance, and carrying out cross-modal fusion on extracted visual features and text features; performing multi-modal reasoning according to the fused features, and outputting a question and answer result; according to the method, the prior branches are introduced to provide the edge prior features for the main branches, so that the edge artifacts of the duplicated-moved tampered image are remarkably recovered and enhanced, and the accuracy of analysis and understanding of the tampered image is improved.

The invention relates to the field of computer vision and artificial intelligence, in particular to a real-time multi-person motion analysis system based on motion perception and semantic understanding, which comprises a motion perception module, a motion tracking module, a motion semantic understanding module and a motion performance analysis module. The motion area segmentation unit adopts a background modeling or inter-frame difference method to identify a motion area and generate a motion mask; the target detection unit is connected with the central processing unit through a data bus and used for processing video frame data and detecting the position and the bounding box of a moving target. The motion area segmentation unit is connected with the target detection unit and is used for generating a motion mask. By integrating a plurality of modules, the motion data can be comprehensively captured and analyzed, and the accuracy and real-time performance of motion analysis are improved. Through combination of the target detection unit and the motion area segmentation unit, the system can accurately identify a motion target and a motion area, and a solid foundation is laid for subsequent tracking and analysis.

The invention provides a self-supervised medical image segmentation method and system based on structural relation modeling, and the method comprises the steps: carrying out the modeling of a complex relation among nodes, sub-graphs and a whole graph, thereby comprehensively capturing the structural information of the graph; by constructing and optimizing the structural relation model of the graph, the performance of graph representation learning is improved, the method is particularly suitable for graph data with a complex topological structure and non-uniform distribution, and the segmentation effect of the medical image is improved by modeling the organization structure relation in the medical image.

The invention discloses an artificial intelligence fake face identification method, apparatus and device, and a storage medium. The method comprises the steps of obtaining a to-be-identified face image; inputting the face image into a pre-trained artificial intelligence fake face identification model, and outputting an identification result of a face in the face image; wherein the artificial intelligence fake face identification model is obtained by training an image processing deep learning model based on a multi-head self-attention mechanism; a training data set of the artificial intelligence fake face identification model is 1000 face images marked with a real face and 1000 artificial intelligence fake faces; the identification result comprises the real face and the artificial intelligence forged face. By utilizing the method, the global information of the image is captured and efficiently processed, so that the accuracy and efficiency of face identification are improved.

The embodiment of the invention provides an image retrieval method and device, equipment, a storage medium and a program product. The method comprises the following steps: on the basis of a local feature vector and a global feature vector corresponding to a to-be-retrieved image, matching a plurality of candidate images from a preset retrieval library; calculating a local feature overlap ratio between the to-be-retrieved image and each candidate image based on the local feature vector corresponding to each candidate image and the local feature vector corresponding to the to-be-retrieved image; for each candidate image, based on the corresponding local feature overlap ratio, fusing the global feature vector corresponding to the candidate image and the global feature vector of the to-be-retrieved image to obtain a fused global feature vector corresponding to the candidate image; and calculating the similarity between the global feature vector corresponding to the to-be-retrieved image and the fused global feature vector corresponding to each candidate image, and determining a retrieval result of the to-be-retrieved image based on the similarity. According to the method, the retrieval accuracy can be improved.

The invention relates to a non-cooperative target satellite large-tolerance docking device, and belongs to the field of satellite on-orbit service docking. Comprising a target satellite bottom plate, three mechanical arm capturing units, three locking units and a docking device mounting plate, wherein the butt joint device mounting plate is of a horizontally-arranged circular plate-shaped structure; the three mechanical arm capturing units and the three locking units are mounted on the upper surface of the docking device mounting plate; the target satellite bottom plate is located above the mechanical arm capturing unit and the locking unit; the target satellite bottom plate is grabbed through the three mechanical arm capturing units, and the target satellite bottom plate is dragged to a locking position; the target satellite bottom plate is locked through the three locking units; the functions of capturing, locking and separating a space non-cooperative target satellite and the like are realized, so that on-orbit services such as maintenance and refueling of a non-cooperative target are facilitated.

The invention discloses a privacy-protected multi-scale multi-attention IoT (Internet of Things) attack detection method, which comprises the following steps of: firstly, acquiring an IoT attack data set, and preprocessing IoT attack data in the IoT attack data set; and secondly, constructing an IoT attack detection deep learning model integrating multi-scale feature extraction and a lightweight multi-attention module, and inputting preprocessed IoT attack data to obtain an IoT attack detection result. And finally, combining the IoT attack detection deep learning model with federated learning, and obtaining an optimal IoT attack detection deep learning model through multiple iterations and parameter tuning. According to the method, the detection capability of multi-stage attacks in a dynamic environment is improved, meanwhile, the lightweight of the module is kept, the method is suitable for IoT equipment with limited resources, and the data privacy is effectively guaranteed.

The invention discloses a small sample action recognition method, electronic equipment and a medium. The method comprises the following steps: acquiring a support set video, a query set video and a category label template; performing feature processing on the support set video, the query set video and the category label template through a preset multi-modal training model to obtain a first visual feature vector, a category-level semantic feature and a second visual feature vector; inputting the three data into a proxy semantic diffusion module for noise diffusion processing; performing feature representation on the first instance-level proxy semantics, the class-level semantic features and the first visual feature vector output by the proxy semantic diffusion module to construct a class prototype, and performing feature fusion on the second instance-level proxy semantics to obtain query feature representation; and performing action recognition on the query feature representation and the category prototype to determine an action type of a second action video frame in the query set video. According to the embodiment of the invention, semantic details of an instance level can be captured, and the action recognition precision is improved.

The invention relates to the technical field of sewage treatment, in particular to a system for efficiently treating livestock and poultry breeding sewage by utilizing algal-bacterial symbiosis, which comprises a solid-liquid separation module arranged on the ground, the solid-liquid separation module comprises a water inlet pipe, a separation box and a filter basket, a nitrification and denitrification tank is arranged on the top of the ground, and a stirring assembly is arranged in the nitrification and denitrification tank; the chemical phosphorus removal module comprises a coagulation, flocculation and sedimentation tank and a stirring assembly, the coagulation, flocculation and sedimentation tank is arranged on the ground, and the stirring assembly is arranged in a coagulation and flocculation area in the coagulation, flocculation and sedimentation tank; according to the system for efficiently treating the livestock and poultry breeding sewage by utilizing the bacterial-algae symbiosis, compared with an existing method for removing the total nitrogen content in the sewage by independently utilizing microorganisms, pollutants in the sewage are synergistically removed in the oxidation pond by utilizing a bacterial-algae symbiosis structure, microalgae and bacteria, so that the energy consumption and the carbon source are saved, and the environmental pollution is reduced. Meanwhile, the sewage treatment effect is improved.

The present invention relates to the technical field of database data capture, and specifically provides a metadata change capture method, which has the following steps: S1, executing CDC statements in a distributed database with a metadata capture function; S2, a user executes a DDL operation, and the database service syntax parses SQL to generate a syntax tree, and then generates an execution plan through the syntax tree; S3, the database sends the execution plan to each execution node for execution, and after the plan is executed, the metadata change record is updated to the system table, and the monitor pushes these metadata change records to the changefeed; S4, judging whether the changefeed sending mode is the DDL sending mode, if yes, directly sending the DDL; if not, waiting for the sending mode to switch to the DDL sending mode before sending. Compared with the prior art, the present invention does not need to change the decentralized structure of the database, does not rely on the timing service of the central node, and ensures the transaction semantics of the DDL is captured through a series of judgments, and does not rely on the central node database to still maintain good scalability.

The invention discloses an automatic news event extraction and combination system based on chapter-level text analysis, which is characterized in that text noise is cleaned and paragraphs are divided through a preprocessing module, and strong event candidates are screened by combining multi-modal conjoint analysis of picture titles and paragraph texts and utilizing semantic similarity calculation and named entity recognition technologies; a multi-task learning framework based on a Transform architecture is adopted to cooperatively extract elements such as time, places, figures, events, causes and passes, and the comprehensiveness and accuracy of event element extraction are improved; and the event similarity is calculated through a Bi-Encoder model, and the events are combined and sorted in combination with timestamps, so that an event chain with clear logic is generated. The system breaks through the limitation of existing sentence-level or paragraph-level analysis, remarkably improves the comprehensiveness, consistency and coherence of news event identification through chapter-level global context, data multi-mode combination and intelligent event chain construction, and has wide application value in the fields of news monitoring, public opinion analysis and the like.

The present invention discloses an automatic installation tool for a lightning protection and electric detection grounding ring for a distribution network based on a drone, including a drone system, an automatic installation system and a hoisting system. The automatic installation system includes a device housing, a mounting platform, a lower grounding ring clamp assembly, a cable clamp assembly, a laser sensor assembly and an upper grounding ring clamp assembly. The mounting platform is installed in the device housing; two laser sensor assemblies are provided and respectively installed on the front and rear end surfaces of the mounting platform; two cable clamp assemblies are provided and respectively installed on the front and rear sides of the upper end surface of the mounting platform; the lower grounding ring clamp assembly is installed below the mounting platform; and the upper grounding ring clamp assembly is installed on the mounting platform. The present invention has a compact structure, utilizes drones and automatic installation tools, realizes the automatic installation of the lightning protection and electric detection grounding ring for the distribution network without power outage, greatly reduces the manpower demand, significantly improves the efficiency and level of live operations in the distribution network, and is safe, stable and reliable.

The invention relates to a polymer and a preparation method thereof, a binder, a diaphragm, a manganese-based lithium battery and a vehicle, and relates to the field of lithium batteries. The polymer comprises a main chain polymer and a first branch chain, the main chain polymer comprises a structural unit shown in the formula I, the first branch chain comprises a chelating agent and # imgabs0, and Y in the formula I represents the bonding position of the main chain polymer and the first branch chain. Therefore, the binder comprising the polymer has excellent transition metal ion capturing ability, bonding effect and mechanical strength, and the corresponding diaphragm has excellent transition metal ion capturing ability, bonding effect and thermal stability, so that the rate capability, cycle life and safety of the manganese-based lithium battery are improved.

The invention discloses a distributed photovoltaic load prediction method and device based on ISO and LightGBM, and a medium, and the method comprises the steps: carrying out the load prediction of meteorological load data through a LightGBM model if the data size of the meteorological load data is larger than a preset threshold range, and obtaining a load prediction value; if the data volume is within the threshold range, performing load prediction on the meteorological load data through a physical model to obtain a load prediction value; if the data volume is smaller than the threshold range, load prediction is carried out on the meteorological load data in a power curve fitting mode, and a load prediction value is obtained. According to the distributed photovoltaic load prediction method and device based on ISO and LightGBM and the medium provided by the invention, a machine learning model, a physical model and a statistical method are comprehensively applied, and an appropriate prediction method is selected according to different data conditions, so that the advantages of various methods can be brought into full play, the prediction accuracy and efficiency are improved, and the prediction efficiency is improved. The problem that the accuracy of distributed photovoltaic load prediction is difficult to improve can be solved.

The invention discloses a ground feature classification method, system and device based on time-space spectrum feature collaboration, and the method comprises the steps: obtaining a time sequence remote sensing image set, and carrying out the processing of the time sequence remote sensing image set through a graph attention model, and obtaining key time spectrum features; processing the time sequence remote sensing image set and the key time spectrum features through a space attention model to obtain space features; converting the key time spectrum features and the spatial features through a feature transformation network to obtain space-time spectrum three-dimensional features; constructing a ground feature classification pre-training model, and training the ground feature classification pre-training model based on the space-time spectrum three-dimensional features to obtain a ground feature classification model; and reasoning the time sequence remote sensing image of the to-be-measured target based on the ground feature classification model to obtain a ground feature classification result. According to the method, high-precision classification based on time-space spectrum multi-dimensional features is realized, and the ground feature classification model not only can realize high-precision classification on a limited data set, but also has relatively strong generalization ability, and can be suitable for ground feature classification tasks in different scenes.

The invention discloses a dynamic sound field modulation and particle control method based on digital microfluidics, modulation of a sound field is achieved through liquid drops, the liquid drops are located between an upper pole plate and a lower pole plate, air is around the liquid drops, sound waves are located on the paths of the lower pole plate, the liquid drops and the upper pole plate, and due to the fact that the acoustic impedance difference between liquid and glass is not large, the sound field is modulated. Most of energy of sound waves can be transmitted through transmission, most of the energy of the sound waves is difficult to transmit due to the fact that acoustic impedance difference between air and glass is large and most of the energy of the sound waves is reflected at an interface of the air and the glass, the sound waves selectively penetrate at specific positions by controlling the number and arrangement of the liquid drops, so that a specific sound field is formed, and real-time movement of the liquid drops is controlled. The sound field formed by penetration is changed in real time, so that the dynamic modulation of the sound field is achieved, the sound field obtained by modulation acts on the solution, the specific movement of the particles is obtained through the sound field, the control of the particles is realized, and the application prospect in the aspects of micro-assembly and cell control is very good.

The invention relates to the technical field of energy utilization, in particular to a selective sunlight absorption / dynamic heat radiation regulator for a temperature difference thermoelectric generator and a manufacturing method of the selective sunlight absorption / dynamic heat radiation regulator. The modulator comprises an antireflection layer, a short wave absorption metal ceramic layer, a metal layer, a visible near-infrared absorption metal ceramic layer, a phase change layer and a substrate layer, the short wave absorption metal ceramic layer, the metal layer, the visible near-infrared absorption metal ceramic layer and the phase change layer form a nano cavity, and the layers are sequentially arranged in an overlapped mode. The modulator can realize excellent selective absorption and radiation refrigeration dynamic spectral response in the whole spectrum range of 250-25000nm.

The invention discloses a device and a method for treating and recycling radioactive CO2 gas in nuclear-grade waste combustion products. The device comprises a combustion unit, a combustion unit and a recycling unit, the filtering unit is connected with the combustion unit and used for filtering mixed gas containing radioactive CO2 generated in the nuclear-grade waste combustion process; the cooling unit is connected with the filtering unit; the purification unit is connected with the cooling unit; the trapping and regenerating unit is connected with the purifying unit; and the circulating unit is connected with the combustion unit and the trapping and regenerating unit. The device is simple in structure, the adsorption material is convenient to regenerate and replace, the device can be repeatedly used for a long time, CO2 can be selectively adsorbed from nuclear power plant mixed waste gas or other mixed gas, the development requirements of carbon emission control and carbon neutralization are met, environmental radiation pollution caused by nuclear power plant waste gas or other waste gas in the nature can be relieved, and the environmental pollution is reduced. Meanwhile, carbon isotope resources can be effectively recycled, the process is safe and controllable, and the risk of radioactive substance leakage is avoided.

The invention relates to a high-efficiency face deep counterfeiting detection method based on multi-frequency feature aggregation, belongs to the technical field of computer vision and pattern recognition, focuses on detection of a face counterfeiting image, and provides a high-efficiency face deep counterfeiting detection method based on multi-scale spatial feature and multi-frequency domain feature fusion. The method aims at improving the robustness and generalization ability of a detection model under the conditions of coping with different counterfeiting means, noise and frequency distortion, and is suitable for various deep counterfeiting detection application scenes. The multi-context and frequency aggregation network provided by the invention aims to realize efficient and robust deep counterfeiting detection through innovative design of a multi-core aggregator and a multi-frequency aggregator module; the defects of an existing deep forgery detection technology in the aspects of space and frequency feature collaborative modeling, multi-scale feature extraction, high-frequency detail processing and cross-domain generalization ability are overcome.

The invention belongs to the technical field of computer communication and artificial intelligence, is applied to the field of smart cities, and relates to a page loading method which comprises the following steps: receiving a page access request of a user; judging whether the target page has an abnormal cache or not; if yes, receiving an instruction sent by the user for emptying the abnormal cache and reloading the target page; according to the instruction, the abnormal cache is emptied, and the target page is reloaded; or triggering a monitoring module pre-deployed on the target page to capture the abnormal cache; and controlling the monitoring module to empty the abnormal cache and reload the target page. The invention further provides a page loading device, computer equipment and a storage medium. In addition, the invention also relates to a block chain technology, and the page cache content of the user can be stored in a block chain. According to the method, the advantages of the cache are reserved, the cache is cleared and the page is reloaded only when the abnormal cache occurs, the fluency when the user uses the H5 page is improved, the customer complaint rate is reduced, and business popularization is facilitated.