166 results about "Collection analysis" patented technology

The invention discloses a collection, analysis and utilization linkage method for the full life cycle of vulnerability intelligence, which comprises the following steps: respectively acquiring original vulnerability intelligence of a plurality of heterogeneous data sources, and carrying out standardization processing on the original vulnerability intelligence to obtain vulnerability description information; constructing an asset knowledge graph, wherein the asset knowledge graph is used for representing IT asset objects in the enterprise and dependency and deployment relationships among the IT asset objects; performing graph traversal query on the vulnerability description information based on an asset knowledge graph, and identifying an affected asset list affected by vulnerabilities; and according to a preset response strategy, generating a safety response instruction from the affected asset list and sending the safety response instruction to the safety protection system. According to the method, the asset knowledge graph is constructed and the multi-source vulnerability information is subjected to standardization and mapping correlation analysis, so that the accurate and rapid identification of the affected assets of the power system and the automatic response strategy generation are realized, and the efficiency and the accuracy of the vulnerability management of the key information infrastructure of the power grid are effectively improved.

The invention belongs to the technical field of cable fault positioning, and discloses a cable branch box fault early warning and positioning system based on online partial discharge monitoring. Comprising a topology building module which is used for building a topology model in a cable branch box and defining a signal propagation path rule; pre-calculating signal expectation characteristics from each topological node to all monitoring sensors to form a structured expectation library; the acquisition and analysis module is used for acquiring multi-source transient signals and performing feature extraction to obtain an information fusion matrix; the decoupling positioning module is used for performing analysis based on the information fusion matrix and the structured expectation library to obtain a fault positioning point; the evaluation and early warning module is used for extracting all transient events in a preset time period corresponding to the fault positioning point, forming a time sequence feature sequence, extracting and analyzing an evolution index of the time sequence feature sequence, obtaining a risk transition probability, dividing risk levels and carrying out corresponding early warning; according to the invention, non-confusion accurate positioning of the partial discharge source of a specific joint or a cable section in the branch box is realized.

An emotional disorder recognition control model acquisition method and system based on near-infrared brain imaging map characteristics, a controller and a control method, and the method comprises the following steps: step 100, fNIRS-based multi-dimensional nerve marker acquisition and analysis, step 200, an integrated handheld diagnostic device, step 300, a dual-mode high-speed intelligent analysis system, step 300, a near-infrared brain imaging map characteristic analysis system, and step 300, a near-infrared brain imaging map characteristic analysis system. And step 400, a closed-loop AI training and database iteration system realizes a link of collection-analysis-diagnosis-display by fusing technologies of signal collection, data preprocessing and normalization, characteristic parameter extraction, intelligent report display and the like, so that the transformation of mood disorder screening from subjectivity to objectivity and from experience driving to data driving is promoted, and the mood disorder screening efficiency is improved. Therefore, the embedded analysis system for the rehabilitation medical scene is provided.

The invention relates to a coal industry chain operation plan assessment method and device, equipment and a storage medium. The method comprises the steps of obtaining to-be-assessed data; determining an evaluation index of the multi-dimensional evaluation index system; determining an evaluation period, a simulation scene and time constraints, space constraints, occurrence probability and influence intensity related to the simulation scene, and obtaining scene setting parameters; a preset objective weighting method is adopted to calculate the weight of each third-level index, and a TOPSIS multi-attribute decision model is combined to calculate and obtain a comprehensive evaluation result; performing problem analysis based on the comprehensive evaluation result, generating optimization suggestions corresponding to the problems, and obtaining a plan analysis result; and generating a target evaluation report. According to the integrated operation plan evaluation method provided by the invention, automatic acquisition and analysis of index data and report generation can be realized, automation and intelligence of an evaluation process are effectively realized, and evaluation efficiency, accuracy and systematicness of a coal industry chain operation plan are greatly improved.

The invention relates to the technical field of digital information monitoring, in particular to a network equipment operation monitoring log integration system based on multi-source data acquisition and analysis. Comprising a multi-source data acquisition unit; a pretreatment unit; the intelligent analysis unit is used for constructing a normal operation behavior baseline of the network equipment, identifying abnormal operation of the equipment and mining a log association relationship among multiple pieces of equipment; a log integration unit; and a visualization unit. According to the invention, an intelligent analysis unit adopts an attention mechanism enhanced time sequence modeling technology to construct an equipment normal operation behavior baseline, sub-fragments are divided according to equipment operation periodicity, and differentiated weights are given according to key index fluctuation characteristics, so that the baseline is more focused on a time period with significant influence on the equipment state; and meanwhile, an improved association mining algorithm is combined to introduce an equipment topology association degree pruning low association item set, so that the rationality of a behavior baseline and the comprehensiveness of multi-equipment log association analysis are improved.

The invention belongs to the field of information network security. The invention specifically discloses an agent-based logic vulnerability detection method, system and device, and a medium. The method comprises the following steps: capturing service system request response data through network traffic, identifying an HTTP protocol through a port and protocol features, filtering non-HTTP request data, extracting and analyzing the request data, standardizing extracted parameters, continuously collecting and analyzing through an API analysis agent, improving the semantic analysis capability of a large model, perfecting a logic relation graph, and improving the semantic analysis efficiency of the large model. A complete vulnerability life cycle data link is constructed through a vulnerability analysis agent, the system logic vulnerability mining efficiency and coverage are improved, and automatic logic vulnerability alarm and disposal can be realized and the risk disposal real-time performance is improved by accessing the agent to a network security protection product, an alarm analysis platform and other systems.

The invention provides an employee work assessment recommendation method and system based on a collaborative filtering algorithm, and relates to the technical field of employee work assessment, and the method comprises the steps: extracting a task dependency relationship and time consumption data from a project management system, and constructing an employee task dependency graph; secondly, collecting and analyzing work interaction information containing document operation and task marking features, and identifying task switching critical points to generate quantitative indexes; constructing a similar employee group and extracting an efficient working mode by means of a collaborative filtering algorithm after establishing an evaluation data set through space-time alignment of the two; by taking the task dependency graph as an environment state and the efficient mode as a reference, generating a task sequence optimization scheme by using a reinforcement learning model; according to the scheme, the staff group composition is adjusted according to the scheme execution effect, the optimization model is fed back, finally, a comprehensive evaluation result is generated based on the scheme and actual task sequence matching degree and the quantitative index stability, the task sequence optimization and physiological state adjustment ability can be accurately evaluated, and the work assessment scientificity and the task execution efficiency are improved.

The invention relates to the technical field of agricultural data visualization, in particular to a salinized cotton field visual multi-point salinity collection and analysis method and system.The method comprises the following steps that soil conductivity distribution data of a salinized cotton field area is obtained, a high-salinity accumulation area and a low-salinity dilution area are recognized, particle gap distribution characteristics are collected, and a salinized cotton field is obtained; and screening areas with loose or compact particle structures, executing temperature gradient analysis, screening surface thermal environment abnormal subareas, detecting wind speed distribution, removing wind field disturbance areas, and obtaining a recommended planting area list. According to the method, high and low salt areas are accurately identified through salt space distribution analysis, distribution precision is improved, humidity dynamic expression is refined, structural abnormal blocks are screened by using a particle gap change trend, soil management is optimized, surface temperature and wind speed data analysis is superposed, thermal disturbance and wind field abnormity are eliminated, and agricultural management precision is improved through multi-dimensional data fusion. The capability of coping with a complex environment is enhanced, efficient decision making is effectively supported, and the overall agricultural production benefit is improved.

The invention relates to the technical field of tunnel settlement data acquisition and analysis, in particular to a tunnel settlement data acquisition method and system based on laser ranging, and the method comprises the following steps: S1, equipment is a monitoring device comprising a holder and two laser ranging sensors with a fixed included angle, the first sensor horizontally points to the side wall of a tunnel, and the second sensor horizontally points to the side wall of the tunnel; the second sensor points to the tunnel vault at an included angle theta; s2, point distribution and base point selection, wherein monitoring points are distributed in the tunnel according to a preset rule, and a zero reference point is arranged in a stable area of a tunnel portal; in the scheme, through integration of the double laser sensors and automatic holder scanning, high-density and high-precision automatic continuous data acquisition of the tunnel section is realized, and the defects of low efficiency and insufficient coverage of a traditional method are overcome; meanwhile, by means of intelligent coordinate transformation, a deformation calculation model and a multi-stage early warning mechanism, real-time calculation, trend prediction and graded automatic alarm of settlement and convergence parameters are achieved.

The invention discloses an acoustic stimulation nerve regulation and control service management system combining AI and big data analysis, and the system comprises a cloud center layer, a region edge layer, a doctor management layer and a patient terminal layer which are connected with one another, the cloud is provided with a client management module, and an exclusive account is established for each patient; and an individualized sound stimulation scheme is generated and dynamically optimized by using an AI algorithm to realize remote closed-loop regulation and control. The platform supports remote demonstration and teaching of experts, autonomous operation and maintenance of clients and a mutual assistance network of the experts, and a cloud-side-end collaborative AIoT intelligent system is formed. The system is provided with a national-level data supervision interface, so that safety and controllability are ensured. Through the above mode, the AI and big data analysis combined acoustic stimulation nerve regulation and control service management system improves the accuracy and efficiency of personalized scheme formulation and optimization by constructing a collection-analysis-decision-execution-feedback-optimization closed-loop process; therefore, cloud-edge-end-human collaborative closed-loop intelligent regulation and control management with individualized nerve regulation and control and compliance safety considered is formed.

The invention discloses an electrophoresis parameter management system and method based on multi-data analysis, and relates to the technical field of data analysis, and the management method comprises the following steps: carrying out the collection and analysis of each electrophoresis parameter of each electrophoretic coating process, and carrying out the evaluation of the quality of a generated coating; identifying a parameter group corresponding to the generated coating based on an evaluation result; analyzing the difference condition of any electrophoresis parameter based on the evaluation result difference; generating a corresponding parameter range for any electrophoresis parameter; identifying a mutual influence relationship among the electrophoresis parameters; analyzing the influence condition of any parameter group on the evaluation result to obtain the influence value of each electrophoresis parameter; performing quality evaluation on the current electrophoretic coating based on the electrophoresis parameters collected in real time; analyzing and judging the rationality of each electrophoresis parameter, and judging the risk according to the influence value of the electrophoresis parameter; the limitation of dependence on manual experience in the prior art is overcome, and scientificity and accuracy are improved.

The invention relates to the technical field of scientific and technological information processing, and discloses a scientific and technological information distributed acquisition and analysis system. A distributed acquisition unit of the system acquires a real-time science and technology information data set containing timestamps, spatial position identifiers, resource state indexes and external environment parameters from a plurality of heterogeneous data sources; the dynamic analysis network unit performs multi-dimensional alignment processing on the data set to form a space-time associated data cube, performs feature decoupling operation, and separates out a core feature matrix and a deviation feature matrix; the strategy generation unit generates a distributed resource collaborative scheduling strategy containing a resource allocation sequence and a task execution node mapping relation based on the core feature matrix and a predefined resource allocation rule; and the adaptive adjustment unit analyzes a real-time operation abnormal event indicated by the deviation characteristic matrix, and dynamically corrects a scheduling strategy to generate an optimized resource collaborative scheduling strategy. The system realizes efficient collection and accurate analysis of scientific and technological information and dynamic optimization of a scheduling strategy, and adapts to complex scientific and technological information processing requirements.

The invention discloses a freight environment state data acquisition and analysis method and system based on star-flash communication, and belongs to the technical field of freight detection, and the method comprises the following steps: a cab data acquisition module acquires vehicle state data and road condition environment data of a configured freight tool in a freight process in real time; receiving the state data of a plurality of containers in the freight vehicle acquired by the container data acquisition module and transmitted based on a star flash wireless communication interaction mode, wherein each piece of container state data comprises temperature and humidity, impact and vibration data; and sending the vehicle state data, the road condition environment data and the plurality of cargo box state data to a terminal data processing center, so that the terminal data processing center sends the vehicle state data, the road condition environment data and the plurality of cargo box state data according to the vehicle state data, the road condition environment data and the plurality of cargo box state data in a preset freight work time period. And carrying out judgment and analysis on the transportation safety risk state of each container in the freight transportation process.

This invention provides an analysis system, analysis program, and analysis method that enable more accurate evaluation of target data. [Solution] In the analysis system 0, the analysis device 1 includes an acquisition unit that acquires an analysis index that serves as an indicator in the analysis of the target of analysis and raw data of the target of analysis for the analysis index; a parameter estimation unit that estimates the parameters for each candidate distribution model of the distribution model of the data distribution of multiple targets of analysis based on the acquired data; a goodness of fit calculation unit that calculates the distribution goodness of fit for each estimated candidate distribution model of the data distribution of the target of analysis; a distribution identification unit that identifies a fitted distribution model that fits the data distribution of the target of analysis based on the distribution goodness of fit; a deviation calculation unit that calculates a theoretical value for the analysis index of the target of analysis based on the parameters of the identified fitted distribution model and calculates the deviation between the theoretical value and the actual value of the analysis index of the target of analysis included in the raw data of analysis; and an evaluation unit that evaluates the target of analysis based on the deviation.

The invention relates to an intelligent monitoring and risk early warning system for cardiovascular acute and critical diseases, which is characterized in that electrocardiosignals are acquired in real time through a wearable twelve-lead ECG monitor (ADS1298 chip, 512Hz sampling), and the electrocardiosignals are encrypted and transmitted to a cloud end through an MQTT protocol in combination with edge end wavelet denoising and QRS wave detection. And the cloud end adopts a multi-modal fusion engine, integrates a random forest-XGBoost hybrid model, performs modeling based on a postoperative patient physiological index time sequence, predicts the risk through Kalman filtering and triggers secondary early warning. Clinical tests in a third-A hospital prove that the end-to-end response time is 2.8 seconds, the D2B time is shortened to 70 minutes, and the occurrence rate of main adverse cardiovascular events in 30 days after operation is reduced by 47%. According to the invention, the limitation of traditional single-mode monitoring is broken through, a whole-process closed loop of collection-analysis-intervention is constructed, and an intelligent solution is provided for prevention and control of cardiovascular acute and critical diseases.

This invention discloses an automatic collection and analysis method for medical quality control indicators in critical care departments. The method includes: establishing a case feature database based on historical monitoring data; constructing and training a quality control indicator model based on the case feature database; establishing a model update database to store newly input cases; periodically inputting data from the model update database into the quality control indicator model; and determining whether the medical quality control indicators in critical care departments are within the acceptable range based on the output of the quality control indicator model display module and providing feedback to relevant personnel.

The application provides an online monitoring device simulation method of power primary equipment, comprising: constructing an online monitoring device model containing an application layer module, the module being composed of a state information acquisition, analysis and transmission submodule, and each submodule realizing data interaction through an event-driven shared memory; during simulation, the state information acquisition submodule analyzes a monitoring data file, stores state data to the shared memory and triggers an event queue 1; the state analysis submodule acquires data according to the event queue 1, generates a state file through an auxiliary diagnosis algorithm and triggers an event queue 2; the state transmission submodule responds to the event queues 1 and 2, converts state data and files into a unified format communication message and transmits the message to a station control layer device model in a server mode. The application adopts a seven-layer network structured simulation model, realizes real-time acquisition, analysis and transmission of device states, solves the response lag problem of a traditional monitoring mode and provides support for performance verification of an online monitoring device.

The utility model discloses a kind of multidimensional data acquisition and analysis device based on street dance competition judgment, including mounting bracket, and the fixed head is fixedly connected with mounting bracket bottom end, and the movable head is rotatably connected in fixed head, and the collection analysis equipment is fixedly connected with movable head bottom end;Still including adjusting mechanism, it is set on movable head, and adjusting mechanism includes convex rod, and the second inner groove of a plurality of settings around movable head, when movable head is active in fixed head, and convex rod is reinforced to movable head angle.The utility model is cooperated through convex rod, spring, extruding rod and fixed rod, realize the accurate adjustment and reinforcement locking of movable head angle, overcome the problem that existing technology universal adjusting structure is easy to wear, contact skid and in the collection process because of structure skid lead to capture deviation inaccuracy.

The invention relates to the field of industrial production safety, in particular to a hidden danger investigation and risk prediction management and control method based on industrial production safety. Hidden danger gas components and hidden danger monitoring points are determined through collection and analysis, and a suspected leakage area set where equipment is located is matched according to the hidden danger gas components; combining the concentration of each hidden danger gas component of each hidden danger monitoring point with a workshop airflow field distribution model, and analyzing to obtain a potential leakage area set; analyzing the distribution condition of each potential leakage area and the suspected leakage area set, and judging a final leakage area; and converting the calculated final leakage area hidden danger gas concentration into a concentration risk characteristic score, carrying out weighted calculation on the concentration risk characteristic score and the personnel distribution characteristic score to obtain a comprehensive risk score, and judging a risk grade according to the score. The method is advantaged in that the final leakage area is accurately positioned, through multi-dimensional quantitative evaluation, evaluation reliability is improved, risk levels in different scenes are accurately reflected, and reasonable formulation and effective execution of subsequent targeted measures are improved.

The invention provides an AI-based intelligent whole-house temperature control method and system, and relates to the technical field of temperature control, and the method comprises the steps: obtaining regional physical data of a plurality of whole-house regional data of whole-house physical data, and generating a reference temperature control instruction for the regional physical data; carrying out regional environment acquisition and analysis on the whole house regional data, determining a temperature rise factor and a temperature drop factor which cause the temperature to deviate from the standard according to the acquired and analyzed data, and carrying out data fusion to obtain multi-factor fusion data; performing positive and negative influence analysis based on an AI model according to the multi-factor fusion data, performing temperature control offset analysis according to the positive and negative influence analysis data, generating regional temperature control adjustment data according to the temperature control offset analysis data, performing temperature control adjustment on the whole house regional data, and obtaining temperature optimization control data. According to the invention, global sensing, change prediction and collaborative optimization intelligent temperature control can be realized, so that the problem of thermal interference between regions can be solved fundamentally.

The invention relates to the technical field of ceramic tile glue detection, and provides a ceramic tile glue performance testing device which comprises a display stand, a water soaking device, an environment simulation device, an anti-sliding strength detection device, a drawing strength detection device, a carrying robot, a carrying guide rail, a control center and a display device. The carrying robot is arranged on the carrying track and is used for carrying the to-be-tested ceramic tiles; the water soaking device, the environment simulation device, the anti-sliding strength detection device and the drawing strength detection device are in data signal connection with the control center, the control center collects and analyzes data, obtains an analysis result and sends the analysis result to the display device, and the display device is used for displaying the analysis result; when the ceramic tile adhesive performance is displayed, the ceramic tile adhesive performance is displayed through the processes of water soaking treatment, environment simulation, strength detection and the like of ceramic tiles, observation is more visual and real, the trust degree of people on the ceramic tile adhesive effect can be improved, and popularization of ceramic tile adhesive products is facilitated.

This invention relates to the fields of medical health and artificial intelligence technology, and particularly to a nasopharyngeal carcinoma patient management robot based on proactive health management and user profiling. It includes a main control module, a hardware layer, and a software functional layer. The hardware layer includes a screen interaction module, a voice interaction module, a data acquisition module, and a communication module for human-computer interaction, user data acquisition, and data communication. The software functional layer includes a user profile construction module, a multimodal emotion recognition module, an evidence-based response module, a multimodal temporal symptom proactive prediction module, a rehabilitation training module, and a diagnosis and treatment assistance module for nasopharyngeal carcinoma patient user profile construction, emotional interaction, evidence-based guidance, complication risk prediction, rehabilitation training, and diagnosis and treatment assistance. The storage module pre-stores a nasopharyngeal carcinoma-specific evidence-based knowledge base. This invention achieves precise symptom management for each nasopharyngeal carcinoma patient, completing a closed-loop management system encompassing multimodal empathic interaction, full-cycle coverage, data collection, analysis, intervention, and iteration.

The invention discloses a multi-technology fusion monitoring and early warning system for elevator safety, which belongs to the technical field of elevator safety monitoring and comprises a historical data analysis module, a knowledge graph construction module, a maintenance plan and early warning module and a visual data output module. The historical data analysis module is used for collecting and analyzing historical operation data and fault data, obtaining health decline curves of all parts of the elevator under different use frequencies, load modes and unstable electric power environments, and constructing a prediction model; through multi-module cooperation and multi-technology fusion, accurate early warning, associated diagnosis and personalized maintenance of elevator faults are achieved, manual dependence and misjudgment are reduced, workers can conveniently master the safety state of the elevator in real time, the problems that traditional elevator maintenance is insufficient in pertinence, fault response lags behind, data integration is difficult and the like are effectively solved, and the maintenance efficiency is improved. And the elevator safety management efficiency and reliability are improved.

The invention belongs to the technical field of intelligent medical control, and particularly relates to a hysteroscope surgical robot control method and system combining visual servo and remote cooperative operation. According to the method, time difference, spatial difference and image error between input of an operation terminal and response of a robot are analyzed through multi-mode information collection, a deviation index set is constructed, an operation mode is judged according to the deviation index set, real-time switching among an autonomous mode, a cooperative mode and a guiding mode is achieved, and a corrected or amplitude-limited control instruction is output. The system further combines image features and tactile feedback to realize real-time adjustment of the position of the instrument, and generates an image prompt layer according to the risk level and outputs visual warning information. And under the condition that the error continuously exceeds the limit, the system can automatically execute forced amplitude limiting and area locking, so that the operation safety is improved. By constructing a closed-loop control link, high-precision collaboration and safe intervention in remote minimally invasive operation are realized, and the method has a good clinical application prospect.

The invention discloses a slope reinforcement full life cycle management system and method based on digital twinning, and the system comprises six cooperation modules, and achieves the real-time interaction and deep fusion of geological parameters, stress field distribution, sliding surface features and other parameters through a data bus. On the basis of anti-sliding stability evolution prediction, three-dimensional stress field multi-scale interpolation and critical sliding surface multi-objective optimization search core technologies, a support effect quantitative evaluation platform and a full-life-cycle dynamic management mechanism are combined, and a closed-loop system from parameter acquisition, analysis and prediction to evaluation and regulation is constructed. According to the method, through multi-module cooperative linkage, the slope state is accurately represented, the critical sliding surface is positioned, the supporting effect is quantified, and dynamic regulation and control are conducted according to the evolution law of all stages of the whole life cycle. According to the system and the method, the precision of slope stability analysis, sliding surface positioning and support evaluation is improved, intelligent and dynamic management and control of the whole process are realized, and the geological disaster risk is effectively reduced.

This invention provides a slope scour simulation test device based on soil particle staining tracers, comprising: an adjustable slope scour section connected downstream of a water supply unit, including a slope flume and a slope adjustment component at its bottom; the slope flume contains, from bottom to top, a concrete layer, transparent test particles, and tracer particles; and a data acquisition and analysis unit for acquiring test data under different working conditions, including test condition data and image sequences of the adjustable slope scour section, identifying and tracking tracer particles of different sizes in the image sequences, and outputting the motion parameters and initiation parameters of the tracer particles. By combining the test condition data under multiple working conditions and the corresponding motion and initiation parameters of the tracer particles, the device quantitatively characterizes the initiation, migration, and deposition mechanism of slope particles under runoff-seepage coupling. This invention can simulate slope scour under runoff-seepage coupling conditions, accurately recording and analyzing the motion behavior of slope particles under different working conditions.

The invention, which relates to the optical cable management field, discloses an optical cable full-life-cycle statistical contrastive analysis system comprising a data acquisition layer, an intelligent analysis layer, an application decision-making layer, a full-process data management module and an intelligent deployment module which perform data interaction in sequence. According to the system, an integrated closed loop of collection-analysis-management-decision-allocation is constructed, the data collection layer guarantees full-amount accurate perception and real-time synchronization, the intelligent analysis layer achieves accurate fault pre-judgment and efficient first-aid repair by means of a CNN-LSTM model and a GIS + improved A * algorithm, information islands are broken through through full-process data management, and the real-time performance of the system is improved. And the application decision and intelligent deployment module realizes full-chain tracing and efficient resource utilization. According to the invention, data driving is used to replace artificial experience, fault pre-warning, meter-scale positioning and optimal resource configuration are realized, operation and maintenance cost and resource waste are reduced to the maximum extent, and optical cable value and management efficiency are comprehensively improved.

A method, device, medium and electronic equipment for implementing traffic collection analysis are disclosed, wherein the method comprises: performing common subexpression extraction processing on each regular rule in a regular rule set to obtain one or more common subexpressions; obtaining to-be-processed data packet information that needs to be subjected to traffic collection analysis; in a process of performing regular rule matching calculation on the to-be-processed data packet information based on each regular rule in the regular rule set, judging whether a current to-be-matched subexpression is a common subexpression; and in a case where it is judged that the current to-be-matched subexpression is a common subexpression and a matching calculation result of the common subexpression is currently stored, obtaining the matching calculation result of the current to-be-matched subexpression from the stored matching calculation result. The present disclosure is conducive to guaranteeing the integrity of traffic collection analysis, the real-time performance of attack interception and the robustness of the system in a high-concurrency scenario.

The invention discloses a graphic processing unit performance acquisition and analysis method, belongs to the field of data processing, and is used for quickly acquiring and analyzing performance information of each graphic processing unit during model heterogeneous mixed training. The method comprises the following steps: acquiring first model information of a plurality of first graphic processing units (GPUs) in each first computing power node, wherein the first computing power nodes are computing power nodes during model heterogeneous mixed training; calling an information acquisition interface corresponding to the first model information, and obtaining a first performance parameter of each first GPU and first event information generated during operation; determining node performance information of each first computing power node based on the first performance parameters and the first event information so as to determine overall performance information of a plurality of computing power nodes during model heterogeneous mixed training; and displaying the overall performance information in a preset interface to enable a user to perform performance analysis on each first GPU during model heterogeneous mixed training.

The invention discloses a visual interface adaptive adjustment system based on user behavior analysis, and relates to the technical field of behavior analysis. The user behavior data acquisition module is used for acquiring multi-dimensional interaction data and encrypting and transmitting the multi-dimensional interaction data; the cognitive calculation engine module is used for mining behavior patterns and intentions based on deep learning; the environment perception and situation modeling module is used for perceiving an environment and a user state in real time; the self-adaptive interface generation module is used for dynamically adjusting layout, style and interaction; and the intelligent decision-making and continuous optimization module realizes intelligent decision-making through reinforcement learning. According to the system, user behaviors are accurately collected and analyzed, environment changes are sensed, and the visual interface layout and interaction are dynamically optimized; the task time can be remarkably shortened, the error rate is reduced, the user satisfaction degree, the use duration and the service conversion rate are improved, the user stickiness is enhanced, personalized experience is provided for users, and benefits are created for enterprises.