679 results about "Data cloud" patented technology

The invention discloses an intelligent self-adaptive industrial robot control system and method, and the method comprises the steps: carrying out the automatic recognition and self-inspection of a sensor, an actuator and a controller module through a standardized interface when the system is started; various sensors are used for environment monitoring, acquired data are stored in a cloud end, and a sensor fusion technology is applied for integration; the system state and the environment condition are monitored in real time, machine learning and an artificial intelligence algorithm are used for analyzing the operation state and the task effect, and a corresponding operation strategy is formulated; based on task execution and environment information, an execution path is generated by adopting a path planning algorithm, and an operation strategy and parameters are dynamically adjusted according to real-time feedback; a real-time safety monitoring mechanism is introduced, and a safety threshold value is set; sensor data are analyzed regularly, potential faults are identified, and fault prediction is carried out; according to the intelligent self-adaptive industrial robot control system and method, when the robot is designed, the safety of man-machine cooperation is considered, and safety protection measures are added.

The invention discloses a battery full life cycle intelligent management method and system based on a large model, and the method comprises the following steps: S1, collecting the operation data of a battery, and carrying out the preprocessing; s2, inputting the pre-processed sample into a pre-trained Transform encoder model, and extracting a time sequence and cross-stage characteristics; s3, executing model reasoning, outputting a state index, and generating a battery state vector; s4, identifying an abnormal category and a position in combination with the working condition information; s5, generating a dynamically optimized battery management strategy based on the battery state index and the abnormity identification result; s6, deploying a lightweight model at the edge device, executing local reasoning and uploading data; s7, the cloud updates the model through self-supervised training and issues the model; and S8, repeatedly executing the steps S1 to S7, and carrying out optimized closed-loop management. According to the invention, large model modeling and an edge cloud cooperation mechanism are fused, and intelligent sensing and dynamic management of the whole life cycle of the battery are realized.

The invention discloses an AI auxiliary limb rehabilitation system, and belongs to the field of communication control systems, the AI auxiliary limb rehabilitation system combines a sensing module, an AI analysis module, a VR interaction module, a feedback module and the like, comprehensively collects limb movement and physiological data, introduces gamification elements, improves the patient participation degree through target setting, an award mechanism and interesting tasks, and improves the rehabilitation effect. A personalized rehabilitation plan is generated by using a deep learning algorithm, an emotion recognition unit can adjust the plan according to a psychological state, environment self-adaption and real-time adjustment of training difficulty to ensure training safety and is challenging, data cloud storage facilitates remote monitoring, third-party equipment is compatible with extended monitoring ability, and the training efficiency is improved. A brand-new, efficient and personalized solution system scheme is provided for limb rehabilitation.

The invention provides a mutual inductor test data cloud edge cooperative processing method and device, and relates to the technical field of data processing, and the method comprises the steps: carrying out the time domain and frequency domain combined feature extraction of the original measurement data of a mutual inductor test through a multi-mode decoupling preprocessing model, and forming a data feature vector set, further generating a confidence label flow through state recognition and Bayesian inference, evaluating a prediction error, and realizing data classification screening and priority queue construction; the data are uploaded to a cloud end in a semantic compression and multi-resolution representation vector mode, so that the transmission efficiency is improved; carrying out deep modeling and model performance monitoring at the cloud, and if degradation is detected, returning edge original data to update the model; and finally, generating a scheduling weight factor according to the prediction error distribution graph, and dynamically optimizing an edge cloud task allocation proportion. According to the invention, the problem of low resource allocation efficiency caused by lack of a dynamic scheduling mechanism based on prediction errors and confidence driving in existing mutual inductor test data edge cloud cooperative processing can be solved.

The invention relates to the technical field of network security, in particular to a network security situation real-time monitoring method and system. The method comprises the following steps: acquiring virtual machine running state data and physical host deployment data, and performing distributed probe deployment in a cloud computing environment according to the virtual machine running state data and the physical host deployment data to obtain monitoring probe data, wherein the monitoring probe data comprises lightweight virtual machine probe data, physical host monitoring probe data and network flow probe data; according to the monitoring probe data, performing global identification mapping on the corresponding relationship between the virtual machine and the physical host so as to obtain identification mapping data; obtaining tenant security data; and performing streaming hierarchical processing on the tenant security data according to the identifier mapping data to obtain tenant isolation data. Through the design of the distributed probe, a virtual machine, a physical host and network flow are covered, comprehensive data can be collected in real time, and a monitoring blind area is eliminated.

The invention discloses a laser radar system for simultaneously measuring an atmospheric cloud height and a wind field. The laser radar system comprises an outer cover module and an inner cavity, the outer cover module comprises an automatic detection station shell, an all-in-one meteorological sensor connected with the automatic detection station shell, a cloud measurement window embedded in the automatic detection station shell and a wind measurement window embedded in the automatic detection station shell; a reference module, a control module, a cloud measurement optical receiving module, a cloud measurement laser transmitting module and a wind measurement optical transceiving module are arranged in the inner cavity; wherein the control module, the cloud measurement optical receiving module, the cloud measurement laser transmitting module and the wind measurement optical transceiving module are all integrated on the reference module; through the above structure, the laser radar system can obtain the wind profile data, the cloud height measurement data, the atmospheric extinction coefficient profile and the visibility of the current environment at the same time.

The invention discloses a distributed photovoltaic short-time output dynamic prediction method and system. The method comprises the following steps: acquiring micro-meteorological data, a cloud picture image and historical output data of a target area; calculating a building shadow shielding coefficient, and generating shadow influence factor time sequence data; obtaining cloud image occlusion index time series data based on the cloud image and a pre-constructed cloud image occlusion prediction model; fusing the data, and constructing and training a short-time output dynamic prediction model based on a long short-term memory network; predicting photovoltaic output by using real-time data and the trained model, and adjusting the model according to a prediction error and a confidence interval; according to the method provided by the invention, various factors such as microclimate, building shadow and cloud layer shielding are comprehensively considered, and the complex nonlinear relationship between the factors and the photovoltaic output is captured by using the deep learning model, so that the accuracy and robustness of distributed photovoltaic short-time output prediction are improved.

The invention relates to the technical field of medicine cold-chain logistics, and discloses a medicine cold-chain logistics dynamic data cloud platform and a route analysis method.The cloud platform comprises a data acquisition unit, a preprocessing unit, a data transmission unit, a storage management unit and a route analysis unit. Key data such as temperature, humidity and position are collected in real time through multiple sensors and are subjected to standardized preprocessing, so that the data are comprehensive, real-time and reliable, and a foundation is laid for subsequent analysis. In the aspect of storage management, partition storage is carried out based on time and space dimensions, and the data retrieval and management efficiency is greatly improved; in the aspect of route planning, factors such as time, temperature and humidity, congestion and the like are comprehensively considered, a scientific route cost evaluation function is constructed, costs such as transportation duration, temperature and humidity stability, road conditions and the like are covered, and in combination with a greedy algorithm, the medicine transportation quality can be guaranteed, the cost can be effectively reduced, and the logistics efficiency can be improved.

The invention relates to the technical field of medical rehabilitation monitoring, in particular to a rehabilitation cloud platform system for cardiopulmonary function monitoring. According to the system, an intelligent monitoring terminal collects physiological data; the cloud computing platform adopts wavelet threshold denoising and an index moving average method to remove noise, combines dynamic time warping to achieve multi-physiological signal time sequence alignment, constructs a double-flow model, generates a health state index and a comprehensive health risk index, introduces a multi-scale branch and differential attention mechanism, and achieves the multi-scale health state index and the comprehensive health risk index. High-precision anomaly detection is realized through prediction deviation analysis, multi-target reinforcement learning is adopted to balance health improvement and risk control, and a personalized training scheme is generated; the remote medical support module supports a doctor to monitor patient data in real time and intervene in an adjustment scheme; the user interaction end provides a real-time monitoring data visualization and individuation rehabilitation plan. Through multi-modal data fusion, time sequence synchronous optimization and adaptive reinforcement learning, accurate monitoring, dynamic rehabilitation optimization and remote medical collaboration are realized.

The invention discloses an encryption protection method for network data backup and recovery, which relates to the technical field of data protection, and comprises the following steps: respectively implementing differential backup strategies by adopting a local, cloud and remote three-terminal collaborative backup architecture; and performing local data backup by a combined scheme of complete backup and incremental backup to generate a chained hash value sequence. Three-terminal key isolation and dynamic derivation are adopted to guarantee data security, an independent key system is adopted locally, cloud and remotely, a chain type key is generated locally based on a hash chain, version independent keys are generated at the cloud and remotely, and data security of other terminals is not affected by key leakage of any storage terminal. The recovery priority is automatically switched from local to cloud to remote, the local decrypts incremental data layer by layer through a hash chain, the cloud directly calls an independent key of a target version and uses a synthetic backup complete copy in remote, and the disaster recovery efficiency is improved through layered recovery and synthetic backup.

The invention discloses a cache data security protection system based on cloud computing, particularly relates to the field of security protection, and comprises a data acquisition module, a cloud storage module, a verification execution module, a security processing module and a protection early warning module. The data acquisition module acquires cache data in the platform in real time, and the cloud storage module adopts a distributed architecture to safely store various types of data, and performs classified preprocessing and encrypted storage; the verification execution module identifies the safety signal, generates an execution signaling, and transmits the execution signaling to the execution node through the encryption channel; the security processing module is used for cleaning, desensitizing and encrypting input and recorded information and converting the input and recorded information into calculation processing data, and the protection early warning module is used for integrating a data source, constructing an analysis data set, detecting abnormal behaviors through an algorithm and a rule engine and generating a security signal; in the security comparison stage, the to-be-approved data is compared in real time, the operation risk is evaluated, the signal data is transmitted through an encryption channel, and a full-link log is stored in the block chain.

The invention discloses an intelligent fitness method and system. The intelligent bodybuilding method comprises the steps that the bodybuilding equipment receives a bodybuilding situation instruction and provides a corresponding motion mode for a user; the fitness equipment collects fitness data of the fitness equipment and feeds back the fitness data to the terminal; the terminal collects the fitness data and sends the fitness data to the cloud server; the cloud server processes the fitness data based on a pre-trained model and returns an interaction command to the terminal according to a processing result; and the terminal sends the interaction command returned by the cloud server to the fitness equipment, so that the fitness equipment updates the exercise mode of the fitness equipment. According to the technical scheme, efficient transmission and comprehensive analysis of the fitness data can be realized through interaction of the fitness equipment, the terminal and the cloud server, and a proper fitness situation instruction can be given by utilizing the pre-trained model in combination with the fitness condition of the user, so that the exercise mode is dynamically adjusted, and the user experience is improved. Personalized and intelligent fitness services and fitness guidance are provided for the user, and the fitness experience of the user is improved.

The invention relates to the technical field of lithium battery energy storage and temperature control, in particular to an electronic radiator intelligent temperature control system based on a lithium battery energy storage cabinet. The system comprises an edge temperature sensing module, a cloud thermal management strategy module, a cloud clustering analysis module, a thermal field topological optimization module, a multi-source thermal feature fusion module, a dynamic temperature control strategy generation module and a grading strategy issuing execution module. An edge temperature sensing module collects real-time temperature data of multiple monitoring points and an edge side preliminary temperature control scene judgment result; the cloud heat management strategy module receives battery pack topology connection data; the cloud clustering analysis module carries out fuzzy clustering on the data to obtain a temperature state and scene discrimination clustering center; the thermal field topological optimization module optimizes a temperature state clustering center in combination with topological data; the multi-source thermal feature fusion module fuses the two clustering centers to form multi-source fusion feature representation; the dynamic temperature control strategy generation module determines a temperature control strategy identifier accordingly, and the grading strategy issuing execution module matches and distributes a gradient temperature control strategy to a heat dissipation actuator.

The invention relates to the technical field of salt tide upstream prediction, in particular to a salt tide upstream prediction method and system based on artificial intelligence. The method comprises the following steps: performing river upstream and downstream fall correlation analysis on an estuary to obtain width change fall correlation data; non-flood season and flood season precipitation extraction is carried out on the historical environment data of the estuary, and downstream runoff change analysis is carried out on the width change fall associated data to obtain non-flood season / flood season downstream runoff change data; according to the non-flood season / flood season downstream runoff change data, carrying out tide superimposed effect analysis and non-flood season / flood season salt tide upstream action analysis to obtain non-flood season / flood season salt tide upstream action data; and based on a recurrent neural network algorithm, constructing a salt tide upstream prediction model for the non-flood season / flood season salt tide upstream action data, and sending the model to a data cloud platform to execute salt tide upstream prediction. According to the method, the salt tide upstream prediction technology is optimized, so that the salt tide upstream prediction technology is more accurate.

The invention relates to the technical field of power system safety monitoring, in particular to an electricity utilization potential safety hazard supervision method and system based on an electricity utilization data cloud platform. Multi-dimensional electricity utilization data are collected in real time and preprocessed, and the preprocessed data are subjected to multi-source fusion in combination with a data fusion algorithm; carrying out hidden danger detection on the fused data by utilizing a multi-layer intelligent learning model; a dynamic scoring mechanism is introduced, multi-dimensional scoring is performed on a detection result, hidden danger risk levels are divided, a multi-level early warning mechanism is triggered based on the risk levels, and automatic emergency response is implemented; potential electricity utilization safety hazards can be recognized in time, early warning is achieved, and lagged detection and processing of a traditional method are avoided. By means of intelligent hidden danger detection and dynamic risk assessment, the monitoring accuracy and efficiency are improved, human intervention is reduced, and the timeliness of emergency response is improved.

The invention provides a cloud control robot method and system, and relates to an intelligent control technology, a robot terminal obtains environment data and object state information based on a sensor module, and uploads the environment data and the object state information to a cloud server based on a communication module, and the cloud server determines an object specification according to the object state information and sends the object specification to a cloud server. The cloud server determines a constraint road section according to the environmental data, calculates a track offset value according to the object specification and the road section parameters, generates a control instruction according to the grabbing point and the track offset value, sends the control instruction to the robot terminal, and obtains real-time state data when the robot terminal executes the control instruction; and the cloud server generates action adjustment information based on the real-time state data to perform instruction updating on the robot terminal, so that more efficient and more flexible object carrying control on the robot can be realized.

The invention discloses a camera monitoring system based on wireless communication and a communication protocol optimization method, and relates to the technical field of communication optimization, the camera monitoring system is composed of a monitoring camera group, a wireless communication network, an edge computing unit, a cloud data center and a user interaction terminal; the monitoring camera group simulates human eye vision, adjusts a visual angle and a resolution ratio according to needs, and collects data; a wireless communication network is combined with quantum encryption and a traditional technology, and a communication mode is intelligently selected; the edge calculation unit processes data with low power consumption by using a neuromorphic chip; the cloud data center fuses the block chain and federal learning analysis data; the user interaction terminal has holographic projection and emotion perception functions and provides personalized services. According to the invention, the modules are collaboratively innovated, the pertinence of collected data is strong, the communication is efficient and safe, the processing is fast and accurate, and the interaction experience is good; the system has the functions of intelligent analysis, energy-saving management and the like, the intelligence and reliability of the system are improved, and the system has wide application prospects and market value.

The invention relates to a distributed fault positioning system and method for a power transmission and distribution line, and the system comprises the power transmission and distribution line, an acquisition unit group, a collection unit, a communication base station, a cloud server, and a data terminal. The acquisition unit group is used for monitoring current, self state parameters and the like of the power transmission and distribution line, the collection unit is used for receiving and sending communication data with the acquisition unit group, the communication base station is used for receiving and sending communication data with the collection unit, and the cloud server is used for communicating with the communication base station and the data terminal. And the data terminal is equipment with a communication function at the user side. According to the scheme, the application range of a line fault positioning system is expanded, including mountainous areas without mobile communication network coverage, a mode of collecting and installing equipment distributed calculation is adopted, the accuracy and efficiency of fault positioning are improved, the line patrol time is saved, the time of line power restoration is shortened, and the reliability of a power supply system is effectively ensured.

The invention relates to the technical field of intelligent traffic systems, in particular to a linear intelligent monitoring system for a bridge fabrication machine, which comprises a multi-source sensing module, a data processing module, a data processing module, a data processing module, a data processing module, a data processing module and a data processing module, and is characterized in that the multi-source sensing module comprises strain sensors, inclinometers, GPS positioning units and machine vision units; the edge computing node is internally provided with a self-adaptive data fusion algorithm and is used for denoising sensor data in real time; the cloud digital twin platform is used for synchronizing the BIM design model and generating a millimeter-level linear deviation early warning signal; the PLC control interface is used for receiving the early warning signal and automatically adjusting parameters of the template system and the walking mechanism; according to the method, environmental noise interference is effectively eliminated through multi-source sensor fusion and a self-adaptive data filtering technology, the monitoring precision is improved to a millimeter level, real-time comparison of construction data and a design model is realized based on a BIM and digital twinning dynamic interaction system, and the timeliness and accuracy of linear control are remarkably improved.

The invention belongs to the technical field of floor support plate construction, and particularly discloses a disassembly-free bottom die steel bar truss floor support plate construction system based on an intelligent construction site, which comprises a BIM (Building Information Modeling) model driving module, a construction progress control module, an Internet of Things sensing module, a visual detection module, an evidence storage module and a cloud platform, the BIM model driving module constructs a three-dimensional model and maps the three-dimensional model to field positioning; the construction progress control module dynamically simulates the process through four-dimensional coupling; the Internet of Things sensing module collects welding current, stress and temperature and humidity data in real time and carries out early warning. The visual inspection module inspects quality parameters such as seam splicing and lap joint based on unmanned aerial vehicle inspection; the evidence storage module encrypts and stores construction data; and the cloud platform integrates the whole-process instruction. Through cooperation of multiple modules, high-precision positioning, process dynamic simulation and data full-process management of floor support plate construction are realized, and the construction precision and the progress management and control efficiency are remarkably improved.

The invention belongs to the technical field of intelligent retrieval, and discloses an archive data intelligent retrieval method and system based on a multi-modal large model. The method comprises the following steps that a cloud data center builds an archive data intelligent retrieval platform, deploys a block chain network and builds an archive data intelligent retrieval engine; the cloud data center is used for carrying out data processing on the real-time multi-modal file big data; the cloud data center is used for generating a knowledge graph according to a plurality of real-time retrieval themes; the cloud data center is used for carrying out data clustering; the cloud data center uses a block chain network to carry out distributed storage; the user terminal is used for uploading real-time query data; the cloud data center is used for performing intelligent retrieval by using an archive data intelligent retrieval engine; and the cloud data center visually displays the target archive data. According to the method, the problems of low retrieval efficiency, insufficient accuracy, weak multi-modal data processing capability and insufficient security in the prior art are solved.

The invention relates to the technical field of quality management, in particular to a concrete building quality monitoring system and method. The problems that in the prior art, concrete building quality monitoring mainly depends on manual sampling inspection and regular inspection modes, and potential quality hazards in a concrete building structure are difficult to find and solve are solved. The optimal sensor deployment scheme is generated through the sensor deployment unit in the cloud computing platform, the sensors are deployed according to the deployment scheme, the data acquisition module acquires key parameters of the concrete building through the sensors, and the data transmission module transmits the acquired data. And the cloud computing platform performs safety assessment on the acquired data through an internal quality assessment unit so as to obtain intelligent monitoring and management of the concrete building. Potential safety hazards of the concrete building on the construction site can be found in time so that measures can be taken for repairing in advance, meanwhile, the maintenance cost and the management cost of the concrete building can be reduced, and the management efficiency is improved.

The invention provides a method and device for carrying out pre-filtering retrieval on a vector database, electronic equipment and a readable storage medium, and relates to the technical field of artificial intelligence such as big data, cloud services and deep learning. The method for performing pre-filtering retrieval on the vector database comprises the following steps: acquiring a pre-filtering retrieval condition of a target object, and obtaining a plurality of target retrieval sub-conditions according to the pre-filtering retrieval condition; according to the condition types of the plurality of target retrieval sub-conditions, determining data indexes respectively corresponding to the plurality of target retrieval sub-conditions in a vector database; and obtaining a pre-filtering retrieval result corresponding to the target object according to each target retrieval sub-condition and the data index corresponding to the target retrieval sub-condition. According to the method and the device, the speed of performing pre-filtering retrieval on the vector database can be increased.

The invention relates to the technical field of electric energy meter fault analysis, and discloses a remote electric energy meter fault analysis method based on an acquisition terminal. According to the method, the operation data and the environment data of the electric energy meter are acquired in real time through the acquisition terminal and are transmitted to the data cloud platform in real time through the network, missing data are complemented in the data cloud platform through an interpolation method, the reliability of data processing is improved, and meanwhile, the calculation capability of the data cloud platform is utilized to improve the reliability of data processing. Fault characteristic parameters such as a voltage effective value, a current effective value, a power factor, instantaneous power and harmonic content of the electric energy meter are automatically calculated, then a fault diagnosis algorithm is used to rapidly identify fault types, a detailed fault analysis report is generated in real time, and the report is fed back to operation and maintenance personnel, so that a scientific decision basis is provided for the operation and maintenance personnel. The real-time performance and the accuracy are improved, so that the fault response time is shortened, the reliability of power supply is enhanced, and the power failure risk caused by the fault of the electric energy meter is reduced.

The invention relates to the technical field of data management, and discloses an economic data management method and system.The economic data management method comprises the steps that firstly, a work log of a cloud platform storing economic data is inquired, various real-time safety parameters of the economic data are obtained, whether safety risks exist in the economic data or not is judged according to the various real-time safety parameters, and if not, the economic data are stored; if the safety risk of the economic data is not found, historical data of various safety parameters of the economic data is obtained, and the safety risk of the economic data is further judged according to the change condition of the safety parameters in various data in the historical time period, so that the situation that the safety risk is misjudged due to the error of instantaneous data and the user experience is influenced is avoided; if yes, evaluating the potential safety of the economic data in the platform according to the change of the safety coefficient of the data in each time period in the historical time period, and ensuring the safety of the economic data in the platform.

The invention relates to the field of gait analysis, and discloses a gait analysis method and system based on intelligent shoes, and the method comprises the steps: obtaining pressure distribution data captured by a plantar pressure sensor, dividing gait stages according to a pressure mode, and recognizing gait phases; the method comprises the following steps: acquiring three-dimensional acceleration, angular velocity and magnetic field data of an inertial measurement unit, and converting the three-dimensional acceleration data from an IMU coordinate system to an earth coordinate system; performing integral calculation on the three-dimensional acceleration data based on the earth coordinate system to obtain speed data; describing the time stage of the gait through the pressure mode of the pressure distribution data, detecting to obtain a zero speed moment, correcting the speed data by applying a zero speed updating algorithm according to the zero speed moment to obtain the corrected speed data, and further performing integral calculation to obtain the motion trail of the foot; through a principal component analysis algorithm, decoupling the displacement in the step length direction and the step width direction in the foot motion trail so as to obtain step length and step width parameters; the plantar pressure distribution data cloud picture and the foot movement track are combined and visualized, visual reproduction of the foot posture in the gait process is completed, and a convenient mode is provided for a patient to monitor and evaluate the gait health condition of the patient.

The invention provides a TFT instrument data interaction and cloud collaborative management system, and relates to the technical field of TFT display, and the system comprises a communication collection module which is used for deploying a plurality of integrated thin film transistor display screen units in a transport vehicle, and each unit collects the operation parameter data of the vehicle in real time through a vehicle-mounted sensor network; and the cloud receiving module is used for receiving and analyzing the operation parameters of the transport vehicle, determining the vehicle position, the operation state, the real-time parameter information content and the display mode position which need to be displayed on the monitoring interface, and mapping related data into interface elements to generate a motorcade monitoring interface. According to the invention, real-time data intelligent analysis and cloud cooperation are realized, full-link visualization and remote optimization of the vehicle state are realized, the operation efficiency is improved, and the maintenance cost is reduced.

The embodiment of the invention provides a data processing method and device, equipment and a storage medium, which are used for realizing rapidness and stability of distributed storage. The method is applied to the fields of block chains, big data, cloud technologies, artificial intelligence and the like, and comprises the steps that a data writing request sent by a client is received, and the data writing request is used for indicating data modification of N copies of target data; a consistency rule of the target storage pool is obtained, the consistency rule is used for indicating that the execution of the feedback data write request to the client is successful when response replies returned by M storage nodes are received, and M is a positive integer smaller than or equal to N; sending a data writing request to the N storage nodes to enable the N storage nodes to execute a writing operation of the target data; counting response replies fed back by the N storage nodes; when the response reply meets the consistency rule of the target storage pool, a response message is replied to the client side, and the response message is used for indicating that the data writing request is executed successfully.

The invention discloses a self-learning power transmission line fault detection system based on cloud-edge cooperation, which comprises an edge end, a cloud end and a user end, and is characterized in that the edge end collects an image sample, obtains a fault identification result and a confidence score by using a lightweight edge end detection model, judges whether an anomaly or a defect exists based on the confidence score, and sends the anomaly or the defect to the user end; when abnormity or defects exist, the side end generates alarm information and transmits the alarm information, a fault identification result and an image sample to the cloud end; the cloud side forwards the alarm information and the image sample to the user side, and if the user side carries out fault label marking and submits the fault label to the cloud side, the cloud side forms training data based on the fault label submitted by the user side; and if the user side does not submit the marked image sample to the cloud side, the cloud side forms training data based on a high-precision integrated model fault identification analysis result, and the cloud side performs incremental training on a lightweight edge detection model stored by the cloud side by using the training data to obtain model parameters and sends the model parameters to the edge side so as to iteratively optimize the model of the edge side.

The invention discloses a privacy protection-oriented cloud side-end collaborative driving style identification method, which comprises the following steps of: screening out effective driving data from a side-end collection machine through data preprocessing, processing the effective driving data into windows with the same size according to requirements, putting the batch of labeled data on a side-end model to train the side-end model and obtain probability distribution, and carrying out cloud side-end collaborative driving style identification. And constructing a probability distribution integration framework, and calculating a global probability distribution according to the probability distribution given by each side end. And guiding cloud model training according to the obtained global probability distribution because the cloud model cannot have data with labels. After cloud model training is completed, probability distribution is obtained according to a batch of driving data without labels, training of each side end model is guided according to the probability distribution, and finally the trained model is put into a driving style recognition task of new data to complete driving style recognition. And accurate identification of the driving style is realized through more comprehensive global probability distribution and an excellent cloud edge-end cooperation model.