926 results about "Early warning model" patented technology

The embodiment of the invention provides a modeling method of an early warning model of mixed failures and a modeling system. The modeling method provided by the invention comprises the following steps of: generating a function analyzing module on the basis of HAZOP (Hazard and Operability Analysis) or FMEA (Failure Mode and Effects Analysis); generating a degeneration analyzing module on the basis of FMEA analyzing results and a theory of stochastic processes; generating an accident analyzing module according to state monitoring data and maintenance action information; generating an action analyzing module according to output results of the function analyzing module and the degeneration analyzing module through combining a DBN (Dynamic Bayesian Network) theory; taking the output of the accident analyzing module as an inference evidence and utilizing a DBN inference algorithm to process forward and backward inferences in the same time period to generate an evaluating module for outputting factors and consequences of system failures; taking the output results of the evaluating module and the accident analyzing module as the inference evidence and utilizing the DBN inference algorithm to process forward and backward inferences in the different time periods to generate a predicating module for outputting prospective degeneration tendencies of each member of the system. The model provided by the invention can be used for tracking the failure factors of the system and inferring possible failure consequences and probability.

The invention discloses a vehicle anti-collision early warning system and method based on vehicle-road cooperation. The early warning system comprises a vehicle device, a roadside device, a control module, a communication module and a roadside prompting device; the vehicle device and the roadside device are subjected to data exchange through the communication module, the various collected data are converged to the control module through the established data channel to perform collision risk assessment, and early warning information or control instructions formed after risk assessment information are issued to a vehicle terminal or the roadside prompting device to prompt the vehicle to take corresponding measures to avoid vehicle collision. According to the invention, the accurate collision risk value of the vehicle can be obtained by means of a vehicle-road cooperation technology and a collision early warning model, the vehicle device can remind the vehicle to take corresponding anti-collision measures through voice and images after receiving early warning information, and the roadside prompting device performs collision risk prompting through lamplight or display device.

The invention provides a decision-making system and method for heavy metal pollution of farmland soil. The method is based on monitoring data of heavy metal pollution of farmland soil in a monitoringarea, constructs a farmland soil heavy metal pollution risk comprehensive assessment index system, a graded early warning model, an emergency disposal model and a crop planting adjustment model to assess the risk of heavy metal pollution in the farmland soil and quickly make an emergency response, and at the same time, can also adjust the planting structure of crops in a heavily polluted area. Thesystem includes data collection, data management, risk assessment, graded early warning, emergency disposal, planting adjustment and information visualization modules which are used for monitoring, assessing and responding quickly to heavy metal pollution in the farmland soil. The decision-making system and method for the heavy metal pollution in the farmland soil according to the invention can significantly improve the automation and intelligence levels of management decision making of the heavy metal pollution in the farmland soil.

The invention discloses an adaptive and extended intelligent monitoring and early warning system for slope stability, which includes a geographic information collection module, a three-dimensional modeling module, a scheme selection module, a data collection module, a statistical analysis module, and an early warning module. The geographic information collection module is used for collecting The geographic location information of the monitoring area c is used to generate a GIS data model. The 3D modeling module is used to generate a 3D model of the monitoring area based on the pictures in the monitoring area. The three-dimensional model of the area determines the slope type, landslide risk, and determines the monitoring point, monitoring plan, and early warning plan. The data acquisition module is arranged at the monitoring point to collect the monitoring data of the monitoring point. The statistical analysis module is used to process the monitoring data based on the monitoring data. Generate various statistical data, and the early warning module determines the early warning model, early warning level, and disposal plan based on the monitoring data and statistical data. The invention can effectively reduce risks such as landslides.

The invention discloses a wind turbine generator system fault intelligent diagnosis and early warning method based on random forests. The wind turbine generator system fault intelligent diagnosis and early warning method based on random forests includes the steps: extracting the historical data of the wind turbine generator system state as the sample data; performing exploratory analysis and preprocessing on the sample data; constructing a wind turbine generator system fault intelligent diagnosis and early warning model based on random forests, and analyzing and evaluating the model according to the model result; utilizing the model after analysis and evaluation to perform real-time diagnosis on wind turbine generator system equipment; and if the diagnosis result is not normal, sending out an alarm information by the model. The wind turbine generator system fault intelligent diagnosis and early warning method based on random forests utilizes the random forest algorithm and considers the overall characteristics of the index, so that the wind turbine generator system fault intelligent diagnosis and early warning method based on random forests can solve the problem that single index decides the equipment state and can also comprehensively consider the concealed knowledge relevance among many indexes so as to make comprehensive judgment on the output result.

The invention discloses a landslide hazard multi-stage comprehensive monitoring and early warning method, and relates to a landslide monitoring and early warning method. The method includes the steps of establishing a four-stage comprehensive early warning model based on three early warning indexes according to the topographic and geological conditions of typical landslide hazards, determining a model parameter type and an early warning critical value index through the indoor and outdoor typical rainfall landslide model experimenting method, carrying out statistic analysis of the correlation between typical landslide hazards and rainfall capacity, and establishing a mathematical calculation method suitable for typical landslide hazard early warning. Through the combination with specific monitoring data, monitoring information is fed back in time, multi-hydrologic-year detection and correction are conducted on a hazard early warning mathematical model, and then the landslide hazard multi-stage comprehensive monitoring and early warning method is obtained. A visual critical value domain discrimination graph, an analysis curve and an early warning standard introduction are given. By means of the method, the quality and effect of rainfall landslide monitoring and early warning are improved, and a scientific theory and a technological method are provided for improving the accuracy in geological hazard monitoring and early warning.

The invention belongs to the technical field of vehicle active safety, and particularly provides a rollover warning method and a rollover warning device for a heavy vehicle. The rollover warning method and the rollover warning device are low in cost, are free of influence on heavy vehicle load and mass center height, and can warn impending rollover accidents of the heavy vehicle in the next period of time in advance. The method solves the problem that an existing heavy vehicle rollover warning method can warn a vehicle rollover risk in a static threshold value mode but cannot warn an impending rollover accident to a driver in advance. The rollover warning method adopts a classical Kalman filtering technology to simplify a rollover warning model in a combination mode, the motion state of the heavy vehicle is accurately estimated, and then a TTR algorithm is utilized for predicting the rollover risks in the next period of time. The warning algorithm has better universality, and meanwhile prevents inaccurate or invalid warning caused by the changes such as the load, center-of-gravity position and rotary inertia of the heavy vehicle.

The invention discloses a cold-chain transport temperature monitoring and early warning method and system. The cold-chain transport temperature monitoring and early warning method is characterized in that a wireless sensing network is established to acquire temperature data in carriers in real time by Zigbee, the temperature during cold-chain transport can be monitored and warned early, and an early model is learned from the historical temperature data acquired based on the BP neural network in real time. The cold-chain transport temperature monitoring and early warning system can analyze the monitored temperature data in real time by the aid of the early warning model, and can automatically warn by short messages if the predicted results exceed the set warning threshold, so that alarming and early warning of the cold-chain temperature can be realized.

The invention relates to a wind turbine generator fault early warning method based on data mining. A method based on the minimal redundancy of maximum weight is adopted to select a fault feature signal and carry out dimensionality reduction on the fault feature signal, so that decision accuracy is guaranteed, and the calculated amount of data processing is reduced. Meanwhile, on the basis of the historical data of an SCADA (Supervisory Control And Data Acquisition) system, the early warning model of each equipment component of a wind turbine generator is established, and a nonlinear state estimation technology is adopted to obtain the real-time prediction value of each piece of equipment and each component. On the basis, an adaptive threshold value is designed, and the false alarm of the system due to interference including environment temperature, wind speed change and the like can be avoided. By use of the method, an abnormal state is identified before faults happen so as to be convenient in adopting a corresponding measure in time, so that preventive repair is carried out, and the method has an important practical application value.

The invention discloses a short-time early warning system for severe convection weather, and the system comprises a business management module which achieves duty shifting registration, system log, forecast grading, and business flow; a data preprocessing module which achieves real-time data collection, data quality control, data preprocessing, and severe convection weather diagnosis; a short-time temporary forecast early warning module which achieves a weather background and case library, severe convection monitoring analysis, severe convection potentiality forecast, severe convection short-time temporary forecast, and the manufacturing of forecast early warning products; and a product distribution module which achieves product packaging, early warning information inter-cut, short-message / audio-signal services, and fax / E-mail. The system constructs base information, and carries out the statistics of severe convection generation factors of the Tibet region in recent years. The system achieves the research of the large-scale conditions of severe convection weather in the Tibet region, and divides different circulation patterns for different regions. The system employs a method of composite analysis to synthesize the characteristics of an environment field, builds an early-warning model, and has higher forecast accuracy and long forecast time efficiency for the severe convection weather in the Tibet region.

The invention belongs to the technical field of equipment state early warning, and particularly relates to an evaluation value calibration method of an equipment intelligent early warning system based on an SVM regression model. The evaluation value calibration method of the equipment intelligent early warning system can remove deviation of an abnormal detecting point with respect to the evaluation values of other normal measuring points in real time sequentially through establishment of an equipment intelligent early warning model, influence relation curve fitting, identification of abnormal measuring points and calibration of interfered evaluation values, has good universality and can be grafted to other regression algorithms for performance optimization. When the evaluation value of an interfered measuring point of equipment is adjusted on line, no excessive delay effect is generated, and a real-time early warning capability of the model for the equipment is ensured; discrimination and analysis for abnormal measuring points can be carried out in real time, so that the real-time performance of judgment is ensured, and online state analysis for equipment data can be carried out; and evaluation values without interference of the normal measuring points can be output steadily. According to the evaluation value calibration method of the equipment intelligent early warning system, the reliability of the early warning model is higher, and the life cycle is long.

The invention relates to the technical field of a medical treatment information decision-making system and provides a method for extracting a morbidity key factor and early warning a disease. The method comprises the following steps: pre-treating a data: converting a medical record file data into a training value matrix; establishing a disease early warning model; and searching for an optimal early warning model through verification. The invention also provides a system for extracting the morbidity key factor and early warning the disease. Through the method and the system for extracting the morbidity key factor and early warning the disease provided by the invention, a high-risk factor of the disease is extracted, the early warning of disease mutation is performed, the realization is simple and the effect is good.

The invention provides an intrusion early warning method, an intrusion early warning device and an electronic device. The method includes the following steps that: whether an abnormal target exists ina specified area is determined according to a video frame image; if the abnormal target exists in the specified area, the target information of the abnormal target is determined, wherein the target information comprises one or a plurality of kinds of information selected from target category, target position, target size, target speed, and target trajectory; an early warning grade corresponding to the target information is generated with a pre-trained graded early warning model, wherein the graded early warning model is a deep neural network model; and early warning is performed according tothe early warning grade. With the intrusion early warning method, the intrusion early warning device and the electronic device of the invention adopted, the information of an intruding object can be better determined; graded warning is performed based on the information of the intruding object; and therefore, relevant personnel can be assisted to take targeted measures to further ensure the security of railway operation.

The invention provides a water environment and water ecology safety guarantee management system, which comprises a safety evaluation model, a risk evaluation model, a crisis early warning model, a water level water quality biomass supervision model, a sudden accident emergency response planning model, a post-accident evaluation model, a safety guarantee measure module and a comprehensive treatment module. Therefore, the water environment and water ecology safety guarantee management system belongs to an overall situation guarantee treatment management system realizing the potential safety hazard identification, the weak link early warning, the key element supervision, the sudden accident emergency handling and the post-accident evaluation and analysis, and the key monitoring and the prevention pertinence on key elements and key regions with potential safety hazards are enhanced, so the comprehensive management on the water environment and water ecology safety of monitored water regions can be enhanced; and in addition, the full-area balance supervision costs of the water environment and water ecology are effectively reduced, the supervision efficiency on the water environment and water ecology is improved, and the important significances of guaranteeing the water environment and water ecology safety of lakes and promoting the harmonious development of the social economy and the ecological environment are realized.

The invention discloses a mining subsidence monitoring and early warning method based on multi-source data. The mining subsidence monitoring and early warning method comprises the steps of: (1) acquiring standard data and GPS (Global Position System) data; (2) generating mine area DEM by utilizing Matlab software; (3) carrying out dual-rail difference interference process by virtue of DEM and SAR images to obtain a mine area surface deformation result; (4) correcting the mine area surface deformation horizontal and vertical monitoring results respectively according to GPS and standard monitoring data so as to obtain a high-accuracy three-dimensional deformation monitoring chart; (5) establishing an early warning model and a corresponding database; (6) repeating steps (2)-(5) according to the multi-phase measured data and InSAR data until the process is finished; (7) establishing time sequence and buffer analysis models of sinking, horizontal shifting, inclining, curvature deformation and the like of a subsidence area; and (8) determining the influence area according to the time sequence and buffering analysis models and achieving early warning. The mining subsidence monitoring and early warning method integrates standard data, GPS data and SAR data organically, so that the accuracy of the monitoring result is greatly improved.

The invention discloses a system and a method for real-time unmanned plane flight state evaluation early warning. The system comprises an evaluation early warning platform, an unmanned plane and a ground monitoring station, wherein the ground monitoring station is used for monitoring the flight state data of the unmanned plane and transmitting the data to the evaluation early warning platform, the evaluation early warning platform is used for reading the flight state information and flight control data of the unmanned plane, through the flight control data of the unmanned plane accumulated at each inspection, an unmanned plane flight state evaluation model is constructed, the flight state of the unmanned plane is determined, a flight instruction transmitted by a host computer is further received, and flight of the unmanned plane is adjusted. According to the method, the flight control data of the unmanned plane can be inspected through a power transmission line, through the combined time sequence analysis, clustering analysis and supporting vector machine analysis method, the flight state evaluation early warning model is constructed, and real-time flight state evaluation early warning of the unmanned plane in the power transmission line inspection process can be realized.

The invention discloses a pump storage unit operation stability deterioration early warning method and a pump storage unit operation stable state deterioration early warning system. The pump storage unit operation stable state deterioration early warning method comprises the steps that stability real-time monitoring of a pump storage unit is carried out to acquire a monitoring value of a stability parameter; according to the working condition parameter of the pump storage unit, an upper limit value and a lower limit value, which are corresponding to a current working condition and a currently-compared stability parameter, are searched in a pre-established numerical early warning model; when the monitoring value is greater than or equal to the lower limit value, and is smaller than or equal to the upper limit value, the stability state of the pump storage unit is determined to be normal, and the monitoring value is stored in a normal state record library; when the monitoring value is smaller than the lower limit value or greater than the upper limit value, early warning information is output, and the monitoring value is stored in an abnormal state record library. By additionally providing the working condition parameter, and the pre-established early warning model is used to provide different alarming limit values for the different working conditions of the pump storage unit, and therefore the effective early warning in different working conditions is realized, the stability problem of the pump storage unit is discovered timely, and the precision of the early warning is improved.

The invention discloses a nuclear power plant equipment fault early warning analysis method and system based on a dynamic simulation model. The system comprises a nuclear power station real-time / historical database which is used for providing basic parameters needed by the dynamic simulation model during normal operation; the dynamic simulation model built based on a nuclear power station total range analog machine, wherein the dynamic simulation model is used for providing reference parameter values of all monitoring equipment; a real-time data collection center which is used for measuring the real-time operation parameter values of all monitoring equipment; and a dynamic early-warning analysis platform which is used for inputting the real-time operation parameter values of all monitoring equipment and the corresponding reference parameter values into a trend early-warning model, carrying out the dynamic early-warning analysis of all monitoring equipment through employing a comprehensive early-warning analysis strategy in the trend early-warning model, and starting early-warning for the monitoring equipment which is determined to be fault equipment through the comprehensive early-warning analysis strategy. The method and system can recognize the fault equipment in a nuclear power plant in advance during operation and start early-warning, so as to prevent the fault equipment from being damaged further, thereby avoiding an accident happening to the nuclear power plant, and improving the operation efficiency of the nuclear power plant.

The invention discloses a real-time logging data acquisition and synchronous early-warning method, comprising the following steps: configuring data acquisition modules in well sites, configuring basic information of the well sites, and establishing an early-warning model; acquiring the real-time data of the well sites, and analyzing and formatting the real-time data; classifying and carrying out statistics on the real-time data according to the early-warning model for carrying out comprehensive analysis and judgment and early warning; and processing the formatted data and the abnormal information obtained based on the early-warning model into a standard format, sending to a base, and carrying out secondary early-warning. In the method, the real-time logging data of a site mud logging unitunder a plurality of operating conditions is acquired and then subject to intellectualized comprehensive analysis and judgment, and primary abnormal early warning is carried out; and meanwhile the data is processed into the standard format and then sent to the base so as to realize a synchronous secondary warning mechanism between the site and the logistical base, thus achieving the purposes of remote automatic monitoring and early warning during the well drilling process in a plurality of the well sites, reducing downhole accidents and meeting the demand for safety drilling production.

The application provides a battery thermal runaway early warning processing method and device, equipment, and a storage medium. The method comprises the steps: acquiring current thermal runaway stateinformation of a target battery at current time; acquiring current fault information of acquisition equipment corresponding to the current thermal runaway state information; determining a current thermal runaway state information change result of the target battery based on the current thermal runaway state information and historical thermal runaway state information of the target battery at a first time, setting the first time to be the time of a preset time before the current time; performing early warning analysis on the current thermal runaway state information, the current fault information and the current thermal runaway state information change result based on a preset early warning model to obtain a current thermal runaway early warning level corresponding to the target battery; and performing thermal runaway early warning processing based on the current thermal runaway early warning level. According to the application, graded alarming and processing can be realized, the earlywarning cost is reduced, and the thermal runaway early warning function can be normally carried out under the condition that some sensors fail.

The invention discloses a novel analytical method of deformation monitoring data of a dam slope. Dam slope displacement and a factor influencing the dam slope displacement are monitored, monitoring data are selected, a regression analysis based on proper orthogonal decomposition is carried out to obtain a feature orthogonal basis of proper orthogonal decomposition; quantile regression analyses are carried out on the feature orthogonal basis and the damp slope displacement to obtain regression equation parameters of all quantiles, and then original data substitution is carried out to obtain a quantile regression equation of the original data; and then changing situations of regression equation estimation values of independent variables under different quantiles as well as whether the regression equation estimation values pass the significance test successfully are analyzed and the influence degrees on the dam slope displacement by the independent variables are determined. According to the invention, the method has the great significance in early-warning model establishment for a slope of a dam, slope prediction and forecasting method providing, and slope sensitive influence factor analyses.

The invention discloses a driving safety monitoring method and device, computer equipment and a storage medium. The method comprises the following steps of: obtaining a real-time video stream in a driving process; dRIVING SCENE DATA, human face feature detection is carried out according to the real-time video stream; determining an auxiliary recognition score corresponding to the detected face feature, the driving behavior is analyzed according to the real-time video stream; obtaining abnormal behavior scores, enabling efficient acquisition of real-time data, the timeliness of driving safety monitoring is improved; meanwhile, based on the driving scene data and the driving scene risk early warning model; determining a driving risk score corresponding to the driving scene data, and determining a target behavior score based on the auxiliary identification score, the abnormal behavior score and the driving risk score, comparing the target behavior score with a preset early warning threshold, sending early warning prompt information if the target behavior score exceeds the preset early warning threshold, and executing real-time early warning measures to ensure the accuracy and efficiency of driving safety monitoring.

The invention discloses an illness risk value prediction method and electronic equipment and is applied to hypertension analysis. The method comprises steps that static data and dynamic data of usersare acquired; the acquired static data and the dynamic data of the users are pre-processed; a dynamic early warning model is constructed through logistic regression and a dynamic vector machine; according to the pre-processed data, the dynamic early warning model is utilized to classify the users; according to classification of the users, corresponding risk values are calculated; the risk values stand for the hypertension risk of the users. The method is advantaged in that a dual-data system is employed, through improved logistic regression and the dynamic vector machine, the chronic illness risk dynamic hybrid early warning model is established, and the accurate illness risk probability value is provided.

The invention discloses a power-grid-GIS-based lightning analysis and early warning method and a system thereof. The method comprises the following steps that: a basic power grid equipment geological information graph layer is established according to the coordinate of the power grid equipment; real-time monitoring data of lightning weather of the power grid equipment coverage area as well as meteorological cloud atlas data are obtained; a lightning weather data model is established; and the real-time monitoring data and the lightning weather data model are matched, a lightning early warning grade is determined, the basic power grid equipment geological information graph layer and the lightning weather chart layer are associated, and the power grid equipment information and lightning early warning data are overlapped and displayed on the power grid GIS platform. The method and system have the following beneficial effects: the lightning falling area is analyzed and a professional power lightning disaster early warning model is established; power equipment is associated; and precise early warning for the single line and equipment is realized. Targeted measures and specifications for geological disaster prevention can be formulated; and the losses of the power grid can be reduced.

The invention relates to a deep neural network in artificial intelligence, and provides a default probability prediction method and device, computer equipment and a storage medium. The method comprises the steps of acquiring an insurance source identifier, and inquiring whether a matched target entity exists in a preset knowledge graph or not according to the insurance source identifier; When thematched target entity exists, obtaining a to-be-predicted entity associated with the target entity in a preset knowledge graph; And based on the state attribute information of the to-be-predicted entity and the association relationship between the to-be-predicted entity and the target entity, obtaining a risk early warning feature vector, inputting the risk early warning feature vector into a trained risk early warning model for prediction, obtaining a default probability vector, and obtaining the default probability of the to-be-predicted entity according to the default probability vector. Byadopting the method, the efficiency and accuracy of default probability prediction can be improved.

A petrochemical engineering wharf handling process safety early-warning system based on a LabVIEW comprises a video monitoring unit, a real-time monitoring unit, a safety management unit and an early-warning unit, wherein the real-time monitoring unit is especially used for monitoring parametric variation generated in liquid handling operation and comprises corresponding testing systems. The real-time monitoring unit collects signals of the concentration, temperature, pressure, flow and the like of liquid to be monitored and obtains related parameter values A, quantized values B of indexes in the safety management unit are manually input, and the parameter values A and the quantized values B are transmitted into a petrochemical engineering wharf handling process extension theory early-warning model developed based on a LabVIEW platform; when all the parameters meet settings, the early-warning model is triggered, the current state of the handling process is provided, and an early warning is given under abnormal or nonstandard conditions to inform related personnel to take pre-control measures. According to the system, the functions of on-line monitoring early warning, data query and visual display of the parameters of liquid in an oil pipeline can be achieved in the handling process.

The invention discloses a multi-dimensional operational risk evaluating method for a communication network and relates to power grid operation risk evaluating methods. According to an existing electric power communication system monitoring system, the change of the network quality can not be represented truly, the guarantee capability of a system can not be evaluated comprehensively, and potential risks of the system can not be managed or controlled accurately. The multi-dimensional operational risk evaluating method for the communication network comprises the following steps that a multi-dimensional operational risk evaluating and early warning model for the communication network is established, the weights of evaluation indexes are determined, a grading rule for the evaluation indexes is determined, and comprehensive grading and managing are achieved through an electric power communication system. According to the multi-dimensional operational risk evaluating method for the communication network, a macroscopic result is reflected by evaluating the difference on the whole, the specific problem is accurately positioned and the future development trend and potential risks are predicted by means of multi-dimensional item evaluation, existing maintenance source investment efficiency and benefits can be evaluated in a quantitative mode, the method has great significance in guiding the total number of continuously-invested resources and the structural configuration, and an important basis is provided for daily management of the system.

The embodiment of the invention provides an end-side-cloud architecture-based community fire monitoring system. The community fire monitoring system comprises 6LoWPAN wireless sensing networks, cloudgateways and a cloud platform, wherein the 6LoWPAN wireless sensing networks have a fire risk sensing function; the cloud gateways are connected with the 6LoWPAN wireless sensing networks and is connected to the cloud platform through the 6LoWPAN wireless sensing networks; the cloud gateway performs data transmission based on an MQTT protocol; and the cloud platform utilizes a community fire fuzzyneural network early warning model and the edge calculation of the cloud gateway to realize real-time early warning and control of fire risk. According to the embodiment of the invention, the low-cost arrangement of the fire monitoring network is realized, the signal quality is improved, and the reliability is enhanced; the data transmission of the wireless sensing networks and the cloud platformis carried out through the single MQTT protocol, and the data transmission efficiency and the timeliness of fire early warning are improved; and the perception, transmission, analysis, early warningand control of the fire risk of the community are comprehensively realized.

Analysis, prediction and warning are performed at first on power grid safety by analyzing a risk type, establishing an early warning model, analyzing a risk index and managing risk early warning, and then under the guidance of an early warning target, a safety risk early warning management organization structure is constructed, a necessary early warning function is set and exerted, and a targeted early warning scheme is formulated, organized and implemented.