387results about How to "Alarm is accurate" patented technology

The invention provides an oil pipeline network leakage intelligent self-adaptation monitoring system and method based on big data, and belongs to the technical field of pipeline network internal detection methods. By means of the system and method, a large number of data collected on site can be effectively analyzed in reasonable time, the state of a pipeline network can be obtained by means of an intelligent self-adaptation method, and accordingly a topological structure of the pipeline network is obtained. By means of a flow balance method and with the combination of an information consistency theory, whether the pipeline network leaks is analyzed, and the method is visual, simple, high in sensitivity and low in false alarm rate. Besides, precise warning is well carried out on detection of small leakage amount and slow leakage. By means of a generalized regression neural network, leakage positioning of the pipeline network is carried out, and accuracy of a result is improved. Accordingly, leakage detection and positioning problems of the pipeline network are solved by means of strategies based on the big data and the intelligent self-adaptation method, and meanwhile the aims of high precision and high accuracy can be achieved.

The invention relates to a mobile phone anti-theft and anti-loss alarming method based on the Bluetooth technology. The method comprises the following steps: creating a Bluetooth wireless connection between a mobile phone and an alarm; setting an alarming environment information value; obtaining an alarming distance threshold value Th; detecting a Bluetooth signal-strength value RSSI; estimating the distance Dis between the mobile phone and the alarm according to the received signal strength value RSSI and the alarming environment information value; and giving an alarm once the distance Dis exceeds the alarming distance threshold value Th. According to the invention, the alarm is more accurate with the environmental factor considered for the alarming distance; the alarming distance can be set according the requirement of the users; further, when the maximal transmission distance of the Bluetooth signal is exceeded, the alarm or the mobile phone can perform automatic detection and give an alarm; the current mobile phone GPS information can be automatically obtained and sent to a mobile phone number arranged in advance; and a method for bidirectionally detecting signal strength is adopted, thereby improving the alarming precision.

The invention discloses a perimeter protection system which comprises a fencing, a video monitoring system, an optical fiber shaking alarm system and a monitoring center, wherein a protective area is encircled by the fencing; a plurality of monitoring centers are distributed by the video monitoring system along a protection perimeter and are communicated with one another; a shaking detection line is paved by the optical fiber shaking alarm system along the protection perimeter; the video monitoring system and the optical fiber shaking alarm system are respectively connected with the monitoring center; the monitoring centers collect video information of the protection perimeter and transmit the video information to the video monitoring center; and the optical fiber shaking detection line detects shaking information of the protection perimeter and transmits the information to the monitoring center. The invention further provides a corresponding perimeter protection method which comprises the step of: sensing shake of the protection perimeter and alarming; judging and transmitting position information of an alarm point; calling a scene image of the alarming point according to the alarm information; and determining and processing the alarm information.

The invention provides a comprehensive safety intelligent monitoring method capable of effectively and quickly judging if such abnormal activities as trapping, falling and violence are generated in a car to timely and accurately give an alarm so as to effectively relieve the burden of monitoring staff. The comprehensive safety intelligent monitoring method specifically comprises the following steps: an image shot by a camera with a fixed position is used as a background model; a differential image is obtained through a difference between a present video frame and the background model by a background subtracting method; the binarization processing is performed for the differential image according to a binarization formula; then, the noise is removed through a morphology filtering algorithm; finally, the area communication analysis is performed for the noise removing result; the edge profile of the binarized image is extracted by a Canny operator; and the extracted foreground profile is compared with a human body image template library by a template matching algorithm to judge if one or more persons are in the car and if the abnormal activities are generated to trigger warning.

The invention provides a leakage rate estimation system and method for product oil pipelines based on zero dimension and belongs to the technical field of leakage rate detection for the product oil pipelines. The method comprises the following steps: effectively analyzing a lot of data collected on site in a reasonable time, obtaining the state of a whole pipe segment by using a pressure sensor array, and therefore obtaining the pressure changing trend of the pipeline; meanwhile, analyzing whether the pipeline has abnormal conditions or not by adopting a flow balance method and combining with an information consistency theory. The method is visualized and simple, is high in sensitivity and low in false alarm rate and can precisely give an alarm for detection of the small leakage rate and slow leakage. A BP neural network is adopted to calculate the rate of pressure decay of the pipeline, so that the accuracy of the final result is improved. Meanwhile, the method adopting dimensionless modeling decreases the calculated amount and increases the calculation accuracy.

The invention discloses a self-adapting compensation detection method for baseline shift of combustible gas detector, belonging to the gas detection technical field, suitable for baseline shift compensation of combustible gas detector and long-time monitoring the outburst combustible gas leakage accident. The baseline shift is as the general problem of the gas sensor and the baseline shift speed due to the environment factor and the slow change of the gas sensor is far less than the change speed of the gas leakage signal. A gas detector control center in the invention presets a changed speed threshold and the concentration alarming threshold; the baseline shift signal and the combustible gas leakage response signal are differentiated by the change speed of the initial gas concentration signal G of the gas sensor; and the gas concentration T and the baseline B are calculated after self-adapting compensation of the baseline shift; and the alarm signal E1 and the severe shift alarm signal E2 are generated based on T and B. The combustible gas detector can automatically judge the baseline shift and the user is reminded of timely calibrating or exchanging the gas sensor.

The invention provides a method and system for allowing a parking sensor to resist interference. The method for allowing the parking senor to resist interference includes the steps of detecting interference waves, namely, counting pulses with set strength interference at an MCU back wave signal end in a set period of time before an obstacle is detected by an ultrasonic transducer, and judging whether the interference is strong or not according to the number of counted interference pulses; detecting the obstacle, namely, prolonging random delay so as to distinguish the obstacle from the interference waves when the ultrasonic transducer detects the obstacle. The method and system for allowing the parking sensor to resist interference have the advantages that interference signals and the obstacle are effectively distinguished, and therefore an alarm is accurately given, and the alarm accuracy is improved.

The invention discloses a railway foreign matter clearance intrusion detection system and method. The railway foreign matter clearance intrusion detection system comprises at least one monitoring unit, an embedded foreign matter clearance intrusion detection unit connected with the monitoring units and a remote monitoring server. The embedded foreign matter clearance intrusion detection unit comprises an image acquisition chip for acquiring image data, shot by the monitoring units, of a railway area to be detected in real time, an FPGA used for judging whether a target exists in the image data or not through background subtraction frame by frame and extracting the features of the target to form feature vectors of the target if the target exists in the image data, a microprocessor for conducting foreign matter clearance intrusion judgment on the image data according to the feature vectors of the target, a remote monitoring server for confirming alarm information sent by embedded foreign matter clearance intrusion detection platforms and informing trains in relevant areas. By the adoption of the technical scheme, the railway foreign matter clearance intrusion detection system is high in recognition speed, high in detection efficiency and accurate in alarming and is suitable for an occasion needing long-distance along-railway foreign matter clearance intrusion detection.

The invention relates to an intelligent monitoring system for a flexible passive protective net. The intelligent monitoring system of the flexible passive protective net includes a dangerous source capture system, a signal processing system and an alarm system. Under a condition that the signal processing system determines that dangers happen in the protective net and / or the protective net reaches a dangerous degree, alarm signals and / or intelligence information are / is sent in one or more methods. Fiber bragg grating sensors are preinstalled on lower support ropes of the flexible passive protective net. Through monitoring the tension of the lower support ropes of the protective net, signals related with rock fall and breakage of the protective net are monitored in a real-time manner and then the signals are returned to a fiber bragg grating demodulator.

The invention provides an intelligent walking stick for tumbling remote monitoring and nursing. A battery main switch, function keys, a display screen and indicator lights are arranged on a walking stick handle. A wireless communication module and a buzzer are internally arranged on the upper portion of a walking stick supporting body, an antenna which is connected with the wireless communication module is placed on the surface of the walking stick supporting body, and an acceleration sensor, a GPS module, a microprocessor and a battery are integrally arranged on the lower portion of the walking stick supporting body. The acceleration sensor is used for being responsible for collecting the motion acceleration of the walking stick, then the microprocessor is used for processing and judging tumbling information, the GPS module is used for tracking the positional information of the walking stick in real time, and when tumbling is detected or an alarm is given actively through the function keys, the microprocessor sends the positional information of the walking stick and alarm signals. According to the intelligent walking stick, a multi-innovation Kalman filtering method is adopted, the information returned by the acceleration sensor is processed, noise signals are eliminated, so that whether an old man / woman tumbles or not is accurately judged, and accurate warning is achieved; through the function of GPS positioning and the function of buzzer warning, a user of the walking stick can be treated and cured conveniently in time.

The invention provides a driver driving state recognition-based lane departure alarming method, which comprises the following steps of (1) acquiring the condition of control of a driver over a steering wheel from a vehicle-mounted sensor; (2) establishing a statistical performance index characteristic set for operation over the steering wheel; (3) judging whether the driver is fatigue or not according to performance index characteristic information; (4) judging whether to give an alarm to the driver or not according to the fatigue state of the driver and a lane line departure condition. The method has the advantages that 1, the vehicle-mounted sensor is used for inputting the condition of operation of the driver over the steering wheel, a vehicle-mounted speed measurement signal and the like, so that dependence on an external sensor is reduced; 2, the fatigue state of the driver is determined only according to the condition of operation of the driver over the steering wheel, i.e. a rotating angle of the steering wheel, the rotating angular rate of the steering wheel, a heading angle, a heading angular rate, a transverse position and transverse speed; 3, whether an alarming signal is to be transmitted to the driver or not is judged according to the fatigue state of the driver and an alarming condition of a lane line departure alarming system, so that alarming accuracy is ensured, and the false alarm rate is greatly lowered.

The invention discloses a passenger invading signal character identifying method in an optical fiber crawl. In the method, the walking morphological feature of people is used to identify whether the signal detected by an optical fiber crawl system is aroused by artificial invasion. The walking morphological feature of people is the walking gait cycle of people; the average value thereof is 1.2 seconds; the average stepping period of people during walking is 0.6 second; wherein, the time for the feet to contact the ground is about 0.2 second; while the time for the feet to leave the ground is about 0.4 second. If the sensor change aroused by one perturbation appears in an average period of 0.5 to 0.7 second, and the continuing time is 0.2 to 0.3 second in average, then the perturbation is judged to be a passenger invasion, which can be realized by carrying out self correlation calculation on the output signal of the optical fiber crawl system; the peak period of a self correlation curve is the period of an invasion signal; the continuing time of a self-correlation peak is the continuing time of a vibration signal aroused by contact of the feet of an invader with the ground.

The invention provides a method and a system for detecting person trapping in an elevator. According to the logic combination between multiple outputted detection signals of the elevator from an elevator controller, the person trapping state of the elevator is detected. The detection signals comprise a person carrying signal, a door lock signal, a fault signal and an overhauling signal; when the person trapping state is determined, a corresponding alarming signal can be sent out. The method and the system have the advantage that the problem of false judging on person trapping in the elevator can be effectively solved.

The invention discloses a method for monitoring novel words on Internet and belongs to the field of Internet information mining. The method comprises the following steps: acquiring all articles of thesame day on home pages and subpages of all subordinates directed to by all internal links of the home pages according to a target information source collection; subjecting the articles to word segmentation, assigning weight for each word in each article, using top a words having the largest weight as alternative novel words of the article, and using the number of a plurality of articles as the freq of an alternative novel word on the same day when the word is an alternative novel word of the articles at the same time; calculating the novelty coefficients n of the alternative words according to the records of the freq of the alternative novel words in b days; and determining novelties theta according to the novelty coefficients and determining an alternative novel word as a novel word of the same day when the novelty of the word is greater than a novelty threshold thetat. The method of the invention can effectively find the novel word of the same day and direct the practice for findingand monitoring Internet information.

The invention discloses a security alarm method. Voice data of a dialogue of characters are analyzed and the language content of the voice data is identified; when the vocabulary contained by the language content is consistent with extreme vocabulary stored in a preset vocabulary database, audio feature data in the voice data are extracted and identification matching is carried out on the audio feature data; and when an emotion identification value corresponding to object audio feature data with the highest matching degree with the audio feature data exceeds a preset security threshold, an alarm message containing a target place location information is sent to a processing center, so that the processing center takes corresponding security protection measures timely based on the alarm message to avoid the potential safety hazard. With the method disclosed by the invention, accurate and real alarming can be realized; and no missing reporting and false reporting are caused. Correspondingly, the invention also discloses a security alarm apparatus, equipment and a computer-readable storage medium that have the same technical effects as the security alarm method.

The invention provides a control instruction error correction system and a control instruction error correction method. The system comprises a data integrated processing module, a comprehensive display module and a system-grade real-time monitoring module, and further comprises a real-time instruction error correction alarm module and a historical alarm comprehensive analysis module. According to the invention, through receiving data and performing data integrated processing, conflict modeling is carried out on an aircraft currently changing an instruction height, then conflict calculation is performed, whether a current control instruction causes a control unsafe accident is identified, and a conflict predetermining function is demonstrated, thus the safety of air traffic control work is improved; the system and method have the advantages of high alarm accuracy, low false alarm rate, real-time and accurate data and wide application scope; and besides, the area transplantability is high, the customization is good, the flexibility is high, alarm information, parameters and the like can be customized for each control area according to actual conditions, the system and method are already applied to guiding daily control operation work, and the alarm accuracy and the false alarm rate reach a domestic advanced level.

The invention discloses a rollover warning method and device based on a three-axis acceleration sensor. The three-axis acceleration sensor is fixedly installed on a vehicle. The rollover warning method includes the following steps that a rollover inclination angle threshold is obtained in advance; an x-axis inclination angle deviant deltax, a y-axis inclination angle deviant deltay and a z-axis inclination angle deviant deltaz of the three-axis acceleration sensor are calculated; whether deltax, deltay and deltaz are larger than the rollover inclination angle threshold or not is judged, if any one of deltax, deltay and deltaz is larger than the rollover inclination angle threshold, it is judged that the vehicle rolls over; when it is judged that the vehicle rolls over, a warning signal is sent out. Whether the vehicle rolls over is determined according to the state of the rolled-over vehicle, and the judgment method is accurate and simple. Therefore, the rollover warning method and device are timely and accurate in warning and high in practicality.

The invention relates to a remote medical monitoring and security monitoring system, and aims to solve technical problems that functions of exiting monitoring systems are simplex and the systems are poor in mutual compatibility, etc. The remote medical monitoring and security monitoring system comprises an all-in-one machine, a center cloud server and a monitoring center. The all-in-one machine integrates a medical monitoring system, a regional environment monitoring system, a video monitoring system, an entrance guard system and an alarm system into one body. Data packets acquired and generated by the all-in-one machine is uploaded to the center could server in real time. The center could server receives and processes and stores data for the monitoring center to use. According to the invention, comprehensive and multi-aspect monitoring can be carried out on monitored areas. Through a cloud technology platform, management can be carried out on the plurality of monitored areas, and large scope monitoring and remote distance observation can be realized.

The invention relates to a security and monitoring all-in-one machine and a remote monitoring system and aims to solve technical problems that existing monitoring systems are monotonous in function and poor in mutual compatibility, etc. The all-in-one machine integrates an area environment monitoring system, a video monitoring system, an entrance guard and an alarm system into one body. The all-in-one machine also includes an embedded-type video main control board and an onboard network module connected with the embedded-type video main control board. Function modules of the various systems are connected with the main control board, and data packets acquired and generated by the function modules are transmitted in real time to the main control board. The onboard network module uploads the data packets in real time to a center cloud server through the Internet. The monitor system comprises the all-in-one machine, the center cloud server and a monitoring center. According to the invention, comprehensive and multi-aspect monitoring can be carried out on monitored areas; and management can be carried out on the monitored areas through a cloud technique platform, so that large-scope monitoring and remote observation can be realized.

The invention discloses a method for detecting intrusion on the basis of video monitoring. The method includes acquiring front-end video stream data packets by the aid of FFMPEG and parallelizing acquired front-end video stream data by the aid of special CUDA (compute unified device architecture) decoding data interfaces; acquiring pedestrian classifiers by the aid of machine learning processes and establishing reference backgrounds by the aid of mathematical statistic processes. The method has the advantages that real-time online learning can be carried out, and the scene adaptability of algorithms can be improved; multiple targets can be tracked in real time by the aid of Kalman filter algorithms; pedestrian targets are tracked, the mass centers of the pedestrians are computed, target boundary violation warning effects can be realized if the mass centers are positioned in scribed regions, and video display can be carried out; the boundary violation targets can be quickly and accurately warned on the basis of automatic and artificial intelligent technologies, accordingly, intrusion of foreign targets can be effectively prevented, and the labor cost and the device-related prevention cost in perimeter prevention can be reduced to a great extent; the shortcoming of influence of change of light rays, swinging of trees and interference of insects can be overcome.

The invention relates to the field of electric power and communication, in particular to a method for alarming an electric power event. The method comprises the following steps: (S1) a terminal mechanism collocates an event rule needed to be uploaded to a server, automatically generates an event and uploads the event needed to be uploaded to the server to the server; (S2) the server acquires the reporting event of the terminal mechanism and collects data; (S3) the server analyzes the event and the data to generate alarming event; and (S4) the server sends the alarming event to a relative responsible party. The invention has the advantages that a method for electric-power line theftproof alarm monitoring with stable performance, high reliability and accurate alarm is designed by utilizing an electric power line carrier wave communication mode to realize informatization, scientific and fine management of electric power distribution management, greatly shortens the time for alarming, finding and processing failure in an electric power system so as to make remarkable contribution for optimizing marketing and reducing line damage.

The invention provides an alarm analyzing method with higher accuracy and lower cost for mobile device, net element system and network management system, which comprises: when processing an alarm happening in the net element, the relevant alarming information is added into each alarming message; the relevant alarming information comprises one or more main alarming identification information; the net element sends the alarming message containing the relevant alarming information to its upper layer management system; after receiving the alarming message, the upper layer management system process the alarming message sent from the net element according to the relevant alarming information.

The invention discloses an intelligent monitoring method aiming at well drilling overflowing and leakage, belonging to the well drilling monitoring technology. On the basis of the existing hardware device, the method comprehensively utilizes the information of a plurality of site sensors, imitates the experience of experts to put forward the reasonable judgment of accidents such as overflowing, leakage and the like as well as the regulation of alarm control, and realizes automatic monitoring and alarm control by a computer. By using the method, the capability of treating multiple information of the computer can be fully exerted, automatic monitoring wok can be realized under the unsupervised condition, leads the alarm of accidents to be more timely and accurate, reduces the probability offalse alarm, and well solves the problems in oil field well drilling production.

An online monitoring and fault diagnosis system for a wind turbine generator set bearing includes: a first step, template library creation: creating a wind turbine generator set bearing vibration template library offline; a second step, data acquisition: synchronously acquiring vibration acceleration signals of the bearing in four directions in a wind turbine generator set online monitoring platform, and allowing a data acquisition interface to complete configuration of a data acquisition card; a third step, signal initial fusion, wherein during a signal acquisition process, a sensor is affected by noise and the sensor itself, the acquired signals are preprocessed, so that the acquired signals can retain the fault information to the greatest extent; a fourth step, feature value extraction:using the EMD algorithm to decompose the fused vibration signal into different frequency bands; and a fifth step, fault monitoring: using the BP neural network and the Euclidean distance to calculatethe similarity between a fault feature vector and a vibration template, sending the results of the two methods as the D-S evidence theory as the evidence body for decision making and fault warning.

The present invention relates to an index data monitoring method and device, a storage medium and a computer device. The method comprises the steps of: obtaining normal historical index data of the same time point in the newest preset period, and calculating first baseline data of index data of the corresponding time point according to the normal historical index data; obtaining historical transaction request amount and historical index data of the same time point in the newest preset period, and calculating second baseline data of the index data of the corresponding time point according to the historical transaction request amount and historical index data; calculating and obtaining third baseline data of the index data of the corresponding same time point at the current transaction dateaccording to the first baseline data and the second baseline data; and performing monitoring of index data generated in the current transaction date according to the third baseline data, and when theindex data generated in the current transaction date is not in a baseline range of the third baseline data, performing alarm. Monitoring is performed according to the third baseline data to more timely and accurately perform alarm for the fluctuation of real-time index data.

The invention relates to a system and a method for monitoring the switching synchronism of a joint three-way / five-way turnout. The system comprises turnout driving devices, wherein an absolute encoder is arranged on a transmission shaft between two-stage speed reducers of each turnout driving device, and is introduced into the conventional turnout state monitoring system through a bus, so that the function of monitoring the synchronism in the turnout switching process is realized by monitoring and comparing readings of the encoders in real time during programmable logic controller (PLC) numerical processing. The system and the method have the advantages that 1, each turnout driving device is continuously monitored in real time, wherein whether each turnout driving device can be normally started or not and whether a transmission chain fails or not are judged; 2, three driving devices can be continuously monitored in real time under the conditions that slip is overhigh, motors and control circuits thereof fail and the like due to driving asynchronism in the turnout switching process; and 3, an alarm is timely and accurately given, the safety of equipment is ensured, and help information is provided for failure processing.

The invention discloses a lane departure warning method and system for preventing repeated warning of one-side lane departure. The method comprises the following steps of acquiring a road image in front of a vehicle; identifying a lane line of a lane in which the vehicle is located from the road image; and because a warning line, a warning end line and an external recount line are arranged at a one-side lane line, when the outer side surface of a tire on one side of the vehicle is beyond the warning line but a driver is unconscious, performing lane departure warning. The lane departure warning method and system for preventing repeated warning of one-side lane departure disclosed by the invention can judge whether the tire on one side of the vehicle is beyond the external recount line, and if not, another lane departure warning is prohibited to execute, so that the problem that the left wheel and the right wheel respectively trigger the lane departure warning is effectively solved; and if yes, a situation whether the tire on one side of the vehicle is beyond the warning end line is judged, and if yes, another lane departure warning is allowed to execute, so that the situation that the warning is missed out cannot occur.

The invention discloses a smart home security system, which comprises an emergency lamp, an alarm host, an alarm light, a loudspeaker, a center control host, a center control display screen, a monitoring module, a camera, an infrared detector, a smoke detector, a gas detector, a gas valve, a rainwater detector, a curtain motor, a window pusher, an entrance guard module, an emergency button, door magnetic detectors, curtain detectors, a remote controller, a fixed telephone, a mobile phone terminal and an alarm center. The beneficial effects are that the smart home security system is mainly formed by three modules, that is, the entrance guard module, the alarm module and the monitoring module; the entrance guard module is provided with the emergency button, which can lock each door and window and disconnect a power source under a dangerous condition, thereby improving processing capacity of the system under emergency; and the door magnetic detectors and the curtain detectors can be arranged in the corner of each door and each window, and can motivate the alarm module to alert a house owner when monitoring a stranger comes in.

The invention discloses a method for accurately detecting fault information of IT equipment in a machine room. The method comprises the steps that S1, a robot moves to a calibrated stationary point position in front of a to-be-detected cabinet; s2, a plurality of cameras on the robot record videos of the to-be-tested cabinet, and first frame images of the videos recorded by different cameras are spliced into an image covering the whole to-be-tested cabinet up and down; s3, picture analysis is carried out; s4, video analysis is carried out; and S5, picture analysis and video analysis results are integrated to obtain indicating lamp state fault information. According to the invention, the state information of the static indicator lamp and the flashing lamp of the IT equipment is analyzed bycombining a deep learning image technical means; the equipment and the fault indicator lamp of the equipment can be detected in time, the information of the equipment and the fault indicator lamp of the equipment can be accurately alarmed, solutions can be pre-judged, workers only need to finally check the inspection report, an innovative technical detection means is provided for data management of the unattended machine room, and the machine room inspection problems of low inspection efficiency, incomplete recording, less inspection, false inspection and the like are effectively solved.

The invention provides a surface detection device used for surface detection of devices on a display panel. The surface detection device comprises an ultrasound transduction detection unit and a processing unit, wherein the ultrasound transduction detection unit is used for sending ultrasound waves to a surface to be detected and obtaining feedback signals fed back by the ultrasound waves after the ultrasound waves reach the surface to be detected, and the processing unit is used for calculating the location of the surface to be detected according to the feedback signals. The surface detection device can be used for precisely detecting shock insulators and monitoring residual quantity of liquid crystal.