85results about How to "Timely alarm" patented technology

The invention provides a service application-oriented IT contralized operation and maintenance analyzing system which comprises an operation and maintenance analyzing platform and an acquisition server, wherein the operation and maintenance analyzing platform is used for sending an acquisition command to the acquisition server, analyzing originally acquired data uploaded by the acquisition server to obtain index data, and processing the index data, wherein the procedure of processing the index data comprises the step of judging whether an alarm event is generated or not according to the index data; and the acquisition server is used for driving a corresponding acquisition plug-in to acquire data for a system/service to be monitored after receiving the acquisition command sent by the operation and maintenance analyzing platform or according to a preset acquisition period, receiving originally acquired data acquired by the acquisition plug-in, and transmitting the received originally acquired data to the operation and maintenance analyzing platform. According to the invention, operation states of an application system and servicees of enterprises are monitored, an alarm can be immediately sent out, failures can be quickly solved, hidden dangers are predicted, and risks are controlled in advance.

The invention discloses an electric power operation personnel safety dressing detection method based on YOLOv3 target detection. According to the technical scheme, the method comprises: firstly, dressing pictures of electric power operators are collected, and a training sample set is made; secondly, constructing a YOLOv3 network model, and training the network model by adopting the training sampleset to obtain a trained YOLOv3 network model; Thirdly, taking the power operation on-site monitoring video as input data of the trained YOLOv3 network model, and detecting the safe dressing conditionof the power operation personnel in the video. Compared with a traditional method of manually designing characteristics, the method has the advantages that the characteristics are automatically extracted to realize detection, the applicability and the generalization performance are high, the safety helmet wearing and work clothes dressing conditions of electric power operation personnel in a video can be detected at the same time, and the method can be applied to a video monitoring system of an electric power operation site.

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 an intelligent identification method for urban violation, which comprises the following steps: step S1, obtaining a plurality of urban violation photos, and labeling the violation categories of the photos manually; step S2, creating an intelligent recognition model, and uploading the artificially labeled photos to the intelligent recognition model for in-depth learning training; step S3, obtaining a city monitoring video, intelligently identify the city monitoring video by using the optimized intelligent identification model, capturing a video frame which is judged to have illegal behavior, and creating a label; step S4, auditing the label of the screenshot manually, and if the label is incorrect, modifying the label, and entering the step S2; if the label is correct, sending the screenshot to the city management system for real-time alarm. The invention has the advantages of realizing the real-time monitoring and intelligent identification of the urban violationbehavior and reducing the cost of supervising the urban violation behavior.

The invention relates to the technical field of network security and discloses a captured host automatic judgment method and device, electronic equipment and a memory medium. The method comprises thesteps of generating security incidents according to logs output by security equipment, wherein the security incidents are incidents which describe activities of attackers, and the security incidents comprise destination IP addresses and attack phases; aggregating the security incidents with the same destination IP address, and obtaining attack chains of hosts corresponding to the destination IP address according to the attack phases of the aggregated security incidents; searching the security incidents of which source IP addresses are the IP addresses of the hosts, and correcting the attack chains according to the searched security incidents; and judging whether the hosts are captured or not according to the corrected attack chains. According to the technical scheme provided by the embodiment of the invention, a function of automatically judging whether the hosts are captured or not is realized, the problem that judgment of whether the hosts are captured or not must depend on a safetyengineer is solved, the captured hosts are alarmed timely, and losses resulting from the fact that the hosts are captured are reduced.

The invention discloses an intelligent controller of a permanent magnet mechanism breaker, which is formed by combining the permanent magnet mechanism control technology and the relay protection technology and comprises a central processing unit (CPU) module, a data collection module, a power supply management module, a permanent magnet driving module, a communication module and a man-machine interface module. The controller can directly drive a mono-stable permanent magnet mechanism, achieve fault isolation and fault location and effectively determine fault lines especially for small current grounding faults by calculating and comparing component amplitude of transient zero-sequence current in feature bands. The controller has a fault wave recording function and can record as many as 5000 accident records. The controller has a self-diagnostic function and can detect and collect faults in passages on line, control faults in the passages in remote control mode and give alarms timely. The intelligent controller is strong in functions and can sufficiently meet requirements for construction of intelligent distribution networks by being matched with the permanent magnet breaker.

The invention discloses a detecting and processing method for validity of a link among multiple nodes, and belongs to the field of multi-node link transmission of a TETRA digital trunking system. A multi-node data mutual transmission link generally has multiple phenomena, including, long-term validity, link long-term disconnection, link recovery after disconnection, disconnection in link validity process, and alternate link disconnection and recovery. Due to the phenomena, various abnormal conditions for applications are aroused in a practical running process. The detecting and processing method for validity of the link among multiple nodes comprises the steps of calculating time when a heartbeat request is sent at each time; dispersing the time on a time grid; detecting the validity of the link in a preset period; and performing detection for link recovery by adopting a hypothetical recovery strategy in a link recovery process. Through the ways mentioned above, feedback of the validity of the link is added, and accurate guarantee is provided for link selection of the applications.

The invention provides an automatic monitoring and warming system for nuclear power plant concrete structure sediment, belonging to the technical field of the nuclear power plant infrastructure construction. The system comprises a plurality of monitor sensing devices and a warning control cabinet. Each monitor sensing device comprises a fixed part and a moving part, wherein the fixed part is a microstroke switch, and the moving part is a sinker bar mounted on the top plate of the gallery. The distance between the end part of the sinker bar and the contact of the microstroke switch is equal to the sediment warning value of the gallery top plate. The contact of the microstroke switch is serially connected with the coil of a corresponding intermediate relay and connected with a power circuit. The normally open contact of the intermediate relay is respectively connected with the electric bell and the corresponding signal lamps in serial and connected in the power circuit. In construction, when the sediment of any monitoring point reaches a set early warning value, the warning control cabinet will automatically displays the audible alarm, and the position of the monitor point which reaches the early warning value is indicated via the signal lamp, so that a construction commander can rapidly understand the monitor condition and start the corresponding handle plan, to ensure the construction safety.

The invention discloses a safety protection device and a safety protection method. The safety protection device comprises a detection module for sending a detection signal and obtaining the feedback signal based on the sent detection signal, an analyzing and comparing module for determining the detection distance from the detection object based on the feedback signal and comparing the detection distance, and a safe distance alarm module for conducting alarm reminding when the detection distance is smaller than the safe distance.

The invention discloses an abnormal air alarming method and apparatus in order to timely discover danger and give off alarm. The method includes the steps of obtaining air component information in an indoor environment, comparing the air component information in the indoor environment with corresponding air component information in air change trend obtained in advance, and giving off an alarm message to a preset target device when the difference between the air component information in the indoor environment and corresponding air component information in air change trend is greater than a preset first air threshold.

The invention discloses a device monitoring apparatus for a photovoltaic power plant. The apparatus comprises a collection module for collecting running data of each device in the photovoltaic power plant in real time and current environmental data; a calculation module for analyzing and calculating and judging the operation according to the running data of each device and the environmental data respectively and sending an alarm signal to an alarm module if a device fault is confirmed; and the alarm module for executing alarm operation after receiving the alarm signal. The apparatus can automatically detect a fault according to the collected photovoltaic power plant data, reduce the workload of staff and achieve the efficiency and accuracy of device inspection.

The invention discloses an intelligent security implementing method for an active home camera, based on the following devices: a camera, a control unit, an image receiving unit, an image processing unit, a security logic unit, a main control unit, a motor circuit unit, and a mechanical unit. The camera control unit controls the image capturing, and the image receiving unit receives the image and then sends to the image processing unit. The image processing unit adopts a deep learning algorithm to automatically recognize various objects and human body motions in the image. The image processingunit outputs a monitoring APP that can be connected to a mobile phone of a user through a network. The user sets up a matching APP having a security logic unit, sets emergency contacts and dangerous scenes, and can remotely view the home situation in real time. A dangerous judgment is made and an alarm is issued if a judgment threshold of the dangerous scene set by the safety logic unit exceeds the threshold standard. No dangerous judgment is made if the threshold standard is not met.

The embodiment of the invention provides an Internet hospital system monitoring method and device, a storage medium and a program product, and the method comprises the steps: collecting business log data of an Internet hospital system, and adding the business log data into a message queue; obtaining a target index through a stream processing engine according to the service log data in the message queue; judging whether the Internet hospital system is abnormal or not according to the target index and a preset alarm rule; and if it is determined that the Internet hospital system is abnormal, sending an alarm message to a user terminal. According to the embodiment of the invention, the business log data of the Internet hospital system is collected, the business log data is analyzed and processed to obtain the target index, and the unified monitoring index system is constructed, so that unified monitoring of different Internet hospitals accessed by the Internet hospital system is realized; and whether the Internet hospital system is abnormal or not is judged based on the target index and a preset alarm rule, potential abnormity of the Internet hospital system is explored, timely alarm is facilitated, and the abnormity is recovered.

The invention relates to a distributed power supply grid connected alarm system and an operating method thereof. The distributed power supply grid connected alarm system comprises a photovoltaic assembly, an inverter, a voltage-drop release breaker, a disconnector, a grid connected electric energy meter, an alarm device and an earth leakage protection breaker. The inverter, the voltage-drop release breaker, the grid connected electric energy meter, the alarm device and the earth leakage protection breaker are connected in sequence. The voltage-drop release breaker is connected to the grid connected electric energy meter through the disconnector. The alarm device comprises a master control MCU, and also an information acquisition module, a voltage abnormity alarm module, a frequency change alarm module, a display module, an uplink communication module, a storage module and a power supply module that are connected to the master control MCU. When the distributed power supply grid connected alarm system is running, the earth leakage protection breaker is connected to a power grid side, and the alarm device can send alarm information, if voltage drops or frequency changes of the power grid side exceed a certain range, to notify management workers of checking the running condition of distributed power supply grid connected system equipment. Therefore, the safety of a power grid, equipment and operation management workers is improved in a wider scope and to a greater extent.

The invention discloses a remote monitoring system and method based on cloud computing. The system comprises a remote monitoring platform, different types of data acquisition devices distributed in different regions and a terminal alarm. The data acquisition device is arranged in different regions and is used for acquiring state data of the corresponding regions according to functions of the dataacquisition device. The remote monitoring platform is used for receiving state data acquired by different types of data acquisition device in different regions through a uniformly packaged API, and storing the acquired data in a distributed cloud database. The remote monitoring platform is also used for judging the state data acquired by the data acquisition device according to the alarm thresholddynamically configured for each piece of data acquisition device so as to determine whether to send an alarm signal to the terminal alarm. The terminal alarm is used for giving out a corresponding alarm according to the alarm signal.

The invention provides a server hard disk security measurement method and device. The method comprises the following steps: pre-storing identification information of related hard disks in a server ina trusted chip; in response to startup and electrification of the server, enabling the BIOS to acquire identification information of the hard disk and compare the acquired identification information with pre-stored identification information; and in response to the comparison failure, enabling the BIOS to control the server to stop starting and/or give an alarm. According to the invention, the hard disk on the computer can be timely alarmed after being replaced, and the method has good popularization and use values in the field of secrecy.

The invention relates to the technical field of computers. The invention particularly relates to an abnormal behavior identification method and device. The method is used for timely and effectively identifying abnormal behaviors. The method comprises the following steps of: obtaining a face image of a target person at the target location, based on a face recognition result and consulting information triggered by a target person, generating corresponding user description information; then, extracting the historical data from the pre-stored historical data; screening out reference user description information meeting a similarity condition, then determining the number of different historical target locations corresponding to each piece of obtained reference user description information and the respective generation time of each piece of reference user description information, determining that the target person has an abnormal behavior when the corresponding condition is met, and giving an alarm. Thus, when the target person appears at the target place, whether the target person has an abnormal behavior can be quickly and accurately identified according to the historical data, and thus an alarm is given in time.