Urban rail transit power distribution room monitoring method and device based on digital-visual fusion

Abstract

The invention relates to the technical field of auxiliary monitoring, in particular to an urban rail transit power distribution room monitoring method and device based on digital-visual fusion, and the method comprises the following steps: carrying out the identity recognition of a person entering a power distribution room, and obtaining the identity recognition information; performing state identification on equipment in the power distribution room to obtain state identification information; performing non-contact temperature measurement management and control on equipment in the power distribution room to obtain temperature data; managing and controlling the environment state of the power distribution room to obtain environment data; acquiring video data, and fusing the data on a picture formed by the video data; and judging the state of the power distribution room according to the fused internet-of-things data. According to the method, video data and internet-of-things data are accessed into the auxiliary monitoring device in the rail transit power distribution room, data cleaning and logic control are carried out on the video data and the internet-of-things data through the device, patrol service is carried out on the power distribution room, and therefore safety early warning, digital-visual fusion, statistical analysis and the like are achieved.

Description

Technical Field

[0001] The present invention relates to the field of auxiliary monitoring of rail transit distribution rooms, and in particular to a monitoring method and device for urban rail transit distribution rooms based on digital and visual fusion. Background Art

[0002] The rail transit auxiliary monitoring device is based on the development of the Internet, the Internet of Things, and the Industrial Internet. It effectively integrates technologies such as neural network video recognition algorithms, data collection and cleaning algorithms, Internet of Things big data analysis algorithms, flexible production and automation. At the same time, after preliminary cleaning, screening, calculation and analysis of the data, it can be forwarded or pushed to the intelligent auxiliary monitoring system to improve the service quality of the rail transit distribution room, improve operation and maintenance efficiency, and ensure operational safety.

[0003] The rail transit distribution room is an important part of the power system in the rail transit system. It is responsible for converting the electric energy of the high-voltage transmission line into low-voltage electric energy suitable for rail transit, and for power supply and distribution protection for each line and station. The design and operation management of the distribution room plays a vital role in the safe operation of the rail transit system and the stable supply of electric energy. The auxiliary monitoring device in the distribution room can provide auxiliary monitoring functions in three dimensions: personnel control, operation safety, and environmental monitoring. According to the specific conditions of the rail transit distribution room and the standards of personnel, operation, and environment, the auxiliary monitoring requirements of the rail transit distribution room can be completed quickly and accurately.

[0004] For example, the Chinese invention patent with announcement number CN108765398A discloses a rail transit monitoring and management platform, including: an image monitoring system, a data identification and integration system, a display system, a data processing system and a communication system; the image monitoring system, the data identification and integration system, the display system and the communication system are all connected to the data processing system. A rapid analysis and processing process of acquiring, locating and finding faults is realized, that is, the rail transit monitoring and management platform quickly processes data and locates and finds the fault area, and quickly notifies the staff at the rail transit site to carry out maintenance. Summary of the invention

[0005] The method of the present invention accesses video data and IoT data in an auxiliary monitoring device in a rail transit distribution room, performs data cleaning and logic control on the video data and IoT data through the device, and inspects the distribution room. A monitoring method and device for an urban rail transit distribution room based on digital and visual fusion are proposed to help managers to more intelligently process distribution room business and manage equipment, making the operation and maintenance of the distribution room safer, more reliable and more efficient.

[0006] In order to achieve the above objectives, the following technical solutions are proposed: A monitoring method for a distribution room of urban rail transit based on digital and visual fusion comprises the following steps: Identify the personnel entering the power distribution room and obtain their identity information; Identify the status of the equipment in the power distribution room and obtain status identification information; Conduct non-contact temperature measurement and control on the equipment in the power distribution room to obtain temperature data; Control the environmental status of the power distribution room and obtain environmental data; Acquire video data, and fuse identity recognition information, state recognition information, temperature data, and environmental data on a screen formed by the video data to obtain fused IoT data; Determine the status of the power distribution room based on the integrated IoT data; The IoT data is stored.

[0007] Preferably, the method for identifying a person entering a power distribution room specifically comprises the following steps: When the radar or intelligent dual-detection detector detects the entry of a person, it calls the camera to the entry area or calls the fixed camera, and then determines the number of people and uploads the personnel information to the monitoring video screen; Import the staff's work clothes compliance base map and identify the personnel through machine learning algorithms. If the person entering is a staff member, the staff entering the area will be matched with the imported work clothes compliance base map for similarity. If they are not wearing work clothes, sound and light alarms will be used to warn them.

[0008] Preferably, the method for identifying the status of equipment in the power distribution room specifically comprises the following steps: Set multiple camera preset coordinates according to the display orientation of electrical equipment and instruments in the power distribution room; Turn on the lighting in the distribution room, and call the camera to move to the preset coordinates to monitor the equipment. Classify the equipment data according to the time period or the type of equipment cabinet, identify the status values ​​of meters, finger pressure plates, and indicator lights, and analyze the health status of the equipment.

[0009] Preferably, the state value includes: the measured values ​​of the current, voltage and SF6 of the pressure gauge, the on or off state of the finger pressure plate, and the on and off state of the indicator light.

[0010] Preferably, the non-contact temperature measurement and control of the equipment in the power distribution room specifically includes the following steps: Configure the status represented by the operating temperature range of equipment components in the equipment room; After receiving or timing the start of the temperature inspection command for equipment components, the preset coordinates of the thermal imaging camera are automatically called to conduct video surveillance of the equipment components and obtain the temperature data measured by the thermal imaging camera. The temperature data is classified according to the time period or the type of equipment cabinet; The temperature data is compared with the operating temperature of the equipment components, and the comparison results are uploaded to the auxiliary monitoring device. The peripherals in the distribution room are turned on or off in response to abnormal temperature data.

[0011] Preferably, the control of the environmental status of the power distribution room specifically includes the following steps: Collect environmental data and upload it to the auxiliary monitoring device, which cleans the environmental data; The environmental data after cleaning is used to judge the environmental conditions of the distribution room according to the operating standards of the distribution room equipment. The auxiliary monitoring device can link the peripherals of the distribution room to turn on or off when the environment is abnormal.

[0012] Preferably, the environmental data is collected by an environmental collection sensor, a water immersion detection sensor, a sulfur hexafluoride detection sensor, and a power monitoring sensor.

[0013] Preferably, the cleaning of the environmental data adopts automatic cleaning, and the data is automatically cleaned by defining data cleaning rules, which specifically includes the following steps: Check if the data conforms to the defined data type; Remove missing values, fill missing values ​​with mean or median; Detect outliers in the data and mark or remove them; Check the data for duplicate values ​​and remove them; Checks whether the data meets predefined consistency rules.

[0014] Based on the same concept, a monitoring device for a distribution room of urban rail transit based on digital-visual fusion is also proposed, including an auxiliary monitoring device and an auxiliary monitoring system, wherein the auxiliary monitoring device includes an interface module, an intelligent analysis module and a business module; Interface module, used to access IoT data and video data and call camera coordinates; Intelligent analysis module, used for data cleaning and logic control of video data and IoT data, and analysis of the identity and work clothes of staff entering the IoT detection area through neural network recognition algorithm; The business module is used to output the processed data, analysis results and prompt information to the auxiliary intelligent monitoring system for recording and display, and to display the numerical values ​​on the video screen. It also provides normal / abnormal fusion prompts for the distribution room status according to business requirements, and provides logical linkage processing for abnormal situations. Perform identity recognition on personnel entering the distribution room to obtain identity recognition information; perform status recognition on equipment in the distribution room to obtain status recognition information; perform non-contact temperature measurement and control on equipment in the distribution room to obtain temperature data; perform control on the environmental status of the distribution room to obtain environmental data; obtain video data, and fuse the identity recognition information, status recognition information, temperature data and environmental data on a screen formed by the video data to obtain fused IoT data; determine the status of the distribution room based on the fused IoT data; and store the IoT data.

[0015] Compared with the prior art, the present invention has the following beneficial effects: the method of the present invention accesses video data and IoT data in the auxiliary monitoring device in the rail transit distribution room, performs data cleaning and logic control on the video data and IoT data through the device, and performs inspection services on the distribution room to achieve safety warning, digital and visual fusion, statistical analysis, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of a monitoring method for a distribution room of urban rail transit based on digital and visual fusion; Figure 2 It is a framework diagram of a method and device for monitoring a distribution room of urban rail transit based on digital and visual fusion in Example 1; Figure 3 is a schematic diagram of the safety management and control method for rail transit power distribution room staff in Example 1; Figure 4 It is a schematic diagram of the method for inspecting and controlling the equipment status in a rail transit power distribution room in Example 1; Figure 5 It is a schematic diagram of the non-contact temperature measurement and control method for key equipment components in a rail transit power distribution room in Example 1; Figure 6 is a schematic diagram of the environmental state control method of the rail transit power distribution room in Example 1; Figure 7 It is a schematic diagram of the application of the present invention in a monitoring scenario of a main control room in a railway traction substation. DETAILED DESCRIPTION

[0017] The present invention is further described in detail below in conjunction with test examples and specific implementation methods. However, this should not be understood as the scope of the above subject matter of the present invention being limited to the following embodiments, and all technologies realized based on the content of the present invention belong to the scope of the present invention.

[0018] Example 1 A monitoring method and device for urban rail transit distribution room based on digital and visual fusion, such as Figure 1 As shown, the following steps are included: Identify the personnel entering the power distribution room and obtain their identity information; Identify the status of the equipment in the power distribution room and obtain status identification information; Conduct non-contact temperature measurement and control on the equipment in the power distribution room to obtain temperature data; Control the environmental status of the power distribution room and obtain environmental data; Acquire video data, and fuse identity recognition information, state recognition information, temperature data, and environmental data on a screen formed by the video data to obtain fused IoT data; Determine the status of the power distribution room based on the integrated IoT data; The IoT data is stored.

[0019] As a specific embodiment, Figure 2 As shown, the auxiliary monitoring device uses radar and intrusion detection IoT data to link the video and use a neural network recognition algorithm to conduct a comprehensive analysis of personnel, analyze the identity and work clothes of the personnel entering the IoT detection area, and display the IoT data in a fusion manner on the video screen. According to business requirements, normal / abnormal video fusion prompts are given to assist management personnel in the safety management of personnel in the distribution room, and provide logical linkage processing for abnormal situations.

[0020] During the inspection of the distribution room, the auxiliary monitoring device controls the camera pan to rotate to the specified coordinates (preset points) during the early debugging according to the display orientation of the electrical equipment and instruments in the distribution room. The monitoring device periodically calls the pan / tilt camera coordinates (preset points) according to the set rules. The camera coordinates are called to use the neural network recognition algorithm to identify the status of the meter, finger pressure plate, and indicator light data, and to display the numerical values ​​on the video screen. According to business requirements, the object status is normal / abnormal fusion prompts are given to assist management personnel in the safety management of the distribution cabinet in the distribution room.

[0021] The auxiliary monitoring device is connected to the thermal imaging camera to obtain the temperature of the observed object without contact, and displays the numerical fusion on the video screen. It gives normal / abnormal fusion prompts of the observed object status according to business requirements, and assists management personnel in the safety management of key components in the distribution room.

[0022] The auxiliary monitoring device accesses the IoT data such as the distribution room environment, component power, SF6-O2, water level, etc., and displays the numerical fusion on the video screen. It gives the normal / abnormal fusion prompt of the distribution room status according to business requirements, assists the management personnel to carry out the safety management of the distribution cabinet in the distribution room, and gives logical linkage processing for abnormal situations. Based on the comprehensive analysis of IoT perception data including video intelligent analysis data and sensor detection data, auxiliary monitoring and analysis are carried out on the daily operation and maintenance of the rail transit distribution room, and linkage processing is carried out for abnormalities to realize the effective application of the operation of the rail transit distribution room.

[0023] Furthermore, in order to achieve the above object, the technical solution adopted by the present invention is as follows: 1. Safety management and control methods for rail transit power distribution room staff, such as Figure 3 As shown, including: 1.1: When a radar detector or intelligent dual-detection detector detects a person entering, it pushes the information to the auxiliary monitoring device.

[0024] 1.2: The auxiliary monitoring device automatically calls the camera to turn to the personnel entry area or directly calls the fixed camera to identify the personnel, and superimposes the detection information on the called camera screen.

[0025] The detection information includes data such as the person’s name / stranger, number, helmet wearing status, and dress code.

[0026] 1.3: The auxiliary monitoring device identifies and confirms the identity information of the person entering. If the person entering is a staff member, the staff member’s identity information will be superimposed on the video screen for fusion display, and the Internet of Things data will be generated and pushed to the auxiliary intelligent monitoring system for recording; if the person entering is not a staff member, the stranger warning information will be superimposed on the video, and the auxiliary monitoring device will output the sound and light alarm information for warning, and the Internet of Things data will be generated and pushed to the auxiliary intelligent monitoring system for recording.

[0027] 1.4: After the personnel identity information is confirmed, the auxiliary monitoring device will conduct video analysis and personnel status recognition analysis on the staff entering the designated area, superimpose the staff's clothing status on the video screen for fusion display, and generate IoT data to push to the auxiliary intelligent monitoring system for recording; if the personnel status does not meet the management requirements of the rail transit distribution room, the auxiliary monitoring device will output sound and light alarm information for warning, and generate IoT data to push to the auxiliary intelligent monitoring system for recording.

[0028] 1.5: When the radar or detector feedbacks that there is no staff in the distribution room, the auxiliary monitoring device determines that there is no staff in the distribution room, and superimposes the distribution room status on the video screen for fusion display, generates IoT data and pushes it to the auxiliary intelligent monitoring system for recording. The data includes the name / stranger of the person, number, single / double / multiple people, helmet wearing status, dress code and other data.

[0029] Among them, as a specific embodiment, the machine learning algorithm identifies the identity of the intruder, and after identifying the identity, it determines whether to alarm or perform the next step of work clothes analysis to determine whether the operation is standardized and give corresponding business events.

[0030] (1) Import the worker's multi-scene and multi-angle tooling compliance base map into the road auxiliary monitoring device management module; (2) After the staff member passes the identity verification, the auxiliary monitoring device pulls the personnel monitoring video stream for work analysis, performs similarity matching with the imported tooling compliance map, and displays the matching results in screen information fusion.

[0031] There are many ways to judge the auxiliary monitoring device: (1) Microwave radar peripheral judgment can determine the person and number of people; (2) Infrared dual identification can determine if an object has entered; (3) Surveillance cameras conduct intrusion detection, support personnel judgment, and filter interference.

[0032] 2. Inspection and control methods for equipment status in rail transit distribution rooms, such as Figure 4 As shown, including: 2.1: The auxiliary monitoring device receives or regularly starts the equipment instrument status inspection command, automatically turns on the distribution room lighting, automatically calls the coordinates of the camera, and conducts video monitoring of the meters, finger pressure plates, and indicator lights on the equipment.

[0033] 2.2: The auxiliary monitoring device receives the video stream data from the camera and identifies the status values ​​of the equipment components (meters, finger pressure plates, indicator lights, etc.) in the video screen. The status values ​​include: the measurement value of the pressure gauge (current / voltage / SF6), the opening or closing of the finger pressure plate, the on or off of the indicator light, etc. And the display identification and analysis information is integrated on the camera screen to generate IoT data to push the auxiliary intelligent monitoring system. Through status identification, the status data information of the equipment components can be extracted and superimposed on the video image. Through the background, the operation and maintenance personnel can fully grasp the equipment operation status, accurately identify equipment abnormalities, and truly realize the monitoring and management of the distribution room, greatly reducing the workload of daily patrol personnel, and providing complete and reliable protection for unmanned distribution stations.

[0034] 2.3: The auxiliary monitoring device can perform statistical analysis on the inspection results and integrate the statistical analysis results into the camera screen for display, so that the management personnel can intuitively understand the status of the distribution room.

[0035] The statistical analysis method is used as a specific embodiment: (1) Data can be classified according to different dimensions. For example, the time dimension can be classified by day, week, month, quarter, year and other time periods. For the space dimension, statistics are focused on different equipment cabinets. (2) Analyze the health status of the equipment by identifying the status values ​​of equipment components in the video images.

[0036] 3. Non-contact temperature measurement and control methods for key equipment components in rail transit distribution rooms, such as Figure 5 As shown, including: 3.1: The auxiliary monitoring device receives or regularly starts temperature inspection instructions for key equipment components, automatically calls the coordinates of the thermal imaging camera, and conducts video monitoring of key equipment components.

[0037] 3.2: The auxiliary monitoring device receives the temperature and video stream data from the thermal imaging camera, collects the temperature values ​​of the equipment components in the video screen, and integrates and displays the temperature value information on the camera screen to generate IoT data and push it to the auxiliary intelligent monitoring system.

[0038] 3.3: The auxiliary monitoring device can link fans, air conditioners and other peripherals to turn on and off according to the operating standards of the distribution room equipment in response to abnormal temperature values, and integrate the linkage results on the camera screen to generate IoT data and push them to the auxiliary intelligent monitoring system.

[0039] 3.4: The auxiliary monitoring device can perform statistical analysis on the temperature results of the inspection, analyze whether the temperature measurement data collected by the thermal imaging camera is within the allowable value, analyze the health status of the equipment, and display the infrared thermal map at the same time, and display the area exceeding the threshold through the area demarcation mark. It can identify and mark the location of the specific items that have been demarcated, display the temperature, set the threshold and issue an alarm. Analyze whether the temperature measurement data collected by the thermal imaging camera is within the allowable value, and analyze the health status of the equipment. The statistical analysis results are integrated and displayed on the camera screen, so that managers can intuitively understand the status of the distribution room.

[0040] The conditions for starting the auxiliary monitoring device are: (1) Receive an operation instruction from the operation and maintenance personnel to start the auxiliary monitoring device; (2) Set up an automatic startup program to automatically start according to the operation and maintenance inspection time.

[0041] 4. Environmental status control methods for rail transit power distribution rooms, such as Figure 6 As shown, including: 4.1: Environmental data collection sensors, water immersion detection sensors, sulfur hexafluoride detection sensors, and power monitoring sensors transmit environmental data to the auxiliary monitoring device.

[0042] 4.2: The auxiliary monitoring device cleans the data, integrates the data in real time and displays it on the camera screen, generates IoT data and pushes it to the auxiliary intelligent monitoring system.

[0043] 4.3: The auxiliary monitoring device can link fans, air conditioners and other peripherals to turn on and off according to the operating standards of the distribution room equipment in case of environmental abnormalities, and integrate the linkage results on the camera screen to generate IoT data and push them to the auxiliary intelligent monitoring system.

[0044] The data from each sensor will have large errors due to interference from environmental factors, so the data needs to be cleaned, sorted and filtered before it can be used (for example, IO input state voltage, raw data 40000+, output data ≤0V after clear sorting); As a specific embodiment, the cleaning step adopts automatic cleaning to automatically clean the data by defining data cleaning rules: 1. Data type check: Check whether the data conforms to the defined data type, such as numbers, dates, strings, etc.

[0045] 2. Missing value processing: Process missing values, such as deleting missing values ​​or filling missing values ​​with mean or median.

[0046] 3. Outlier detection: Detect outliers in the data and mark or delete them.

[0047] 4. Duplicate value processing: Check the duplicate values ​​in the data and delete them.

[0048] 5. Data consistency check: Check whether the data meets the predefined consistency rules, such as whether the relationship between the data is correct and whether the data format is consistent.

[0049] Example 2 As a practical application, Figure 7 As shown in the figure, the monitoring scene of the main control room in the railway traction substation uses a complete system solution to collect video, intelligent events, lighting control, sensor and other events and data in the main control room, including comprehensive application services, video management substation, dynamic environment measurement and control management substation, etc. Establish an information-based main control room with "vision" as the core and "hearing", "smell" and "touch" as important supports to fully perceive the real-time operating status of the main control room.

[0050] There is also an intelligent inspection subsystem, as follows: (1) System composition The intelligent inspection subsystem consists of the line network management and control layer and the station and section front-end perception layer.

[0051] The management and control layer provides comprehensive visual recognition algorithm services, voice recognition, robot navigation and control and other multimodal artificial intelligence services, and is deployed on the existing network-level power supply intelligent operation and maintenance platform of the Huangdao Control Center.

[0052] The front-end perception layer consists of mobile intelligent robots + cameras deployed in the substation.

[0053] (2) System functions Environmental testing: ground water, wall testing, pipeline testing, temperature and humidity testing.

[0054] Equipment inspection: equipment appearance damage, cabinet door switch status, sign compliance, and water accumulation on the cabinet roof.

[0055] Equipment detection: abnormal noise, temperature measurement, partial discharge detection.

[0056] Fault linkage: fault / fire linkage, weather linkage, gas extinguishing mode linkage, switch operation linkage.

[0057] Finally, it should be noted that the embodiments described in detail above are only the best practices of the invention and cannot be used to limit the scope of rights of the invention. Equivalent replacement of the technical solutions recorded in the aforementioned embodiments does not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the invention, and they should all be included in the scope of the claims and description of the invention.

Claims

1. A monitoring method for urban rail transit distribution room based on digital and visual fusion, characterized in that: The following steps are involved: Identify the personnel entering the power distribution room and obtain their identity information; Identify the status of the equipment in the power distribution room and obtain status identification information; Conduct non-contact temperature measurement and control on the equipment in the power distribution room to obtain temperature data; Control the environmental status of the power distribution room and obtain environmental data; Acquire video data, and fuse identity recognition information, state recognition information, temperature data, and environmental data on a screen formed by the video data to obtain fused IoT data; Determine the status of the power distribution room based on the integrated IoT data; The IoT data is stored.

2. The urban rail transit distribution room monitoring method based on digital and visual fusion as claimed in claim 1 is characterized in that: The method for identifying a person entering a power distribution room specifically comprises the following steps: When the radar or intelligent dual-detection detector detects the entry of a person, it calls the camera to the entry area or calls the fixed camera, and then determines the number of people and uploads the personnel information to the monitoring video screen; Import the staff's work clothes compliance base map and identify the personnel through machine learning algorithms. If the person entering is a staff member, the staff entering the area will be matched with the imported work clothes compliance base map for similarity. If they are not wearing work clothes, sound and light alarms will be used to warn them.

3. The urban rail transit distribution room monitoring method based on digital and visual fusion as claimed in claim 1 is characterized in that: The method for identifying the status of equipment in the power distribution room specifically comprises the following steps: Set multiple camera preset coordinates according to the display orientation of electrical equipment and instruments in the power distribution room; Turn on the lighting in the distribution room, and call the camera to move to the preset coordinates to monitor the equipment. Classify the equipment data according to the time period or the type of equipment cabinet, identify the status values ​​of meters, finger pressure plates, and indicator lights, and analyze the health status of the equipment.

4. The urban rail transit distribution room monitoring method based on digital and visual fusion as claimed in claim 3 is characterized in that: The state values ​​include: the measured values ​​of the current, voltage and SF6 of the pressure gauge, the on or off state of the finger pressure plate, and the on and off state of the indicator light.

5. The urban rail transit distribution room monitoring method based on digital and visual fusion as claimed in claim 1 is characterized in that: The non-contact temperature measurement and control of the equipment in the power distribution room specifically includes the following steps: Configure the status represented by the operating temperature range of equipment components in the equipment room; After receiving or timing the start of the temperature inspection command for equipment components, the preset coordinates of the thermal imaging camera are automatically called to conduct video surveillance of the equipment components and obtain the temperature data measured by the thermal imaging camera. The temperature data is classified according to the time period or the type of equipment cabinet; The temperature data is compared with the operating temperature of the equipment components, and the comparison results are uploaded to the auxiliary monitoring device. The peripherals in the distribution room are turned on or off in response to abnormal temperature data.

6. The urban rail transit distribution room monitoring method based on digital and visual fusion as claimed in claim 1 is characterized in that: The control of the environmental status of the power distribution room specifically includes the following steps: Collect environmental data and upload it to the auxiliary monitoring device, which cleans the environmental data; The environmental data after cleaning is used to judge the environmental conditions of the distribution room according to the operating standards of the distribution room equipment. The auxiliary monitoring device can link the peripherals of the distribution room to turn on or off when the environment is abnormal.

7. The urban rail transit distribution room monitoring method based on digital and visual fusion as claimed in claim 6 is characterized in that: The environmental data is collected by an environmental collection sensor, a water immersion detection sensor, a sulfur hexafluoride detection sensor, and a power monitoring sensor.

8. The urban rail transit distribution room monitoring method based on digital and visual fusion as claimed in claim 6 is characterized in that: The cleaning of the environmental data adopts automatic cleaning, and the data is automatically cleaned by defining data cleaning rules, which specifically includes the following steps: Check whether the data conforms to the defined data type; Remove missing values, fill missing values ​​with mean or median; Detect outliers in the data and mark or remove them; Check the data for duplicate values ​​and remove them; Checks whether the data meets predefined consistency rules.

9. A monitoring device for urban rail transit distribution room based on digital and visual fusion, characterized in that: It includes an auxiliary monitoring device and an auxiliary monitoring system, wherein the auxiliary monitoring device includes an interface module, an intelligent analysis module and a business module; Interface module, used to access IoT data and video data and call camera coordinates; Intelligent analysis module, used for data cleaning and logic control of video data and IoT data, and analysis of the identity and work clothes of staff entering the IoT detection area through neural network recognition algorithm; The business module is used to output the processed data, analysis results and prompt information to the auxiliary intelligent monitoring system for recording and display, and to display the numerical fusion on the video screen, give the normal / abnormal fusion prompt of the distribution room status according to business requirements, and give logical linkage processing for abnormal situations; Perform identity recognition on personnel entering the distribution room to obtain identity recognition information; perform status recognition on equipment in the distribution room to obtain status recognition information; perform non-contact temperature measurement and control on equipment in the distribution room to obtain temperature data; perform control on the environmental status of the distribution room to obtain environmental data; obtain video data, and fuse the identity recognition information, status recognition information, temperature data and environmental data on a screen formed by the video data to obtain fused IoT data; determine the status of the distribution room based on the fused IoT data; and store the IoT data.

Images