Safety helmet monitoring system for petroleum downhole operation environment

Abstract

The invention discloses a safety helmet monitoring system for a petroleum downhole operation environment, and relates to the technical field of petroleum operation safety, the safety helmet monitoring system comprises a monitoring layer and an application layer, the application layer comprises a plurality of information monitoring modules deployed at an intelligent safety helmet terminal; the information monitoring module comprises a wearing state recognition unit which judges whether the current corresponding safety helmet is in a wearing state or not; the dynamic identification unit is used for acquiring behavior dynamic data of a corresponding safety helmet user through preset power consumption control scheduling logic; the environment sensing unit is used for collecting environment data related to operation safety in a working scene; the monitoring layer divides a working scene into a plurality of working areas. According to the invention, the system can adjust the working mode of the safety helmet in real time by comprehensively collecting environment data and behavior dynamic data of an operator; moreover, a dynamic power consumption management mechanism is introduced, and the working mode of the monitoring equipment is automatically adjusted according to the danger level of the operation area.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil operation safety, in particular to a safety helmet monitoring system for an oil downhole operation environment. BACKGROUND

[0002] In modern industrial operations, the operation environment is often full of various potential dangerous factors, especially in high-risk environments such as oil downhole operations. In order to ensure the safety of the operation personnel, various monitoring systems have been deployed in many operation environments, among which the safety helmet monitoring system as an important personal protective equipment is widely used in various operation scenes. These systems usually include various sensors such as gas sensors, temperature and humidity sensors and motion sensors, which can monitor the operation environment and the state of the operation personnel in real time and issue timely warnings.

[0003] Through retrieval, a safety helmet environment monitoring and early warning method and system are disclosed in Chinese patent (publication number: CN118692219A). This patent continuously and dynamically monitors the environmental parameters in a sealed space, timely captures and analyzes the environmental change trend, evaluates the environmental conditions at different heights, provides a scientific basis for predicting harmful gas concentration changes, realizes personalized early warning and in-depth analysis of environmental parameters, and improves safety protection in different spatial operation environments.

[0004] Although the existing safety helmet monitoring system has made significant progress in basic environmental monitoring and personnel state identification, in actual application, many systems have energy efficiency problems. Traditional safety helmet monitoring systems often rely on high-frequency data collection and processing, which can improve the accuracy of data, but also significantly increases the power consumption of the system, limiting the endurance of the device, especially in long-time, high-load operation environments, resulting in the inability of the device to continuously and effectively operate, therefore the present application proposes a safety helmet monitoring system for an oil downhole operation environment. SUMMARY

[0005] The purpose of the present application is to provide a safety helmet monitoring system for an oil downhole operation environment to solve the problems mentioned in the background.

[0006] The present application can be realized by the following technical scheme: a safety helmet monitoring system for an oil downhole operation environment, comprising: a monitoring layer and an application layer; The application layer includes a plurality of information monitoring modules deployed on intelligent safety helmet terminals; The information monitoring module includes a wearing state identification unit, a dynamic identification unit and an environmental perception unit; The wearing state identification unit is used to determine whether the corresponding safety helmet is in a wearing state; The dynamic identification unit is used to acquire behavior dynamic data of the corresponding safety helmet user, so as to judge whether the safety helmet user is in a moving state, and includes the following working logic: Extremely low power consumption state: The dynamic identification unit collects images at a preset low frequency, does not perform grid division and pixel color unit extraction operation, only performs gray mean value or brightness distribution feature extraction on the whole frame of image, and judges whether there is obvious environmental change based on the change amplitude of the whole image gray mean value; Low power consumption state: The dynamic identification unit collects images, divides the images into a plurality of grid regions, and extracts the pixel color unit in each grid region and records the time sequence thereof; The dynamic identification unit judges whether there is a moving behavior based on the change characteristics of the pixel color unit in each grid region in the time dimension; High power consumption state: Collect complete images and upload to the edge computing module for further identification or modeling; The environment perception unit is used to collect environment data related to work safety in the work scene, and the environment data includes but is not limited to temperature, humidity, harmful gas concentration, and light intensity; The environment perception unit uploads the collected environment data to the monitoring layer; The monitoring layer includes a plurality of edge computing modules and processing platforms; The monitoring layer divides the work scene into a plurality of work areas, and each edge computing module is arranged in each work area and receives the behavior dynamic data and environment data uploaded by the information monitoring module in each safety helmet in the corresponding work area, and sends the data to the processing platform after preprocessing; The processing platform gathers the environment data and behavior dynamic data of the plurality of edge computing modules, establishes a space-time mapping relationship based on the two, constructs a cooperative judgment mechanism, and is used for source point positioning, trend inference and propagation path simulation of dangerous events.

[0007] Further technical improvements of the application are that the monitoring layer associates each work area with the on-site deployment in the work scene, divides the risk level of each work area, and the safety level of different work areas is stored in the database of the monitoring layer or pre-stored in the information monitoring module of the application layer.

[0008] Further technical improvements of the application are that in the low power consumption state, the dynamic identification unit collects images and divides them into a plurality of grid regions, extracts the pixel color unit in each grid region and records the time sequence thereof, and judges whether there is a moving behavior based on the change characteristics of the pixel color unit in the time dimension; If it is judged that the static state lasts for a preset time length, the information monitoring module is triggered to enter a high-power consumption state.

[0009] Further technical improvements of the present application are as follows: the working method of the dynamic identification unit in the low-power consumption state comprises: A1, image acquisition and division processing: The dynamic identification unit acquires images at a preset sampling period, and divides the acquired images into a plurality of regular grid regions; A2, pixel color unit extraction and sequence construction: In each grid region, the representative pixel color unit in the grid region is extracted, and a time sequence of the pixel color unit is established for each grid; And in each grid in the continuous image frame, the change process of the pixel color unit is recorded; A3, local grid change judgment: For each grid region, the difference value of the adjacent frame pixel color unit in the time dimension is calculated; If the difference value exceeds a preset color change threshold, it is judged that the grid region has moved; A4, overall movement state judgment: If the number of grid regions that meet the pixel change condition simultaneously exceeds a preset number threshold, the dynamic identification unit judges that the safety hat wearer is in a moving state; Otherwise, it is judged that the safety hat wearer is in a static state; A5, if the dynamic identification unit judges that the safety hat wearer is in a moving state, the dynamic identification unit maintains the current low-power consumption sampling logic; If the dynamic identification unit judges that the safety hat wearer is in a static state, and the duration reaches a preset duration, enter a high-power consumption state, enable image acquisition upload and environment full parameter sampling, and increase the sampling frequency.

[0010] Further technical improvements of the present application are as follows: the method for obtaining the representative pixel color unit comprises: extracting the color channel value of a plurality of pixel points at a preset position in each grid region in the image, and simply averaging the extracted pixel values to obtain the representative pixel color unit of the grid region, specifically: Z1, the image is divided into m*n grid regions of equal size, each grid is numbered , and each grid region includes K pixel points; Z2, a plurality of preset position pixel points are selected in each grid region, in this embodiment, the center point and the 4 corner points are selected, a total of 5 pixels are selected as representative sample pixels; Obtain the primary color channel value representing the sample pixel, and then take a simple average of the selected sample pixel values ​​in integer form to obtain the representative pixel color unit of the grid, denoted as: In the formula, N is the number of samples; This is the color channel value of the k-th selected pixel; Z3. Finally, each grid area corresponds to a representative pixel color unit. value; All grid areas The values ​​form a two-dimensional array and change over time to form a time series; In subsequent judgments, use The time difference is used to determine whether a movement has occurred.

[0011] A further technical improvement of the present invention is that: the processing platform constructs a time series based on environmental data collected in each work area, and determines whether there is an increasing trend by analyzing the changing trend of environmental data in the same area, and further determines whether there is a propagation path of environmental anomalies based on the relationship between the changing gradient and time delay of environmental data between adjacent work areas.

[0012] A further technical improvement of the present invention is that: the method for the processing platform to determine whether a propagation path exists includes: S1. Time series trend analysis of a single grid area: For the same work area In, at multiple consecutive time points , ,..., Environmental data V (such as temperature, humidity, or concentration of harmful gases) collected online are used to construct time series data. ; Processing platform for time series Trend fitting is performed; in this embodiment, a sliding window is used to calculate the local mean increase. If time series If a preset trend threshold is met, an abnormal growth trend is determined in the corresponding grid area. S2. Analysis of the propagation trend in adjacent areas: The calculation module considers adjacent work areas, for example... , Etc., respectively construct the synchronization value vector of their environmental data at the same time node; Then compare the current value of the grid region component with the value at the previous moment to determine if it exists: The values ​​of environmental data exhibit a "spatial gradient" (e.g., gas concentration increases from the inner zone to the outer zone). In terms of time, changes in the outer region are delayed compared to those in the inner region (there is a propagation lag). If the above conditions are met at the same time, it is judged that the environment data has an outward diffusion trend from the core area; S3, space-time mapping relationship and path simulation; The judgment result in S2 is written into the three-dimensional model M[i][j][t] to form a multi-dimensional abnormal change graph based on area and time; According to the maximum spatial gradient path and the shortest time delay path, a propagation path graph is constructed.

[0013] Further technical improvements of the present application are that the processing platform adjusts the propagation path of the environment data according to the type of environment data and its triggering condition, specifically including: When the type of environment data changes, the propagation path of the environment data is adjusted based on related environmental conditions such as air flow, temperature, wind direction, etc.; When the environment data is affected by process or equipment factors in the work scene, the propagation path of the environment data is adjusted based on job factors such as device status, position, etc., and the diffusion trend of the data in the adjacent work area is further predicted.

[0014] Compared with the prior art, the present application has the following beneficial effects: The present application can greatly enhance the intelligence and adaptability of the system by comprehensively collecting environment data and behavior dynamic data of the workers, and the system can adjust the working mode of the safety helmet in real time to cope with different working environments and risk levels; Moreover, the present application can automatically adjust the working mode of the monitoring device according to the danger level of the work area by introducing a dynamic power consumption management mechanism. For example, in areas with low danger level, the system will enter a low-power state to extend the battery life; while in high-risk environments, the system will switch to a high-power state to enable more sensors and more frequent data collection, thereby improving response speed and safety. In addition, the system can intelligently adjust the frequency and accuracy of data collection according to real-time environmental changes and the behavior state of the workers, achieving a balance between low power consumption and high efficiency; On the other hand, the present application also has strong multi-scene adaptability. In complex environments, the system can analyze and predict possible risk events in real time according to the surrounding environmental changes and dynamic behaviors of the workers, and through the intelligent warning function, the system can issue an alarm in advance and provide emergency response suggestions, effectively improving the safety of the workers, and through the propagation path analysis, the system can simulate the diffusion path of dangerous events such as gas leakage, further enhancing the scientificity and operability of emergency management. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to facilitate understanding of those skilled in the art, the present application will be further described below with reference to the accompanying drawings.

[0016] Figure 1 A system block diagram of the present application. DETAILED DESCRIPTION

[0017] In order to further clarify the technical means and effects taken by the present application to achieve the predetermined inventive purposes, the specific embodiments, structures, features and effects thereof according to the present application are described in detail below in combination with the drawings and preferred embodiments.

[0018] Embodiment 1 Please refer to Figure 1 As shown in the figure, the present application provides a safety helmet monitoring system for oil downhole operation environment, which comprises a monitoring layer and an application layer. The monitoring layer associates each operation area with the field deployment in the work scene, divides the risk level of each operation area, and the safety level of different operation areas is stored in the database of the monitoring layer or pre-stored in the information monitoring module of the application layer. In this embodiment, the association of the operation area and the field deployment is shown in the following table 1: Table 1 Region number Work content Key equipment Dangerous factors A area Well repair operation Elevator equipment, winch, derrick High-altitude falling, personnel falling B area Fracturing operation Fracturing pump, hydraulic pipeline High pressure, risk of bursting C area Wellhead operation Valve, well killing manifold Toxic gas leakage D area Inspection channel No equipment or away from equipment Relatively safe The application layer comprises a plurality of information monitoring modules deployed in the intelligent safety helmet terminal. The information monitoring module comprises a wearing state recognition unit, a dynamic recognition unit and an environment perception unit. The wearing state recognition unit is used to judge whether the current corresponding safety helmet is in the wearing state, and in this embodiment, any one of pressure sensing, capacitance or infrared is used to recognize the state of the safety helmet contacting the human body. When the safety helmet is in the unworn state, other functional units in the information monitoring module do not work. The dynamic recognition unit is used to obtain the behavior dynamic data of the corresponding safety helmet user to judge whether the safety helmet user is in the moving state, including the following working logic: Very low power consumption state: The dynamic recognition unit collects images at a preset low frequency, does not perform grid division and pixel color unit extraction operation, only performs gray mean value or brightness distribution feature extraction on the whole frame image, and judges whether there is obvious environmental change based on the change amplitude of the whole image gray mean value. Low power consumption state: The dynamic recognition unit collects images, divides the images into a plurality of grid areas, and the dynamic recognition unit extracts the pixel color unit in each grid area and records its time sequence. The dynamic recognition unit judges whether there is a moving behavior based on the change characteristics of the pixel color unit in each grid area in the time dimension. If it is judged that the static state lasts for a preset time length, the information monitoring module is triggered to enter a high-power consumption state; The working method of the dynamic recognition unit in the low-power consumption state comprises: A1, image acquisition and division processing: The dynamic recognition unit acquires images at a preset sampling period, and divides the acquired images into a plurality of regular grid regions; A2, pixel color unit extraction and sequence construction: In each grid region, a representative pixel color unit in the grid region is extracted, and a time sequence of pixel color units is established for each grid; And in each grid in the continuous image frame, the change process of the pixel color unit is recorded; The method for obtaining the representative pixel color unit comprises: extracting color channel values of a plurality of pixel points at preset positions in each grid region in the image, and simply averaging the extracted pixel values to obtain the representative pixel color unit of the grid region, specifically: Z1, the image is divided into m*n grid regions of equal size, each grid is numbered , and each grid region includes K pixel points; Z2, a plurality of preset position pixel points are selected in each grid region, in this embodiment, the center point and the 4 corner points are selected, a total of 5 pixels are selected as representative sample pixels; The main color channel value of the representative sample pixel is obtained, the value of the selected sample pixel is simply averaged in integer form to obtain the representative pixel color unit of the grid, denoted as: ; In the formula, N is the number of samples; is the color channel value of the kth selected pixel; Z3, finally, each grid region corresponds to a representative pixel color unit value; The values of all grid regions form a two-dimensional array, and form a time sequence with time change; In subsequent judgment, the time difference value of is used to judge whether movement occurs currently; A3, local grid change judgment: For each grid region, the difference value of the adjacent frame pixel color unit in the time dimension is calculated; If the difference value exceeds a preset color change threshold, it is judged that the grid region has moved; A4, overall movement state judgment: If the number of grid regions satisfying the pixel change condition simultaneously exceeds the preset number threshold, the dynamic recognition unit determines that the safety hat wearer is in a moving state; Otherwise, it is determined that the safety hat wearer is in a stationary state; A5, if the dynamic recognition unit determines that the safety hat wearer is in a moving state, the dynamic recognition unit maintains the current low-power sampling logic; If the dynamic recognition unit determines that the safety hat wearer is in a stationary state, and the duration reaches a preset duration, enter a high-power state, enable image collection upload and environment full parameter sampling, and increase the sampling frequency; High-power state: Collect complete images and upload to the edge computing module for further identification or modeling; In actual use, after the user wears the safety hat into the corresponding work area, the edge computing module positions the information monitoring module, the information monitoring module obtains the danger level of the corresponding work area, and then the information monitoring module dynamically adjusts the activation state, sampling period, and upload strategy of each unit based on the preset power consumption control scheduling logic, to ensure that it is in a comprehensive perception state in high-risk areas, and reduce energy consumption and prolong the endurance in low-risk areas; In this embodiment, the power consumption control scheduling logic is shown in Table 2: Table 2 Danger level Power consumption mode Control strategy 1-2 (low risk) Very low power consumption Only periodic whole frame image acquisition is performed, gray scale or brightness change is extracted, no image upload, and the sampling frequency is low 3 (medium risk) Low power consumption Start low-power image grid analysis, monitor movement state, and upload frequency is moderate 4-5 (high risk) High power consumption Image acquisition upload + environment full parameter sampling is enabled, sampling frequency is improved, and high alert state is entered The environment perception unit is used to collect environment data related to work safety in the work scene, which includes but is not limited to temperature, humidity, harmful gas concentration, and light intensity. Specifically, based on the actual use scene, the corresponding data collection device can be installed on the safety hat to obtain the data; The environment perception unit uploads the collected environment data to the monitoring layer; The monitoring layer includes multiple edge computing modules and processing platforms; The application layer includes multiple information monitoring modules deployed on the intelligent safety hat terminal; The monitoring layer divides the work scene into multiple work areas, and each edge computing module is arranged in each work area and receives the behavior dynamic data and environment data uploaded by the information monitoring module in each safety hat in the corresponding work area, and sends the data to the processing platform after preprocessing. Specifically, the data preprocessing includes: Preliminary filtering: eliminate redundant reporting data, invalid frame data, and repeated timestamp data; Formatting: unify the data uploaded by different safety hats into a structured format; Time labeling: use the processing platform as the time source for timestamp standardization to ensure that the time dimension is aligned in subsequent model calculations; Neighborhood aggregation: aggregate multiple node data in the same job area into neighborhood cluster data package, facilitate subsequent diffusion modeling; The processing platform aggregates the environmental data and behavior dynamic data of multiple edge computing modules, and establishes a space-time mapping relationship based on the two, constructs a collaborative judgment mechanism, which is used for source positioning, trend inference and propagation path simulation of dangerous events; The processing platform constructs a time series based on the environmental data collected in each job area, and judges whether there is a growth trend by analyzing the change trend of the environmental data in the same area. Further, based on the change gradient and time delay relationship of the environmental data between adjacent job areas, it is judged whether there is a propagation path of environmental abnormality. The method for the processing platform to judge whether there is a propagation path comprises: S1, time series trend analysis of single grid area: In the same job area , the environmental data V (such as temperature, humidity or harmful gas concentration) collected at consecutive time points , ,..., is constructed into a time series ; The processing platform performs trend fitting processing on the time series , and in this embodiment, the local mean increment is calculated by using a sliding window; If the time series satisfies a preset change trend threshold, it is judged that the corresponding grid area has an abnormal growth trend; S2, adjacent area propagation trend analysis: The calculation module constructs a synchronous value vector of the environmental data of adjacent job areas, such as , , etc., at the same time node; And compare the current value and the previous time value of the grid area, judge whether there is: The value of the environmental data shows a "spatial gradient" (such as the concentration of the gas increasing from the inner area to the outer area); At the time point, the outer area changes later than the inner area (there is a propagation lag); If the above conditions are met at the same time, it is judged that the environmental data has a diffusion trend from the core area to the outside; S3, space-time mapping relationship and path simulation; The judgment result in S2 is written into the three-dimensional model M[i][j][t] to form a multi-dimensional abnormal change graph based on area and time; According to the maximum spatial gradient path and the shortest time delay path, a propagation path graph is constructed; The processing platform adjusts the propagation path of the environmental data according to the type of the environmental data and the triggering condition, specifically including: When the type of environmental data changes, the propagation path of the environmental data is adjusted based on related environmental conditions such as air flow, temperature, wind direction, etc. When the environmental data is affected by process or equipment factors in the work scene, the propagation path of the environmental data is adjusted based on job factors such as equipment status, location, etc., and the diffusion trend of the data in the adjacent work area is further predicted.

[0019] In this embodiment, each safety helmet is equipped with multiple sensors for collecting data such as temperature, humidity, gas concentration, etc. in the environment where the workers are located, including but not limited to: Temperature and humidity sensor: monitors temperature and humidity in the work environment; Gas sensor: monitors harmful gas concentration (such as , ); The collected environmental data is uploaded to the edge computing module in the corresponding work area in real time; The computing platform analyzes and adjusts the propagation path based on the uploaded environmental data (such as gas concentration, temperature, humidity, etc.) in each work area, combined with environmental conditions such as wind direction, air flow, etc. After obtaining the environmental data and analyzing its trend, the propagation path analysis module adjusts the propagation path of the data in the work area; The processing platform receives environmental data from different work areas, performs data preprocessing and space-time modeling, and based on the space-time data model, predicts the diffusion path and trend of the environmental data (such as gas), generates a path map of environmental data propagation, and issues a warning to the workers; Specifically, the processing platform performs time series analysis on the data to determine whether there is an upward trend in gas concentration, temperature, etc. For example, if the carbon dioxide concentration gradually increases from 200 ppm to 250 ppm within a continuous time period (such as 10 minutes), the system will determine that there is a growth trend in the gas concentration in this area; When the gas concentration, temperature, etc. data meets the preset growth trend threshold, the propagation path analysis module is triggered and enters the path adjustment mode; Adjustment based on the type of environmental data: when the system detects an increase in gas concentration, the propagation path analysis module combines data from anemometers, weather stations, and work equipment to predict the diffusion path of the gas. For example: If the wind speed is 3 m / s and the wind direction is southeast, the system will predict that the gas will spread along the wind direction to adjacent work areas (such as areas B and C); If the temperature is too high, the speed of gas diffusion may increase, and the system will further adjust the prediction of the diffusion path; Specifically, the current wind speed and direction are read from the anemometer or micro weather station module accessed by the edge computing module, and the wind direction is converted into a coordinate axis vector direction; The initial diffusion path is projected and expanded along the coordinate axis vector direction to generate an offset path; If the temperature is higher than the preset temperature threshold, the system updates the gas diffusion speed by increasing the correction coefficient to update the diffusion radius, i.e., the new diffusion radius is the original diffusion radius multiplied by the correction coefficient; Adjustment based on job site conditions: When the gas concentration is affected by the working equipment or process (such as the operation of a fracturing pump), the propagation path analysis module will consider the working state of the equipment and the location of the equipment exhaust. If the exhaust of the equipment is located in a certain direction of the working area, the system will adjust the diffusion path of the gas to ensure that a more realistic path is simulated; For example, if the device P1 (fracturing pump) releases a large amount of gas when it is working, the propagation path analysis module will simulate that the gas first expands in the direction of the device exhaust; Finally, according to the changes in environmental data and the adjusted propagation path, the processing platform generates a gas diffusion path diagram to predict the diffusion trend of the gas in the future; Specifically, the running state of all high-emission equipment in the current working area (such as the operation of a fracturing pump, the location of the exhaust air outlet) is retrieved, and it is determined whether there is a running behavior associated with the gas concentration (such as exhaust frequency, power level); A direction vector is generated based on the orientation of the exhaust and the emission intensity, the direction vector and the wind direction are multiplied by a preset weight respectively, and added to obtain a multi-factor diffusion direction; The original diffusion path is adjusted to expand along the multi-factor diffusion direction, and the concentration weight is adjusted to form a high concentration block in combination with the emission intensity of the equipment; The processing platform sends a warning notice to the workers according to the predicted diffusion path, and the processing platform sends an audible and visual or vibration warning to the safety helmet in the diffusion path area, the workers will be informed of the high-risk area in the working area, and are prompted to evacuate the area; The system also automatically adjusts the working state of the safety helmet, switches the information monitoring module corresponding to the working area from a low-power state to a high-power state, and enables higher frequency image acquisition and uploading to monitor environmental changes in real time; And by positioning the safety helmet, the personnel evacuation situation in each working area is judged; Specifically, the system superimposes the path model on the working area and marks the diffusion boundary at different time points to generate a visual diffusion layer; At the same time, the system obtains the positioning information of each safety helmet in real time, and judges whether it is in the updated diffusion path, and then generates a list of affected personnel; Finally, a hierarchical warning signal is sent to the safety helmet in the personnel affected list, and the evacuation direction is prompted, while instructing the information monitoring module to switch to a high-power mode to improve the sensing accuracy and upload frequency.

[0020] Embodiment 2 On the basis of embodiment 1, embodiment 2: in a weak light or night environment, the dynamic recognition unit compensates the brightness of the collected image through any one of image normalization and threshold self-adaptation; Specifically, the system performs a linear normalization operation on the pixel gray scale range in the current frame image to improve the image contrast; and dynamically adjusts the change threshold of the grid pixel color unit according to the average brightness of the image to adapt to the trend of the change value being reduced in weak light; If necessary, a lighting assembly is arranged on the safety helmet, and the lighting assembly is used to assist sampling to improve the availability of the pixel color unit and ensure that the system can still recognize the dynamic state under low power consumption.

[0021] The above formulas are dimensionless values calculated, the formulas are obtained by software simulation of a large amount of data to obtain a formula of the latest real situation, and the preset parameters and threshold values in the formulas are set by a person skilled in the art according to the actual situation.

[0022] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above with a preferred embodiment, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content without departing from the scope of the technical solution of the present application to obtain equivalent embodiments with equivalent changes. Any simple modification, equivalent change and modification of the above embodiments made according to the technical essence of the present application, without departing from the technical solution of the present application, still belongs to the scope of the technical solution of the present application.

Claims

1. A safety helmet monitoring system for oil downhole operation environment, comprising a monitoring layer and an application layer, characterized in that: the application layer comprises a plurality of information monitoring modules deployed on intelligent safety helmet terminals; the information monitoring module comprises: a wearing state recognition unit to determine whether the corresponding safety helmet is in a wearing state; a dynamic recognition unit to obtain the behavior dynamic data of the user of the corresponding safety helmet through a preset power consumption control scheduling logic; an environment sensing unit to collect environment data related to job safety in the work scene; the monitoring layer divides the work scene into a plurality of work areas, and comprises: an edge computing module, each of which is arranged in each work area and receives and pre-processes the behavior dynamic data and environment data uploaded by each information monitoring module in the corresponding work area; a processing platform to gather the environment data and behavior dynamic data of the plurality of edge computing modules and establish a space-time mapping relationship based on the two.

2. A hard hat monitoring system for use in a downhole oil environment according to claim 1, characterized in that, The monitoring layer associates each work area with the field deployment in the work scene and classifies the risk level of each work area; the edge computing module of the corresponding work area locates the information monitoring module, the information monitoring module obtains the danger level of the corresponding work area, and then the information monitoring module works based on the preset power consumption control scheduling logic.

3. A hard hat monitoring system for use in a downhole oil environment according to claim 2, characterized in that, The power consumption control scheduling logic comprises: an extremely low power consumption state: the dynamic recognition unit collects images at a preset low frequency, performs gray mean value or brightness distribution feature extraction on the entire image, and determines whether there is a moving behavior based on the change amplitude of the overall gray mean value of the image; a low power consumption state: the dynamic recognition unit collects images and divides them into a plurality of grid areas, and determines whether there is a moving behavior based on the change characteristics of the pixel color unit in each grid area in the time dimension; a high power consumption state: collecting complete images and uploading them to the edge computing module.

4. A hard hat monitoring system for use in a downhole oil environment according to claim 3, characterized in that, The working method of the dynamic recognition unit in the low power consumption state comprises: A1, image acquisition and division processing: the dynamic recognition unit collects images at a preset sampling period and divides the collected images into a plurality of regular grid areas; A2, pixel color unit extraction and sequence construction: in each grid area, extract the representative pixel color unit in the grid area, and establish a time sequence of pixel color units for each grid; and in each grid in the continuous image frame, record the change process of the pixel color unit; A3, local grid change judgment: for each grid area, calculate the difference value of the adjacent frame pixel color unit in the time dimension; if the difference value exceeds the preset color change threshold, it is determined that the grid area has moved; A4, overall moving state judgment: if the number of grid areas that meet the pixel change condition exceeds the preset number threshold, the dynamic recognition unit determines that the safety helmet wearer is in a moving state; otherwise, it is determined that the safety helmet wearer is in a stationary state; A5, if the dynamic recognition unit determines that the safety helmet wearer is in a moving state, the dynamic recognition unit maintains the current low power consumption sampling logic; If the dynamic identification unit judges that the safety helmet wearer is in a stationary state and the duration reaches the preset duration, the high-power consumption state is entered.

5. A hard hat monitoring system for use in a downhole oil environment according to claim 4, characterized in that, The method for obtaining a representative pixel color unit comprises: Z1, divide the image into m*n equal-sized grid regions, each grid numbered as and each grid region includes K pixel points; Z2, selecting a plurality of preset position pixel points in each grid region; The main color channel value of the representative sample pixel is obtained, and the value of the selected sample pixel is simply averaged in integer form to obtain the representative pixel color unit of the grid, denoted as: ; where N is the number of samples; is the color channel value of the kth selected pixel; Z3, finally, each grid region corresponds to a representative pixel color cell value; All grid areas The values ​​form a two-dimensional array and change over time to form a time series.

6. A hard hat monitoring system for use in a downhole oil environment according to claim 1, characterized in that, The processing platform constructs a time sequence based on the collected environmental data in each work area, determines whether there is a growth trend by analyzing the change trend of the environmental data in the same area, and further determines whether there is a propagation path of environmental abnormalities based on the change gradient and time delay relationship of the environmental data between adjacent work areas.

7. A safety hat monitoring system for use in a downhole oil environment according to claim 6, characterized in that, The method for determining whether there is a propagation path by the processing platform comprises: S1, single grid region time sequence trend analysis: for the same work area In this case, the environmental data V acquired at a plurality of successive points in time , ..., are used to construct a time series ; Processing platform for time series Performing trend fitting processing; If the time series satisfies a preset change trend threshold, it is determined that the corresponding grid region has an abnormal growth trend. S2, adjacent area propagation trend analysis: The calculation module constructs a synchronous value vector of the environmental data of the adjacent work area at the same time node, and compares the current value and the previous time value of the grid region to determine whether there is: The size of the environmental data value presents a "spatial gradient"; At the time point, the outer area changes later than the inner area; If the above conditions are met at the same time, it is judged that the environmental data has a diffusion trend from the core area to the outside; S3, space-time mapping relationship and path simulation; The judgment result in S2 is written into the three-dimensional model M[i][j][t] to form a multi-dimensional abnormal change graph based on the area and time; The propagation path graph is constructed according to the maximum spatial gradient path and the shortest time delay path.

8. A safety hat monitoring system for use in a downhole oil environment according to claim 7, characterized in that, The processing platform adjusts the propagation path of the environmental data according to the type of the environmental data and the triggering condition thereof.

Images