A big data-based safety monitoring method and system for a cement production line

Abstract

This invention discloses a method and system for safety monitoring of cement production lines based on big data. The method includes: acquiring alarm information containing timestamps and location stamps; determining whether the personnel alarm information and sensor alarm information are in the same spatial grid based on their location stamps; if so, labeling the alarm information as target alarm information; analyzing the timestamp difference between the personnel alarm information and sensor alarm information in the target alarm information, and taking the absolute value of the difference; if the absolute value is less than or equal to the current time threshold, determining the target alarm information as real alarm information; uploading the real alarm information to a cloud database; adding the obtained absolute value of the timestamp difference as a new sample to a sliding window; updating the time threshold according to a preset statistical method; the sliding window adopts a first-in-first-out update mechanism. This invention overcomes the shortcomings of single information sources being susceptible to environmental interference or human error leading to high false alarm rates.

Description

Technical Field

[0001] This invention relates to the field of safe production, specifically to a method and system for safety monitoring of cement production lines based on big data. Background Technology

[0002] Cement production lines are characterized by long processes, widely distributed equipment, and high-temperature, dusty operating environments, leading to frequent safety hazards such as preheater blockage, excessive mill vibration, belt tearing, and high-temperature gas leaks. With the development of industrial internet technology, safety monitoring systems based on big data and sensor networks have been widely applied in the cement production field. These systems aim to achieve early warning of accidents by collecting environmental parameters (such as temperature, vibration, and carbon monoxide) in real time and combining them with video monitoring.

[0003] Currently, existing safety monitoring methods for cement production lines typically employ a "threshold trigger + location reporting" model. This means that when sensor values ​​exceed preset thresholds or personnel press an alarm button, the system generates an alarm record containing time and location information. This approach has the following shortcomings: most existing systems handle sensor alarms or personnel alarms in isolation, but due to the harsh environment of cement production lines, a single sensor is highly prone to false alarms. Furthermore, on-site personnel inevitably make accidental triggers or provide unclear alarm information. In summary, traditional single-source verification mechanisms cannot guarantee the accuracy of safety warnings and are prone to false alarms. Summary of the Invention

[0004] This invention provides a method for safety monitoring of cement production lines based on big data, which at least partially solves the above-mentioned technical problems.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a method for safety monitoring of cement production lines based on big data, comprising: Obtain alarm information containing timestamps and location stamps, including personnel alarm information and sensor alarm information; Based on the location stamps of the personnel alarm information and the sensor alarm information, it is determined whether they are in the same spatial grid. If so, the alarm information is marked as the target alarm information. Analyze the timestamp difference between the personnel alarm information and the sensor alarm information in the target alarm information, and take the absolute value of the difference. If the absolute value is less than or equal to the current time threshold, the target alarm information is determined to be a real alarm information. The real alarm information is uploaded to the cloud database. The absolute value of the timestamp difference obtained this time is added to the sliding window as a new sample. The time threshold is updated according to the preset statistical method. The sliding window adopts a first-in-first-out update mechanism.

[0006] In conjunction with the first aspect above, in one possible implementation, the location stamp is generated by a unique code of a spatial grid, and the method for dividing the spatial grid is as follows: The entire cement production line is divided into multiple basic areas; The historical security event density of each basic grid within a preset statistical period is retrieved from the cloud database. The historical security event density is the number of historical security events that occur within the basic grid per unit time. The base area with a historical security event density greater than or equal to a first density threshold is designated as a refined area, and the refined area is divided into multiple refined grids; The base region with a historical security event density less than or equal to the second density threshold is designated as the coarsened region, and the coarsened region is divided into multiple coarsened grids; The basic region with a historical security event density between a first density threshold and a second density threshold is designated as a standard region. The standard region is then divided into multiple standard grids, where the first density threshold is greater than the second density threshold, and the side length of the standard grid is between that of a refined grid and a coarse grid.

[0007] In conjunction with the first aspect above, in one possible implementation, the personnel alarm information includes communication information and limb information, wherein the limb information is collected by a camera device and identified and confirmed by a central processing unit.

[0008] In conjunction with the first aspect above, in one possible implementation, the limb information being collected by a camera device and identified and confirmed by a central processing unit includes: The system acquires on-site personnel limb information collected by camera equipment, extracts features from the acquired limb information, extracts posture features or directional movements from the limb information, and compares and matches them with a preset action feature database. The system analyzes and determines whether the limb information represents the person pointing to their current spatial grid. If it is determined that the person is pointing to the current spatial grid, the location stamp of the alarm information remains unchanged. If it is determined that the limb information does not point to the current spatial grid, the spatial direction vector of the intention of the person is calculated by combining the spatial position code and shooting angle parameters of the camera device that collected the limb information. The spatial direction vector is compared with the spatial grid model to determine the target spatial grid to which the limb information points. The location stamp of the personnel alarm information associated with the limb information is then updated to the location stamp of the target spatial grid.

[0009] In conjunction with the first aspect above, in one possible implementation, when it is determined that the personnel alarm information and the sensor alarm information are not in the same spatial grid, the spatial grid corresponding to the location stamp in the sensor alarm information is designated as the spatial grid to be confirmed. At the same time, communication information is sent to nearby personnel so that they can go to the corresponding spatial grid to perform detection and provide feedback on the detection results. The detection results include confirmation of alarm information and alarm cancellation information.

[0010] In conjunction with the first aspect above, in one possible implementation, each sensor is configured with an initial confidence level and a defined extreme value, the extreme values ​​of which include a minimum confidence level and a maximum confidence level. When the detection result indicates that the alarm has been cleared, the number of false alarms for the corresponding sensor is accumulated, and the confidence level of the sensor is adjusted according to a preset rule within a fixed period. The preset rule is as follows: The false alarm rate of each sensor within a fixed time period is calculated as follows: False alarm rate = Number of false alarms within the period / Total number of triggers within the period. If the false alarm rate of the sensor is greater than or equal to the first false alarm rate threshold, the confidence level of the sensor is reduced by a fixed adjustment step size. When the confidence level is reduced to the minimum confidence level, a sensor replacement prompt message is triggered. If the false alarm rate of the sensor is less than or equal to the second false alarm rate threshold, the confidence level of the sensor is increased by a fixed adjustment step size. When the confidence level is increased to the maximum confidence level, the confidence level value is kept unchanged. If the false alarm rate of the sensor is between the first false alarm rate threshold and the second false alarm rate threshold, then the confidence level of the sensor remains unchanged. When the false alarm rate of a sensor exceeds the first false alarm rate threshold for X consecutive cycles, the sensor is calibrated as a low-confidence sensor. If a sensor marked as low confidence recovers its false alarm rate to the second false alarm rate threshold within the subsequent Y consecutive cycles, its confidence level is restored to the initial value, and the low confidence mark is cleared. Both X and Y are integers greater than or equal to one.

[0011] In conjunction with the first aspect mentioned above, in one possible implementation, when multiple alarm messages exist simultaneously, the alarm priorities are arranged in descending order of sensor confidence.

[0012] In conjunction with the first aspect above, in one possible implementation, the preset statistical method is a kernel density estimation method, which fits the absolute value of the timestamp difference within the sliding window and takes the horizontal coordinate corresponding to the main peak of the probability density curve as the current time threshold.

[0013] In conjunction with the first aspect mentioned above, one possible implementation also includes: When the target alarm information is determined to be a real alarm information, the associated information group is retrieved based on the timestamp and location stamp in the target alarm information. The information group includes instruction information and operation information. At the same time, a security event file with a unique identifier is generated for the real security event, and the security event file and the retrieved information group are stored together in the cloud database.

[0014] Secondly, the present invention provides a cement production line safety monitoring system based on big data, comprising: The acquisition module is used to acquire alarm information containing timestamps and location stamps, including personnel alarm information and sensor alarm information; The judgment module is used to determine whether the personnel alarm information and the sensor alarm information are in the same spatial grid based on the location stamps. If so, the alarm information is marked as the target alarm information. The analysis module is used to analyze the timestamp difference between personnel alarm information and sensor alarm information in the target alarm information, and take the absolute value of the difference. If the absolute value is less than or equal to the current time threshold, the target alarm information is determined to be a real alarm information. The processing module is used to upload the real alarm information to the cloud database, add the absolute value of the timestamp difference obtained this time as a new sample to the sliding window, and update the time threshold according to a preset statistical method. The sliding window adopts a first-in-first-out update mechanism.

[0015] Thirdly, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the above-mentioned cement production line safety monitoring method based on big data.

[0016] Fourthly, the present invention provides a computer-readable storage medium comprising a computer program, which, when executed on an electronic device, causes the electronic device to perform the steps of the above-described big data-based cement production line safety monitoring method: Obtain alarm information containing timestamps and location stamps, including personnel alarm information and sensor alarm information; Based on the location stamps of the personnel alarm information and the sensor alarm information, it is determined whether they are in the same spatial grid. If so, the alarm information is marked as the target alarm information. Analyze the timestamp difference between the personnel alarm information and the sensor alarm information in the target alarm information, and take the absolute value of the difference. If the absolute value is less than or equal to the current time threshold, the target alarm information is determined to be a real alarm information. The real alarm information is uploaded to the cloud database. The absolute value of the timestamp difference obtained this time is added to the sliding window as a new sample. The time threshold is updated according to the preset statistical method. The sliding window adopts a first-in-first-out update mechanism.

[0017] In summary, the big data-based cement production line safety monitoring method provided by this invention overcomes the shortcomings of single information sources, which are susceptible to environmental interference or human error leading to high false alarm rates. Based on timestamps and location stamps, it determines a real alarm only when human and machine alarms are highly correlated in both time and space dimensions, thus reducing the probability of false alarms and improving alarm reliability compared to existing technologies. Furthermore, by adding the absolute value of the time difference of real alarms as a new sample to the sliding window and dynamically updating the time threshold according to a preset statistical method, adaptive adjustment is achieved, greatly enhancing the system's robustness in the complex industrial environment of cement production. Additionally, the non-uniform spatial grid partitioning mechanism enables focused monitoring of high-risk areas while reducing resource consumption in low-risk areas. While ensuring monitoring accuracy, it also achieves rational allocation of resources. By recognizing the posture characteristics or directional movements of personnel through the central processing unit, when it is determined that the alarm is not pointing to the personnel's current grid, it calculates the spatial direction vector by combining the camera equipment parameters to determine the target area, breaking the limitation of existing technologies that can only locate alarms and improving the effectiveness of safety monitoring. By introducing a dynamic confidence management mechanism for sensors, it achieves quantitative assessment of the health status of sensors, which not only avoids the interference of low-quality data on decision-making, but also provides convenience for operation and maintenance personnel and significantly reduces maintenance costs. By setting up safety event files and retrieving related information groups based on timestamps and spatial stamps, it provides structured data support for subsequent accident analysis, realizing the digitalization and traceability of safety management. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0019] To better understand the present invention and its beneficial effects, the following description will be made in conjunction with the accompanying drawings, wherein the same reference numerals denote the same parts in the following description.

[0020] Figure 1 This is a flowchart of a cement production line safety monitoring method based on big data provided in an exemplary embodiment of the present invention.

[0021] Figure 2This is a schematic diagram of the structure of a cement production line safety monitoring system based on big data provided in an exemplary embodiment of the present invention.

[0022] Figure 3 This is a schematic diagram of the structure of a computer device provided in an exemplary embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the protection scope of the present invention.

[0024] Figure 1 This is an exemplary embodiment of the cement production line safety monitoring method based on big data provided by the present invention. The process steps of this method are merely one possible implementation of the present invention. Figure 1 As shown, the cement production line safety monitoring method based on big data provided in this embodiment of the invention includes the following steps: S10, acquire alarm information including timestamp and location stamp, the alarm information including personnel alarm information and sensor alarm information.

[0025] In this embodiment, the timestamp and location stamp record the time and spatial parameters of the alarm, respectively, to characterize when and in what area the cement production line has a problem. The timestamp can be determined by the alarm time of the sensor or personnel, while for the location stamp, this embodiment introduces a spatial grid model.

[0026] S20, determine whether the personnel alarm information and the sensor alarm information are in the same spatial grid based on the location stamps. If so, mark the alarm information as the target alarm information.

[0027] S30, Analyze the timestamp difference between the personnel alarm information and the sensor alarm information in the target alarm information, and take the absolute value of the difference. If the absolute value is less than or equal to the current time threshold, then determine that the target alarm information is a real alarm information.

[0028] In this embodiment, the time before and after the alarm by personnel and the alarm by the sensor is uncontrollable. Therefore, the absolute value of their difference is taken as the comparison data with the time threshold. For example, if the alarm time of personnel is 10:30:13 and the alarm time of the sensor is 10:30:10, then the absolute value of the difference between the two is 3 seconds. Assuming the time threshold is 3.5 seconds, the target alarm information is determined to be a real alarm information.

[0029] S40, the real alarm information is uploaded to the cloud database, the absolute value of the timestamp difference obtained this time is added to the sliding window as a new sample, and the time threshold is updated according to the preset statistical method. The sliding window adopts a first-in-first-out update mechanism.

[0030] In this embodiment, the sliding window is a window with a fixed length that stores multiple timestamp differences. The first-in-first-out update mechanism means that each time new data enters the sliding window, the oldest data in the sliding window is automatically discarded, so that the sliding window always stores the timestamp differences of the most recent N real security events.

[0031] As an example, assuming the sliding window length N=20, and a timestamp difference data is generated for each real safety event occurring in the cement production line, the update process of the sliding window can be represented as follows: Window 1 (Events 1-20): [Data 1, Data 2 ... Data 20] Window 2 (Event 2-21): [Data 2, Data 3 ... Data 21] Window 3 (Event 3-22): [Data 3, Data 4 ... Data 22] ... And so on.

[0032] Based on historical security event data stored in the cloud database, the sliding window is constantly updated. For each updated window, the system recalculates the time threshold according to a preset statistical method, thereby dynamically updating the time threshold and achieving adaptive adjustment of the time threshold.

[0033] It should be noted that the above values ​​are for illustrative purposes only, and can be flexibly configured according to the specific needs of the cement production line in actual applications.

[0034] In summary, the embodiments of the present invention overcome the shortcomings of single signal sources being susceptible to environmental interference or human error leading to high false alarm rates. Based on timestamps and location stamps, the system is only considered a real alarm when human and machine alarms are highly correlated in both time and space dimensions. Compared with existing technologies, this reduces the probability of false alarms and improves the reliability of alarms. At the same time, by adding the absolute value of the time difference of real alarms as a new sample to the sliding window and dynamically updating the time threshold according to a preset statistical method, adaptive adjustment is achieved, which greatly enhances the robustness of the system in complex industrial environments such as cement production.

[0035] As mentioned above, this embodiment introduces a spatial grid model for location stamps. In existing technologies, the spatial grid is generally set uniformly, but this setting has a problem: unreasonable resource allocation. Based on this, in one embodiment of the present invention, the location stamp is generated by the unique code of the spatial grid, and the method for dividing the spatial grid is as follows: The entire cement production line is divided into multiple basic areas; The historical security event density of each basic grid within a preset statistical period is retrieved from the cloud database. The historical security event density is the number of historical security events that occur within the basic grid per unit time. The base area with a historical security event density greater than or equal to a first density threshold is designated as a refined area, and the refined area is divided into multiple refined grids; The base region with a historical security event density less than or equal to the second density threshold is designated as the coarsened region, and the coarsened region is divided into multiple coarsened grids; The basic region with a historical security event density between a first density threshold and a second density threshold is designated as a standard region. The standard region is then divided into multiple standard grids, where the first density threshold is greater than the second density threshold, and the side length of the standard grid is between that of a refined grid and a coarse grid.

[0036] As an example, assuming the first density threshold is 10.5 times / month and the second density threshold is 5.5 times / month, the historical safety event density of base area A is 13 times / month, indicating a high frequency of safety events and classifying it as a high-risk area. Base area A is therefore classified as a refined area, with a grid size of 2m*2m*2m. The historical safety event density of base area B is 8 times / month, indicating a normal frequency of safety events and classifying it as a standard area, with a grid size of 3m*3m*3m. The historical safety event density of base area C is 4 times / month, indicating a relatively high frequency of safety events and classifying it as a low-risk area, and classifying it as a coarse area, with a grid size of 4m*4m*4m. It should be noted that the above values ​​are only examples, and in actual applications, they can be flexibly configured according to the specific needs of the cement production line.

[0037] In summary, the embodiments of the present invention employ a non-uniform spatial grid division mechanism, which enables key monitoring of high-risk areas and reduces resource consumption in low-risk areas, achieving rational allocation of resources while ensuring monitoring accuracy.

[0038] In the actual production process of cement production lines, there are sometimes problems with personnel communication equipment failure, which can lead to the failure to send alarm information in a timely manner. Based on this, in one embodiment of the present invention, the personnel alarm information includes communication information and limb information, wherein the limb information is collected by a camera device and identified and confirmed by a central processing unit.

[0039] In this embodiment, when the communication equipment of on-site personnel malfunctions, information can be transmitted through body language to ensure that alarm information can be delivered in a timely manner. As for the identification and confirmation by the central processing unit, it needs to be achieved with the help of machine vision technology, which is existing technology and will not be described in detail here.

[0040] Furthermore, in the above embodiments, the limb information being collected by the camera device and identified and confirmed by the central processing unit includes: The system acquires on-site personnel limb information collected by camera equipment, extracts features from the acquired limb information, extracts posture features or directional movements from the limb information, and compares and matches them with a preset action feature database. The system analyzes and determines whether the limb information represents the person pointing to their current spatial grid. If it is determined that the person is pointing to the current spatial grid, the location stamp of the alarm information remains unchanged. If it is determined that the limb information does not point to the current spatial grid, the spatial direction vector of the intention of the person is calculated by combining the spatial position code and shooting angle parameters of the camera device that collected the limb information. The spatial direction vector is compared with the spatial grid model to determine the target spatial grid to which the limb information points. The location stamp of the personnel alarm information associated with the limb information is then updated to the location stamp of the target spatial grid.

[0041] In actual production processes, situations sometimes arise where the "alarm area is not the area where the personnel are located." For example, an inspection worker may be in area A, but because there is no camera in area A or the camera is malfunctioning, they can only go to area B to trigger the alarm. In this case, the personnel can transmit information through specific postures or actions. The central processing unit has a preset "raising a hand and clenching a fist" gesture to indicate that "the alarm area is not the person's location." When the system detects that the personnel have made the "raising a hand and clenching a fist" gesture, it determines that the limb information is not pointing to area A. The system further analyzes the personnel's posture characteristics or directional actions and detects that the personnel have also made the action of "pointing their arm horizontally in the northeast direction." Combining the position encoding of the camera equipment itself, as well as the pixel coordinates and angle of the arm pointing in the image, the system calculates the spatial direction vector and compares it with the spatial grid model to finally determine the target spatial grid corresponding to area B.

[0042] In summary, the embodiments of the present invention identify the posture characteristics or directional movements of personnel through a central processing unit. When it is determined that the alarm is not directed at the grid where the personnel are currently located, the spatial direction vector is calculated in combination with the parameters of the camera equipment to determine the target area. This breaks through the limitation of the prior art that can only locate alarms and improves the safety monitoring effect.

[0043] In one embodiment of the present invention, each sensor is configured with an initial confidence level and a defined extreme value, wherein the extreme values ​​of the confidence level include a minimum extreme value and a maximum extreme value. When the detection result indicates that the alarm has been cleared, the number of false alarms for the corresponding sensor is accumulated, and the confidence level of the sensor is adjusted according to a preset rule within a fixed period. The preset rule is as follows: The false alarm rate of each sensor within a fixed time period is calculated as follows: False alarm rate = Number of false alarms within the period / Total number of triggers within the period. If the false alarm rate of the sensor is greater than or equal to the first false alarm rate threshold, the confidence level of the sensor is reduced by a fixed adjustment step size. When the confidence level is reduced to the minimum confidence level, a sensor replacement prompt message is triggered. If the false alarm rate of the sensor is less than or equal to the second false alarm rate threshold, the confidence level of the sensor is increased by a fixed adjustment step size. When the confidence level is increased to the maximum confidence level, the confidence level value is kept unchanged. If the false alarm rate of the sensor is between the first false alarm rate threshold and the second false alarm rate threshold, then the confidence level of the sensor remains unchanged. When the false alarm rate of a sensor exceeds the first false alarm rate threshold for X consecutive cycles, the sensor is calibrated as a low-confidence sensor. If a sensor marked as low confidence recovers its false alarm rate to the second false alarm rate threshold within the subsequent Y consecutive cycles, its confidence level is restored to the initial value, and the low confidence mark is cleared. Both X and Y are integers greater than or equal to one.

[0044] In this embodiment, the initial confidence level is a preset value. The specific value can be flexibly set according to the needs. No specific example is given here. Similarly, the fixed time period, threshold, and X and Y are also preset values. The "period" mentioned later refers to this preset fixed event period.

[0045] In summary, by introducing a dynamic confidence management mechanism for sensors, this invention enables a quantitative assessment of sensor health status. This not only avoids interference from low-quality data in decision-making but also provides convenience for maintenance personnel and significantly reduces maintenance costs.

[0046] In one embodiment of the present invention, when multiple alarm messages exist simultaneously, the alarm priorities are arranged in descending order of sensor confidence.

[0047] In this embodiment, as mentioned above, the dynamic confidence management mechanism of the sensor enables the sorting of multiple alarm messages in a definite manner, which is more scientific and reasonable and improves the processing efficiency of alarm messages.

[0048] In one embodiment of the present invention, the preset statistical method is a kernel density estimation method, which fits the absolute value of the timestamp difference within the sliding window and takes the horizontal coordinate corresponding to the main peak of the probability density curve as the current time threshold.

[0049] In this embodiment, the specific execution process of the kernel density estimation method is as follows: A Gaussian kernel function is used, combined with a fixed bandwidth or adaptive bandwidth strategy, to perform kernel density fitting on the sample set, generating a continuous probability density curve; all peaks on the curve are identified, the peak with the largest amplitude is selected as the principal peak, and the abscissa value corresponding to the principal peak is determined as the updated time threshold. In actual cement production line conditions, the absolute value of timestamp differences is usually discrete and unevenly distributed. The kernel density estimation method can adaptively fit the true distribution of the data, effectively improving the accuracy of the time threshold setting. Of course, in practical applications, other preset statistical methods, such as the average value method, can also be used.

[0050] In one embodiment of the present invention, the method further includes: S50, when the target alarm information is determined to be a real alarm information, the associated information group is retrieved according to the timestamp and location stamp in the target alarm information. The information group includes instruction information and operation information. At the same time, a security event file with a unique identifier is generated for the real security event, and the security event file and the retrieved information group are stored together in the cloud database.

[0051] In this embodiment, the associated information group refers to the information cluster that matches or is adjacent to the target alarm information in the time dimension and spatial dimension. Since the present invention sets the two concepts of timestamp and location stamp, the association in the time dimension can be realized through timestamp, while the association in the spatial dimension can be realized through location stamp.

[0052] This invention provides structured data support for subsequent accident analysis by setting up security event files and retrieving associated information groups based on timestamps and spatial stamps, thereby achieving digitalization and traceability of security management.

[0053] Figure 2 This is an exemplary embodiment of a cement production line safety monitoring system based on big data, which is merely one possible implementation of the present invention. Figure 2 As shown, the cement production line safety monitoring system based on big data provided in this embodiment of the invention includes: The acquisition module is used to acquire alarm information containing timestamps and location stamps, including personnel alarm information and sensor alarm information; The judgment module is used to determine whether the personnel alarm information and the sensor alarm information are in the same spatial grid based on the location stamps. If so, the alarm information is marked as the target alarm information. The analysis module is used to analyze the timestamp difference between personnel alarm information and sensor alarm information in the target alarm information, and take the absolute value of the difference. If the absolute value is less than or equal to the current time threshold, the target alarm information is determined to be a real alarm information. The processing module is used to upload the real alarm information to the cloud database, add the absolute value of the timestamp difference obtained this time as a new sample to the sliding window, and update the time threshold according to a preset statistical method. The sliding window adopts a first-in-first-out update mechanism.

[0054] Figure 3 This is a schematic diagram of the structure of the computer device provided in an embodiment of the present invention, such as... Figure 3 As shown, the computer device includes a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the big data-based cement production line safety monitoring method described in the above embodiments.

[0055] Furthermore, those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0056] To this end, embodiments of the present invention provide a computer-readable storage medium storing multiple computer programs. These computer programs can be loaded by a processor to execute the big data-based cement production line safety monitoring and control method provided in these embodiments. The computer program can execute the following steps of the big data-based cement production line safety monitoring method: Obtain alarm information containing timestamps and location stamps, including personnel alarm information and sensor alarm information; Based on the location stamps of the personnel alarm information and the sensor alarm information, it is determined whether they are in the same spatial grid. If so, the alarm information is marked as the target alarm information. Analyze the timestamp difference between the personnel alarm information and the sensor alarm information in the target alarm information, and take the absolute value of the difference. If the absolute value is less than or equal to the current time threshold, the target alarm information is determined to be a real alarm information. The real alarm information is uploaded to the cloud database. The absolute value of the timestamp difference obtained this time is added to the sliding window as a new sample. The time threshold is updated according to the preset statistical method. The sliding window adopts a first-in-first-out update mechanism.

[0057] The computer program instructions described in this application embodiment may be stored in a computer-readable storage medium capable of directing a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, the instruction means being implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0058] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0059] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, like read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0060] Computer-readable media include both permanent and non-permanent, removable and non-removable media, which can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient media, such as modulated communication signals and carrier waves.

[0061] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0062] The embodiments, implementation methods, and related technical features of this invention can be combined and substituted for each other without conflict.

[0063] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for safety monitoring of cement production lines based on big data, characterized in that, include: Obtain alarm information containing timestamps and location stamps, including personnel alarm information and sensor alarm information; Based on the location stamps of the personnel alarm information and the sensor alarm information, it is determined whether the two are in the same spatial grid. If so, the alarm information is marked as the target alarm information. Analyze the timestamp difference between the personnel alarm information and the sensor alarm information in the target alarm information, and take the absolute value of the difference. If the absolute value is less than or equal to the current time threshold, the target alarm information is determined to be a real alarm information. The real alarm information is uploaded to the cloud database. The absolute value of the timestamp difference obtained this time is added to the sliding window as a new sample. The time threshold is updated according to the preset statistical method. The sliding window adopts a first-in-first-out update mechanism.

2. The method for safety monitoring of cement production lines based on big data according to claim 1, characterized in that, The location stamp is generated by a unique code of a spatial grid, and the spatial grid is divided as follows: The entire cement production line is divided into multiple basic areas; The historical security event density of each basic grid within a preset statistical period is retrieved from the cloud database. The historical security event density is the number of historical security events that occur within the basic grid per unit time. The base area with a historical security event density greater than or equal to a first density threshold is designated as a refined area, and the refined area is divided into multiple refined grids; The base region with a historical security event density less than or equal to the second density threshold is designated as the coarsened region, and the coarsened region is divided into multiple coarsened grids; The basic region with a historical security event density between a first density threshold and a second density threshold is designated as a standard region. The standard region is then divided into multiple standard grids, where the first density threshold is greater than the second density threshold, and the side length of the standard grid is between that of a refined grid and a coarse grid.

3. The method for safety monitoring of cement production lines based on big data according to claim 2, characterized in that, The personnel alarm information includes communication information and limb information, wherein the limb information is collected by the camera equipment and identified and confirmed by the central processing unit.

4. The method for safety monitoring of cement production lines based on big data according to claim 3, characterized in that, The limb information, collected by the camera device and identified and confirmed by the central processing unit, includes: The system acquires on-site personnel limb information collected by camera equipment, extracts features from the acquired limb information, extracts posture features or directional movements from the limb information, and compares and matches them with a preset action feature database. The system analyzes and determines whether the limb information represents the person pointing to their current spatial grid. If it is determined that the person is pointing to the current spatial grid, the location stamp of the person's alarm information remains unchanged. If it is determined that the limb information does not point to the current spatial grid, the spatial direction vector of the intention of the person is calculated by combining the spatial position code and shooting angle parameters of the camera device that collected the limb information. The spatial direction vector is compared with the spatial grid model to determine the target spatial grid to which the limb information points. The location stamp of the personnel alarm information associated with the limb information is then updated to the location stamp of the target spatial grid.

5. The method for safety monitoring of cement production lines based on big data according to claim 4, characterized in that, When it is determined that the personnel alarm information and the sensor alarm information are not in the same spatial grid, the spatial grid corresponding to the location stamp in the sensor alarm information is designated as the spatial grid to be confirmed. At the same time, communication information is sent to nearby personnel so that they can go to the corresponding spatial grid to perform detection and report the detection results. The detection results include confirmation of alarm information and alarm cancellation information.

6. The method for safety monitoring of cement production lines based on big data according to claim 1, characterized in that, Each sensor is configured with an initial confidence level and a defined extreme value, which includes a minimum confidence level and a maximum confidence level. When the detection result indicates that the alarm has been cleared, the number of false alarms for the corresponding sensor is accumulated, and the confidence level of the sensor is adjusted according to a preset rule within a fixed period. The preset rule is as follows: The false alarm rate of each sensor within a fixed time period is calculated as follows: False alarm rate = Number of false alarms within the period / Total number of triggers within the period. If the false alarm rate of the sensor is greater than or equal to the first false alarm rate threshold, the confidence level of the sensor is reduced by a fixed adjustment step size. When the confidence level is reduced to the minimum confidence level, a sensor replacement prompt message is triggered. If the false alarm rate of the sensor is less than or equal to the second false alarm rate threshold, the confidence level of the sensor is increased by a fixed adjustment step size. When the confidence level is increased to the maximum confidence level, the confidence level value is kept unchanged. If the false alarm rate of the sensor is between the first false alarm rate threshold and the second false alarm rate threshold, then the confidence level of the sensor remains unchanged. When the false alarm rate of a sensor exceeds the first false alarm rate threshold for X consecutive cycles, the sensor is calibrated as a low-confidence sensor. If a sensor marked as low confidence recovers its false alarm rate to the second false alarm rate threshold within the subsequent Y consecutive cycles, its confidence level is restored to the initial value, and the low confidence mark is cleared. Both X and Y are integers greater than or equal to one.

7. The method for safety monitoring of cement production lines based on big data according to claim 6, characterized in that, When multiple alarm messages exist simultaneously, the alarm priorities are arranged in descending order of sensor confidence level.

8. The method for safety monitoring of cement production lines based on big data according to claim 1, characterized in that, The preset statistical method is a kernel density estimation method, which fits the absolute value of the timestamp difference within the sliding window and takes the horizontal coordinate corresponding to the main peak of the probability density curve as the current time threshold.

9. The method for safety monitoring of cement production lines based on big data according to claim 1, characterized in that, Also includes: When the target alarm information is determined to be a real alarm information, the associated information group is retrieved based on the timestamp and location stamp in the target alarm information. The information group includes instruction information and operation information. At the same time, a security event file with a unique identifier is generated for the real security event, and the security event file and the retrieved information group are stored together in the cloud database.

10. A safety monitoring system for a cement production line based on big data, characterized in that, include: The acquisition module is used to acquire alarm information containing timestamps and location stamps, including personnel alarm information and sensor alarm information; The judgment module is used to determine whether the personnel alarm information and the sensor alarm information are in the same spatial grid based on the location stamps. If so, the alarm information is marked as the target alarm information. The analysis module is used to analyze the timestamp difference between personnel alarm information and sensor alarm information in the target alarm information, and take the absolute value of the difference. If the absolute value is less than or equal to the current time threshold, the target alarm information is determined to be a real alarm information. The processing module is used to upload the real alarm information to the cloud database, add the absolute value of the timestamp difference obtained this time as a new sample to the sliding window, and update the time threshold according to a preset statistical method. The sliding window adopts a first-in-first-out update mechanism.

Images