Methods, systems, equipment, and readable storage media for identifying unlicensed operations in thermal power plants.

Abstract

This invention belongs to the field of power plant information technology and safety supervision, and discloses a method, system, equipment, and readable storage medium for identifying unlicensed operations in thermal power plants. The method for identifying unlicensed operations in thermal power plants includes the following steps: acquiring monitoring image data of the thermal power plant production area through cameras installed in the production area, inputting the data into a pre-trained behavior detection model for detection, and obtaining the probability P of personnel operation behavior appearing in the monitoring image data. 作业 And the corresponding location POS where the operation occurs; the probability P 作业 The results of comparing the results with a preset threshold P0 are used to identify unlicensed operations in thermal power plants. This invention can identify and monitor unlicensed operations, unauthorized expansion of work areas, and other violations, significantly improving the real-time performance, accuracy, and coverage of monitoring.

Description

Technical Field

[0001] This invention belongs to the field of power plant information technology and safety supervision technology, and specifically relates to a method, system, equipment and readable storage medium for identifying unlicensed operations in thermal power plants. Background Technology

[0002] Thermal power plants have formulated a complete set of safety production management systems and requirements in accordance with relevant national laws and regulations or industry standards. These systems are used to strengthen on-site management of power plants, regulate the behavior of various personnel, and ensure the safety of personnel, power grid, and equipment.

[0003] Currently, during actual maintenance and repair work in thermal power plants, there are often multiple personnel and multiple areas involved in on-site operations. Maintenance personnel, especially outsourced maintenance teams, have a high degree of mobility and arbitrariness on the production site, making safety management and supervision difficult. In particular, for some relatively simple maintenance tasks, some personnel take chances and frequently engage in violations such as working without a permit or expanding the scope of work without authorization.

[0004] Given the above situation, it is necessary to supervise violations such as operating without a permit and unauthorized expansion of the work scope. Currently, manual inspection is used. However, manual inspection is labor-intensive, and safety supervisors and managers are not on the front line of business risk control. They cannot promptly detect and track dynamic risks on site, and risks during the operation process are difficult to effectively supervise and control. This results in a lack of effective safety control plans in the implementation of management systems and requirements, and the inability to detect violations in a timely manner, which may lead to serious consequences such as personal injury, equipment damage, etc. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, equipment, and readable storage medium for identifying unlicensed operations in thermal power plants, thereby solving one or more of the aforementioned technical problems. The technical solution provided by this invention can identify and monitor unlicensed operations, unauthorized expansion of the work scope, and other violations, significantly improving the real-time performance, accuracy, and coverage of monitoring.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] The first aspect of this invention provides a method for identifying unlicensed operations in thermal power plants, comprising the following steps:

[0008] Surveillance video data of the power plant's production area is acquired by cameras installed in the plant's production area and input into a pre-trained behavior detection model for detection, thereby obtaining the probability P of human work behavior appearing in the surveillance video data. 作业 and the corresponding POS location where the operation occurred;

[0009] The probability P 作业Compare with the preset threshold P0, and obtain the identification result of unlicensed operation in the thermal power plant based on the comparison result; where

[0010] If P 作业 ≥P0, then P 无票作业 =P 作业 α 定位 η 持票 β 缺陷 ; If P 作业 <P0, then P 无票作业 =0; In the formula, 0<P0<1, P 无票作业 is the probability of unlicensed operation occurrence determined finally; α 定位 is the positioning improvement coefficient; η 持票 is the ticket-holding coefficient; β 缺陷 is the defect improvement coefficient.

[0011] A further improvement of the method of the present invention is that the behavior detection model adopts any one of C3D, P3D, I3D, SlowFast and Two-stream.

[0012] A further improvement of the method of the present invention is that the training acquisition step of the trained behavior detection model includes:

[0013] On the basis of the pre-trained model, weight adjustment is carried out by using the actual on-site video sample data to obtain the finally trained behavior detection model;

[0014] Among them, the actual on-site video sample data is the sample data obtained by intercepting and marking the videos with operation behaviors and the videos without operation behaviors, which are shot in the production area of the thermal power plant by monitoring cameras.

[0015] A further improvement of the method of the present invention is that the acquisition step of the positioning improvement coefficient includes:

[0016] The monitoring image data is obtained by the cameras installed in the production area of the thermal power plant;

[0017] Obtain the angle, installation height and installation position coordinates of the camera; input the angle, installation height of the camera and the position POS where the operation behavior occurs into the pre-trained relative position calculation model of personnel to obtain the relative coordinates of the operator and the camera; according to the relative coordinates of the operator and the camera and the installation position coordinates of the camera, obtain the three-dimensional position coordinate range A of the operator; among them, the three-dimensional position coordinate range A includes the position center point A0 and the preset activity radius R;

[0018] The personnel set M is obtained by using a personnel positioning device and located within the three-dimensional position coordinate range A. Then, a maintenance team personnel set N is selected from personnel set M. The maximum value t of the dwell time of all personnel in the maintenance team personnel set N within the three-dimensional position coordinate range A is obtained. m The positioning lift coefficient α is obtained by comparing it with the preset reference time t0. 定位 The calculation formula is α 定位 =tanh(t) m / t0-1)+1.

[0019] A further improvement to the method of the present invention is that the step of obtaining the voting coefficient includes:

[0020] Obtain the set P of work members included in the set of work tickets that have commenced work within the three-dimensional position coordinate range A, determine whether all personnel in the maintenance team personnel set N are within set P, and obtain the ticket holding coefficient η. 持票 The expression is,

[0021]

[0022] A further improvement to the method of the present invention is that the step of obtaining the defect enhancement coefficient includes:

[0023] Obtain the unresolved equipment defects and the probability of defect occurrence within the three-dimensional position coordinate range A, and determine the defect increase coefficient β. 缺陷 The expression is,

[0024]

[0025] Where λ and τ are preset constant terms, λ>1, 0<τ<1, P 缺陷 This represents the maximum probability of equipment defects occurring within the three-dimensional position coordinate range A.

[0026] A further improvement to the method of the present invention lies in the pre-trained personnel relative position calculation model.

[0027] The system is trained using a multilayer sensor, with the POS location where the work activity occurs, the camera angle, and the installation height as inputs, and the relative coordinates of the worker and the camera as outputs.

[0028] A second aspect of the present invention provides a system for identifying unlicensed operations in thermal power plants, comprising:

[0029] The behavior detection module is used to acquire surveillance video data of the production area of ​​the thermal power plant, input it into a pre-trained behavior detection model for detection, and obtain the probability P of human work behavior appearing in the surveillance video data. 作业 and the corresponding POS location where the operation occurred;

[0030] A comparison and recognition module for comparing the probability P 作业 with a preset threshold P0 and obtaining an identification result of un-ticketed operations in a thermal power plant based on the comparison result; where

[0031] if P 作业 ≥P0, then P 无票作业 =P 作业 α 定位 η 持票 β 缺陷 ; if P 作业 <P0, then P 无票作业 =0; in the formula, 0<P0<1, P 无票作业 is the probability of un-ticketed operations determined finally; α 定位 is the positioning improvement coefficient; η 持票 is the ticket-holding coefficient; β 缺陷 is the defect improvement coefficient.

[0032] An electronic device provided by the third aspect of the present invention includes:

[0033] At least one processor; and,

[0034] A memory communicatively connected to the at least one processor; where

[0035] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method for identifying un-ticketed operations in a thermal power plant according to any one of the first aspects of the present invention.

[0036] A computer-readable storage medium provided by the fourth aspect of the present invention stores a computer program, and when the computer program is executed by a processor, the method for identifying un-ticketed operations in a thermal power plant according to any one of the first aspects of the present invention is implemented.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] In the method for identifying un-ticketed operations in a thermal power plant provided by the present invention, by using the monitoring image data and behavior detection model in the production area of the thermal power plant, video monitoring can be associated with personnel positioning; by using the provided calculation expression P 无票作业 =P 作业 α 定位 η 持票 β 缺陷This invention can link production management operations with actual on-site conditions; ultimately, it can proactively and intelligently detect whether there are personnel working without work permits or personnel with work permits who have expanded their work scope without authorization. Compared with existing manual inspections, the technical solution of this invention can significantly improve the real-time performance, accuracy, and coverage of safety supervision, avoid safety risks caused by illegal operations, and greatly reduce the workload of safety supervisors. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art are briefly introduced below; obviously, the drawings described below are some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart illustrating a method for identifying unlicensed operations in a thermal power plant, provided by an embodiment of the present invention.

[0041] Figure 2 This is a schematic diagram of the specific process of a method for identifying unlicensed operations in a thermal power plant, provided by an embodiment of the present invention.

[0042] Figure 3 This is a schematic diagram of a ticketless operation identification system for thermal power plants provided in an embodiment of the present invention;

[0043] Figure 4 This is a schematic diagram of the specific architecture of a thermal power plant no-ticket operation identification system provided in an embodiment of the present invention. Detailed Implementation

[0044] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0045] It should be noted that the terms "first", "second", etc. in the description, claims and above-mentioned drawings of the present invention are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order different from those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily limit to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0046] The following further describes the present invention in detail with reference to the drawings:

[0047] Please refer to Figure 1 , a method for identifying un-ticketed operations in a thermal power plant disclosed in an embodiment of the present invention includes the following steps:

[0048] Step 1, obtain the monitoring image data of the production area of the thermal power plant and input it into a pre-trained behavior detection model for detection to obtain the probability P of the occurrence of personnel operation behavior in the monitoring image data 作业 and the corresponding position POS of the occurrence of the operation behavior (explanatory, this position represents the position of the operation behavior in the monitoring screen);

[0049] Step 2, compare the operation behavior probability P 作业 with a preset threshold P0, and obtain the identification result of un-ticketed operations in the thermal power plant based on the comparison result; wherein,

[0050] If P 作业 ≥P0, then P 无票作业 =P 作业 α 定位 η 持票 β 缺陷 ; if P 作业 <P0, then P 无票作业 =0;

[0051] In the formula, 0<P0<1, P 无票作业 is the probability of the occurrence of un-ticketed operations finally determined; α 定位 is the positioning improvement coefficient; η 持票 is the ticket-holding coefficient; β 缺陷 is the defect improvement coefficient.

[0052] Please refer to Figure 2 , a method for identifying un-ticketed operations in a thermal power plant provided by another embodiment of the present invention includes the following steps:

[0053] S1: Input the real-time monitoring video data of the cameras in the production area of the thermal power plant into a pre-trained job behavior detection model, and calculate the probability P of the occurrence of personnel job behaviors in the real-time monitoring video stream 作业 and the corresponding position POS of the occurrence of the job behavior (this position represents the position of the job behavior in the monitoring screen). Compare P 作业 with a preset threshold P0. If P 作业 ≥P0, then transfer to S2; if P 作业 <P0, then P 无票作业 =0, and the process ends. Among them, 0<P0<1, and P 无票作业 is the probability of the occurrence of un-ticketed operations determined finally.

[0054] In the embodiments of the present invention, the architecture of the job behavior detection model adopts mature behavior recognition model architectures such as C3D (3D Convolutional Networks) or P3D (Pseudo-3D Residual Net) or I3D (Inflated 3D ConvNet) or SlowFast Networks or Two-stream CNN. When training the model, weight adjustment is performed based on the actual on-site video sample data on the basis of the pre-trained model. The actual on-site video sample data is obtained by intercepting and marking the videos with job behaviors and the videos without job behaviors captured by the on-site monitoring cameras respectively. Among the video samples with job behaviors, as many types of job behaviors as possible should be covered.

[0055] S2: Input the position POS of the occurrence of the job behavior, the angle and installation height of the camera into the personnel relative position calculation model to obtain the relative coordinates of the job personnel with respect to the camera; obtain the three-dimensional position coordinate range A of the job personnel according to the relative coordinates of the job personnel with respect to the camera and the installation position coordinates of the camera, including the position center point A0 and the preset activity radius R. Then transfer to S3.

[0056] Specifically示例性, in the embodiments of the present invention, the personnel relative position calculation model takes the position POS of the occurrence of the job behavior, the angle and installation height of the camera as inputs and the relative coordinates of the personnel with respect to the camera as outputs, and is trained and predicted through a multi-layer perceptron.

[0057] S3: Obtain the set M of personnel currently located within the three-dimensional position coordinate range A from the personnel positioning information; screen out the set N of maintenance team personnel from the set M of personnel; obtain the maximum value t m of the residence time of all people in the set N of personnel within the three-dimensional position coordinate range A from the personnel positioning information, and calculate and compare it with the preset reference time t0 to obtain the positioning improvement coefficient α 定位 , and then transfer to S4; α 定位The calculation formula is as follows:

[0058] α 定位 =tanh(t) m / t0-1)+1;

[0059] For example, the maintenance team personnel set N can be selected from the personnel set M based on user attribute information (including the user's department and work group).

[0060] S4: Obtain the set P of work members included in the set of work tickets that have commenced work within the three-dimensional location coordinate range A, determine whether all personnel in set N are in set P, and obtain the ticket holding coefficient η. 持票 Switch to S5.

[0061]

[0062] Specifically, for example, the set of work members P included in the set of work orders that have started work within the three-dimensional location coordinate range A can be obtained based on the equipment location information and work order information; wherein, for example, the work order information may include the work members of the work order, the equipment functional location code (hereinafter referred to as the equipment KKS code), and the working time; for example, the equipment location information may be the equipment KKS code and the three-dimensional coordinates of the location of the equipment.

[0063] S5: Obtain the unresolved equipment defects and their occurrence probability within the three-dimensional position coordinate range A, and determine the defect increase coefficient β. 缺陷 Switch to S6;

[0064]

[0065] Where λ and τ are preset constant terms, λ>1, 0<τ<1, P 缺陷 This represents the maximum probability of equipment defects occurring within the three-dimensional position coordinate range A.

[0066] Specifically, for example, based on equipment location information, information on unresolved equipment defects, and equipment defect warning information, the unresolved equipment defects and their probability of occurrence within the three-dimensional location coordinate range A can be obtained; wherein, for example, the equipment defect warning information can be the equipment KKS code, possible defect category, and probability of occurrence; and for example, the information on unresolved equipment defects can be the equipment KKS code and defect type of the unresolved defects.

[0067] S6: Calculate the probability of a no-ticket operation occurring in the final determination: P 无票作业 =P 作业 α 定位 η 持票 β 缺陷 The location for determining a no-ticket operation is a three-dimensional location coordinate range A, and the personnel involved in the no-ticket operation are a set NP.

[0068] This invention discloses a method for identifying unlicensed operations in thermal power plants. It can link video surveillance, personnel positioning with defects, work permits and other production management operations to detect in real time whether there are personnel working without work permits or personnel with work permits but who have expanded the scope of work without authorization. This greatly improves the real-time performance, accuracy and coverage of safety supervision and avoids safety risks caused by illegal operations.

[0069] Please see Figure 3 This invention provides a system for identifying unlicensed operations in thermal power plants, comprising: a personnel positioning subsystem, a video surveillance subsystem, a work permit management subsystem, a defect management subsystem, an equipment information subsystem, a robot inspection subsystem, a user management subsystem, and a violation identification subsystem; wherein,

[0070] The personnel positioning subsystem is used to provide personnel positioning information in the production area of ​​the thermal power plant. Specifically, mature personnel positioning methods such as radio frequency identification (RFID) positioning or ultra-wideband (UWB) positioning can be used. Multiple positioning base stations are installed in the production area, and corresponding positioning tags are worn by users. The three-dimensional position coordinates of users are calculated in real time through personnel positioning calculation services.

[0071] The video surveillance subsystem is used to provide real-time monitoring images of the production area of ​​the thermal power plant; specifically, for example, the images can be acquired by cameras fixedly installed in the production area of ​​the thermal power plant, and the cameras basically meet the requirement of full coverage of the production area.

[0072] The work order management subsystem is used to provide work order information, which may include the work members, equipment KKS codes, and work time of the work order.

[0073] The defect management subsystem is used to provide information on unresolved equipment defects; specifically, this may include the KKS code and defect type of the unresolved defect.

[0074] The device information subsystem is used to provide device location information; specifically, it may include the device's KKS code and the three-dimensional coordinates of its location.

[0075] The robot inspection subsystem is used for equipment inspection via inspection robots. It includes robot management services, a robot charging device, the inspection robot itself, and its mounted cameras, infrared thermal imaging devices, sound acquisition devices, wireless vibration measurement devices, and wireless temperature measurement devices. After each inspection, the robot inspection subsystem provides equipment defect warning information to the violation identification subsystem. This warning information is provided in the form of the equipment KKS code, possible defect category, and probability of defect occurrence.

[0076] The user management subsystem is used to provide user attribute information;

[0077] The violation identification subsystem acquires personnel location information of the thermal power plant production area provided by the personnel positioning subsystem, real-time monitoring images of the thermal power plant production area provided by the video surveillance subsystem, work ticket information provided by the work ticket management subsystem, information on unresolved equipment defects provided by the defect management subsystem, equipment location information provided by the equipment information subsystem, equipment defect early warning information provided by the robot inspection subsystem, and user attribute information provided by the user management subsystem. Through a preset method for identifying unlicensed operations in thermal power plants, it acquires the location, probability, and personnel involved in unlicensed operations.

[0078] In a further improvement to the method of this invention, the preliminary preparation process for acquiring data may specifically include the following steps:

[0079] The personnel positioning subsystem provides personnel positioning information (exemplary, specifically the user's three-dimensional location coordinates) for the production area of ​​the thermal power plant.

[0080] The video surveillance subsystem provides real-time monitoring images captured by cameras in the production area of ​​the thermal power plant.

[0081] Work order information is provided through the work order management subsystem (for example, it may include the work members, equipment KKS codes, and working hours of the work orders that have been started);

[0082] Information on unresolved equipment defects is provided through the defect management subsystem (for example, the KKS code and defect type of the unresolved defect).

[0083] The device location information (for example, the device's KKS code and its three-dimensional coordinates) is provided through the device information subsystem.

[0084] The robot inspection subsystem provides early warning information on equipment defects (for example, the equipment KKS code, possible defect categories, and probability of defect occurrence).

[0085] User attribute information is provided through the user management subsystem.

[0086] In summary, the embodiments of the present invention specifically disclose a method for identifying unlicensed operations in thermal power plants, including: obtaining information such as personnel positioning, video surveillance, work tickets, defects, equipment locations, defect warnings, and user attributes respectively; inputting the video surveillance images into an operation behavior detection model to calculate the probability of personnel operation behaviors occurring in the monitored video stream and the corresponding positions where the operation behaviors occur; calculating the three-dimensional position coordinate range A of the operating personnel according to the position where the operation behavior occurs, the camera angle, the installation height, and the installation position coordinates; obtaining the set N of maintenance team personnel within the range A from the personnel positioning information and the user attribute information; calculating the positioning improvement coefficient according to the residence time of the personnel in the set N within the range A; obtaining the set P of work members included in the work tickets that have been started within the range A, comparing with the personnel in the set N, and obtaining the ticket-holding coefficient; obtaining the uneliminated equipment defects and the probability of defect occurrence within the range A to determine the defect improvement coefficient; calculating the probability of unlicensed operations occurring, the positions of unlicensed operations, and the personnel of unlicensed operations. This method can detect in real time whether there are situations of unlicensed operations by personnel without work tickets or personnel with work tickets but unauthorized expansion of the operation scope and other illegal operations on site, greatly improving the real-time performance, accuracy, and coverage of safety supervision, and avoiding safety risks caused by illegal operations.

[0087] The following is the device embodiment of the present invention, which can be used to execute the method embodiment of the present invention. For the details not disclosed in the device embodiment, please refer to the method embodiment of the present invention.

[0088] Please refer to Figure 4 , in another embodiment of the present invention, a system for identifying unlicensed operations in thermal power plants is provided, including:

[0089] A behavior detection module, configured to obtain the monitoring image data of the production area of the thermal power plant and input it into a pre-trained behavior detection model for detection, and obtain the probability P of personnel operation behaviors occurring in the monitoring image data 作业 and the corresponding position POS where the operation behavior occurs;

[0090] A comparison and identification module, configured to compare the probability P 作业 with a preset threshold P0, and obtain the identification result of unlicensed operations in the thermal power plant based on the comparison result; wherein,

[0091] If P 作业 ≥P0, then P 无票作业 =P 作业 α 定位 η 持票 β 缺陷 ; if P 作业 <P0, then P 无票作业 =0;

[0092] In the formula, 0<P0<1, P 无票作业 is the probability of unlicensed operations occurring in the final determination; α 定位η is the positioning enhancement factor; 持票 β is the voting coefficient. 缺陷 This is the defect enhancement factor.

[0093] In another embodiment of the present invention, a computer device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to achieve a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used for the operation of a method for identifying unlicensed operations in thermal power plants.

[0094] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the method for identifying unlicensed operations in thermal power plants in the above embodiments.

[0095] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0096] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0097] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0098] 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.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for identifying a ticketless operation in a thermal power plant, characterized by, The method comprises the following steps: The monitoring image data of the thermal power plant production area is obtained through the camera installed in the thermal power plant production area, and is input into a pre-trained behavior detection model for detection to obtain the probability of occurrence of personnel operation behavior in the monitoring image data and the corresponding operation behavior occurrence position POS; probabilities comparing the probabilities with preset threshold values In contrast, the no-ticket operation identification result of the thermal power plant is obtained based on the comparison result. If , then ; if , then ; in the formula, , is the final determination of the probability of no ticket operation; is the positioning promotion coefficient; is the ticket holding coefficient; is the defect promotion coefficient; The behavior detection model adopts any one of C3D, P3D, I3D, SlowFast and Two-stream. The step of obtaining the positioning improvement coefficient comprises: obtaining an angle, an installation height, and an installation position coordinate of the camera; inputting the angle, the installation height, and a position of occurrence of a work behavior (POS) of the camera into a pre-trained personnel relative position calculation model to obtain a relative coordinate between a work personnel and the camera; obtaining a three-dimensional position coordinate range A of the work personnel according to the relative coordinate between the work personnel and the camera and the installation position coordinate of the camera; wherein the three-dimensional position coordinate range A comprises a position center point A0 and a preset activity radius R; obtaining a personnel set M positioned in the three-dimensional position coordinate range A through a personnel positioning device, and screening out a maintenance team personnel set N from the personnel set M; and obtaining a maximum value of a staying time of all people in the maintenance team personnel set N in the three-dimensional position coordinate range A t m , and performing calculation and comparison with a preset reference time to obtain a positioning improvement coefficient , and a calculation formula is ; The obtaining step of the ticket holding coefficient comprises: obtaining a working member set P contained in a working ticket set which has been started in the three-dimensional position coordinate range A, judging whether all the personnel in the maintenance team personnel set N are in the set P, and obtaining the ticket holding coefficient , the expression is, ; The obtaining step of the defect promotion coefficient comprises: obtaining the uneliminated device defects in the three-dimensional position coordinate range A, determining the defect promotion coefficient according to the defect occurrence probability , and the expression is ； wherein, and is a preset constant term, , , is the maximum value of the probability of occurrence of defects of the device in the three-dimensional position coordinate range A.

2. The method according to claim 1, wherein The training steps of the trained behavior detection model comprise:

3. The method according to claim 2, wherein On the basis of the pre-trained model, weight adjustment is performed on the actual field video sample data to obtain the final trained behavior detection model. The actual field video sample data is obtained by shooting the production area of the thermal power plant by a monitoring camera to obtain videos with work behavior and videos without work behavior, and then the sample data is obtained by cutting and marking. The pre-trained personnel relative position calculation model, 4. The method of claim 1, wherein the method further comprises: The input is the work behavior occurrence position POS, the angle and the installation height of the camera, and the output is the relative coordinates of the worker and the camera, and the multi-layer perceptron is trained. The method comprises the following steps:

5. A system for identifying no-ticket operation in a thermal power plant, characterized by, The method comprises the following steps: The behavior detection module is configured to acquire monitoring image data of a production area of the thermal power plant, and input a pre-trained behavior detection model to perform detection, so as to obtain a probability of a personnel operation behavior appearing in the monitoring image data and a corresponding operation behavior occurrence position POS. The comparison identification module is configured to compare the probability with a preset threshold In contrast, the result of the comparison is used to obtain the result of identifying the no-ticket operation of the thermal power plant; wherein, If , then ; if , then ; in which , is the final determination of the probability of no ticket operation; is the positioning promotion coefficient; is the ticket holding coefficient; is the defect promotion coefficient; At least one processor; The acquisition process of the positioning promotion coefficient comprises: acquiring an angle, an installation height and an installation position coordinate of the camera; inputting the angle, the installation height and a position of occurrence of a work behavior (POS) of the camera into a pre-trained personnel relative position calculation model to obtain a relative coordinate between a work personnel and the camera; obtaining a three-dimensional position coordinate range A of the work personnel according to the relative coordinate between the work personnel and the camera and the installation position coordinate of the camera; wherein the three-dimensional position coordinate range A comprises a position center point A0 and a preset activity radius R; acquiring a personnel set M positioned in the three-dimensional position coordinate range A through a personnel positioning device, and screening a maintenance team personnel set N from the personnel set M; acquiring a maximum value of a staying time of all people in the maintenance team personnel set N in the three-dimensional position coordinate range A t m , and performing calculation and comparison with a preset reference time to obtain a positioning promotion coefficient , and the calculation formula is ; The obtaining process of the ticket holding coefficient comprises: obtaining a working member set P contained in a working ticket set that has been started in the three-dimensional position coordinate range A, judging whether all the personnel in the maintenance team personnel set N are in the set P, and obtaining the ticket holding coefficient , the expression is, ; The acquisition process of the defect promotion coefficient comprises: acquiring the uneliminated device defects in the three-dimensional position coordinate range A, determining the defect promotion coefficient according to the defect occurrence probability , the expression is ； wherein, and is a preset constant term, , , is the maximum value of the probability of occurrence of defects of the device in the three-dimensional position coordinate range A.

6. An electronic device, comprising: And The memory is in communication connection with the at least one processor; wherein The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the thermal power plant no-ticket work identification method according to any one of claims 1 to 4. The computer program is executed by the processor to implement the thermal power plant no-ticket work identification method according to any one of claims 1 to 4. ​ 7. A computer readable storage medium storing a computer program, characterized in that, ​

Images