A power digital safety monitoring system and operator supervision method

Abstract

The present invention relates to the field of power management technology, and in particular to a power digital intelligence safety monitoring system and an operator supervision method. By comparing the target movement trajectory distribution information and the on-site movement trajectory distribution information, the on-site operation score of the operator is calculated. The present invention provides the operation benchmark time distribution information for fixed common operation types and operation locations. When the actual operation process of the operator differs greatly from the benchmark, the situation can be discovered in time. After professional analysis, it can be determined whether it is a human problem or an equipment problem, thereby realizing the supervision of the operation process. In addition, the technical solution can be implemented through software algorithms, with low cost and wide application range.

Description

Technical Field

[0001] The present invention relates to the field of power management technology, and in particular to a power digital intelligent safety monitoring system and an operator supervision method. Background Art

[0002] With the rapid development of information technology, traditional power operation and maintenance management can no longer meet the current requirements of safe production, and new information-based supervision systems have gradually been applied to the power operation and maintenance management process.

[0003] The current power operation and maintenance information management system is generally used for simple process management and authority management. It lacks the necessary management means to determine whether the diverse operations of the power system are reasonable, resulting in uneven quality of power operation and maintenance operations. Summary of the Invention

[0004] The present invention discloses an electric power digital safety monitoring system and an operator supervision method, which can not only realize basic process management, but also supervise the operation quality of operators and improve the quality of electric power operation and maintenance work.

[0005] To achieve the above objectives, on the one hand, a power digital intelligence safety monitoring system is provided, the system comprising:

[0006] Personnel information storage module, which records basic information of operators, operation scene information and operation information of operators;

[0007] The operation target storage module records the target movement trajectory distribution information corresponding to each operation task at each operation site;

[0008] The entry management module uses the first camera to obtain facial information of the operator at the work site, compares the facial information with the basic information of the operator, and determines whether the operator is allowed to enter the work site;

[0009] The on-site trajectory acquisition module uses the second camera to obtain the full-process operation video of the operator at the operation site, records the on-site movement trajectory information of the operator based on the full-process operation video, and calculates the on-site movement trajectory distribution information based on the on-site movement trajectory information;

[0010] The operation supervision module compares the on-site movement trajectory distribution information with the corresponding target movement trajectory distribution information and calculates the on-site operation score of the operator.

[0011] Furthermore, the on-site trajectory acquisition module includes an on-site movement trajectory information calculation submodule, which divides the on-site operation space into several sub-areas and calculates the operator's stay time in each sub-area based on the operator's on-site movement trajectory information;

[0012] The target movement trajectory distribution information and the on-site movement trajectory distribution information both include a time ratio of the stay time in each sub-area.

[0013] The advantage of this embodiment is that it provides a digital safety and supervision management system for power operation and maintenance, which, in addition to completing the digital management of traditional process permissions, can also evaluate the actual work status of operation and maintenance personnel and improve the overall operation and maintenance work quality.

[0014] In order to achieve the above purpose, on the other hand, a method for supervising operators is provided, and the specific method is as follows:

[0015] Determine the operator's work site and specific work type;

[0016] Read the target movement trajectory distribution information corresponding to the work site and work type;

[0017] Record the entire operation video of the operator and obtain the on-site movement trajectory information of the operator based on the entire operation video;

[0018] Based on the on-site movement trajectory information, the proportion of the workers' stay time in each sub-area is counted, that is, the on-site movement trajectory distribution information;

[0019] Compare the target movement trajectory distribution information with the on-site movement trajectory distribution information to calculate the on-site operation score of the operator;

[0020] The target movement trajectory distribution information and the on-site movement trajectory distribution information both include a range of the proportion of the stay time between each sub-area of ​​the work space.

[0021] The advantage of this embodiment is that it provides benchmark time distribution information for common fixed operation types and locations. When an operator's actual operation differs significantly from the benchmark, this information can be promptly identified. Professional analysis can determine whether it is a human error or equipment problem, thus enabling oversight of the operation process. Furthermore, this technical solution can be implemented using software algorithms, resulting in low cost and wide applicability.

[0022] Furthermore, the on-site movement trajectory information of the operator is obtained based on the full operation video. The specific method is as follows:

[0023] The shooting time is marked for the image data of the entire operation video at fixed time intervals;

[0024] Identify operators in the entire operation video;

[0025] The distance between the operator and the depth camera is measured by the depth camera;

[0026] Combining the depth camera's coordinates, the shooting angle, the distance to the operator, and the spatial coordinate system, the operator's coordinates in the spatial coordinate system are calculated;

[0027] The coordinates of the workers in the spatial coordinate system are associated with the image shooting time, arranged in chronological order, and the on-site movement trajectory information of the workers with the shooting time is constructed.

[0028] The advantage of this embodiment is that different work locations and types of work have different work requirements, and currently it is impossible to achieve comprehensive supervision. This embodiment innovatively uses the trajectory distribution of workers to achieve supervision. The trajectory distribution of workers is obtained through video images, and then the coordinates of the workers' tracks are obtained through a ranging algorithm. Finally, the coordinates are associated with the shooting time to provide a basis for subsequent analysis.

[0029] Furthermore, to identify the workers in the entire operation video, a human body recognition convolutional neural network model or a marker recognition convolutional neural network model is used.

[0030] The advantage of this embodiment is that the convolutional neural network model is a mature existing technology in or special object recognition, and the convolutional neural network can ensure target tracking.

[0031] Furthermore, the on-site movement trajectory distribution information is obtained as follows:

[0032] Divide the work space into several sub-areas;

[0033] Mark the spatial coordinates of each sub-region;

[0034] Divide the on-site movement trajectory information of the operators into several sub-areas;

[0035] Count the image capture time associated with the on-site movement trajectory information of the workers in each sub-area to calculate the total stay time in each sub-area;

[0036] According to the total stay time in each sub-area, the stay time ratio between the sub-areas is calculated.

[0037] The advantage of this embodiment is that the on-site movement trajectory of workers is simply a series of coordinate changes, making it difficult to extract phase difference data features and correlate them with the type of work. This embodiment utilizes sub-area division and sub-area dwell time to significantly extract data correlations between worker movement trajectory information and the type of work being performed, providing a foundation for evaluating work quality. Specifically, if the actual dwell time of a worker in a sub-area differs significantly from the expected dwell time for the primary work type, this clearly does not conform to the expected behavior of the work.

[0038] Furthermore, the on-site performance score of the operator is calculated as follows:

[0039] Calculate the residence time in each sub-area in the on-site movement trajectory distribution information, and the part that exceeds the proportion range of the residence time between each sub-area in the target movement trajectory distribution information in real time. Accumulate the proportion of all exceeding parts, and subtract the proportion of all exceeding parts from 1 to calculate the on-site operation score of the operator.

[0040] The advantage of this embodiment is that if the overall time distribution of the operator is similar to the target time distribution, it can be preliminarily determined that the operator's specific work is correct. If the overall time distribution of the operator differs from the target time distribution, the difference can be quantified. The use of a ratio takes into account the speed differences of each operator, reducing the possibility of misjudgment.

[0041] Furthermore, the feedback of the operators on the on-site operation scores is counted. If the feedback of the operators on the on-site operation scores is lower than a preset value, the proportion range of the stay time between each sub-area in the target movement trajectory distribution information is updated.

[0042] The advantage of this embodiment is that when the operator reports a problem with the rating, it may be that a new technical problem has occurred during the operation, resulting in a change in the time allocation. Therefore, this embodiment provides a channel for updating the target time allocation range.

[0043] Furthermore, the target movement trajectory distribution information is updated with the following method:

[0044] Obtain the over-stay sub-area of ​​the feedback operator's over-stay time ratio range;

[0045] Obtain feedback on the operator's overstay value in the overstay sub-area;

[0046] The upper limit of the updated stay time ratio range of the super-stay sub-area is obtained by multiplying the super-stay value by the correction coefficient and adding the original upper limit range of the super-stay sub-area. The lower limit of the stay time ratio range remains unchanged.

[0047] The correction factor is the predicted probability of an event causing the operator to exceed the stay time in the future.

[0048] The advantage of this embodiment is that it provides a quantitative method for updating the proportion range of the stay time of each sub-area in the target movement trajectory distribution information. The use of this method can ensure that the field operation score can be updated at any time.

[0049] Other advantages, objects, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings of the present invention are described below.

[0051] Figure 1 Schematic diagram of the measurement principle of Example 2.

[0052] Figure 2 This is a schematic diagram of the system architecture of Example 1.

[0053] Figure 3 This is a schematic diagram of the process of Example 2. DETAILED DESCRIPTION

[0054] The present invention will be further described below with reference to the accompanying drawings and examples.

[0055] Example 1:

[0056] A power digital safety and monitoring system, such as Figure 2 As shown, the system includes:

[0057] Personnel information storage module, which records basic information of operators, operation scene information and operation information of operators;

[0058] The operation target storage module records the target movement trajectory distribution information corresponding to each operation task at each operation site;

[0059] The entry management module uses the first camera to obtain facial information of the operator at the work site, compares the facial information with the basic information of the operator, and determines whether the operator is allowed to enter the work site;

[0060] The on-site trajectory acquisition module uses the second camera to obtain the full-process operation video of the operator at the operation site, records the on-site movement trajectory information of the operator based on the full-process operation video, and calculates the on-site movement trajectory distribution information based on the on-site movement trajectory information;

[0061] The operation supervision module compares the on-site movement trajectory distribution information with the corresponding target movement trajectory distribution information and calculates the on-site operation score of the operator.

[0062] Example 2:

[0063] A method of monitoring operators, such as Figure 3 The specific steps are as follows:

[0064] S1. Determine the operator's work site and specific work type;

[0065] S2. Read the target movement trajectory distribution information corresponding to the work site and work type. in, is the lower limit of the target stay time in the fourth sub-area, The target upper limit of stay time in the fourth sub-area.

[0066] In this embodiment, the work area is divided into four sub-areas.

[0067] S3. Record the operator's full operation video and obtain the operator's on-site movement trajectory information based on the full operation video in The coordinates corresponding to the k-th picture of the operator.

[0068] To obtain the on-site movement trajectory information of the operator, the specific steps are as follows:

[0069] S31, marking the shooting time for the image data of the entire operation video at fixed time intervals in The shooting time of the kth picture of the operator.

[0070] S32. Identify the operators in the entire operation video. Identify the operators in the entire operation video using a human body recognition convolutional neural network model.

[0071] S33, measuring the distance between the operator and the depth camera through the depth camera in, is the distance between the operator and the depth camera when the kth picture is taken.

[0072] In step S33, the depth camera distance measurement principle is as follows Figure 1 As shown in the figure, the binocular stereo vision system matches feature points of the RGB image and calculates the depth information of the image using triangulation. In the figure, P is a point where the operator is located in the image, I is the imaging plane, A is the imaging distance between the left and right cameras, B is the center length of the camera aperture, C is the distance between the camera and the camera, and f represents the focal length of the camera.

[0073] Assume that the imaging points on the left and right sides of the operator are P L 、P R , the aperture center of the left and right cameras is O L , O R , the coordinates of the two imaging points in the image are U L 、U R , disparity value d = U L -UR , according to similar triangles we can get:

[0074]

[0075] Among them, A=Bd, and the depth information of the image is calculated by calculating the disparity value d using triangulation.

[0076] S34, combining the coordinates (x0, y0) of the depth camera, the shooting angle θ, and the distance between the operator and the camera And the spatial coordinate system, calculate the coordinates of the operator in the spatial coordinate system

[0077] S35. Associate the coordinates of the operator in the spatial coordinate system with the image shooting time, arrange them in chronological order, and construct the operator's on-site movement trajectory information with the shooting time

[0078] S4. Based on the on-site movement trajectory information, the proportion of the operator's stay time in each sub-area is counted, that is, the on-site movement trajectory distribution information.

[0079] To obtain on-site mobile trajectory distribution information, the specific steps are as follows:

[0080] S41, dividing the working space into four sub-areas;

[0081] S42. Mark the spatial coordinates of each sub-region;

[0082] S43, dividing the on-site movement trajectory information of the operator into several sub-areas;

[0083] S44, counting the image capture time associated with the on-site movement trajectory information of the workers in each sub-area, and calculating the total stay time in each sub-area;

[0084] S45. Calculate the ratio of the stay time between the sub-areas based on the total stay time in each sub-area.

[0085] S5. Compare the target movement trajectory distribution information with the on-site movement trajectory distribution information, and calculate the on-site operation score of the operator.

[0086] The on-site performance score of the operator is calculated as follows:

[0087] Calculate the residence time in each sub-area in the on-site movement trajectory distribution information, and the part that exceeds the proportion range of the residence time between each sub-area in the target movement trajectory distribution information in real time. Accumulate the proportion of all exceeding parts, and subtract the proportion of all exceeding parts from 1 to calculate the on-site operation score of the operator.

[0088] In this embodiment, and The remaining sub-areas are within the range, then:

[0089]

[0090] S6. Count the feedback of the operators on the on-site operation scores. If the feedback of the operators on the on-site operation scores is lower than a preset value, update the proportion range of the stay time between each sub-area in the target movement trajectory distribution information.

[0091] Update the target movement trajectory distribution information for each sub-area's residence time ratio range. The specific method is as follows:

[0092] S61 : Obtain and feedback the over-stay sub-region within the range of the operator's over-stay time ratio, which is the fourth sub-region in this embodiment.

[0093] S62: Obtain feedback on the operator's overstay value in the overstay sub-area

[0094] S63, multiply the excess dwell value by the coefficient Plus the original upper limit range of the super stay sub-region As the upper limit of the stay time ratio range after the super stay sub-area is updated, the lower limit of the stay time ratio range remains unchanged, that is,

[0095]

[0096] S64. The correction coefficient is the predicted probability of an event causing the operator to exceed the stay time in the future.

[0097] 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, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A digital power safety and monitoring system, characterized in that: The system comprises: Personnel information storage module, which records basic information of operators, operation scene information and operation information of operators; The operation target storage module records the target movement trajectory distribution information corresponding to each operation task at each operation site; The entry management module uses the first camera to obtain facial information of the operator at the work site, compares the facial information with the basic information of the operator, and determines whether the operator is allowed to enter the work site; The on-site trajectory acquisition module uses the second camera to obtain the full-process operation video of the operator at the operation site, records the on-site movement trajectory information of the operator based on the full-process operation video, and calculates the on-site movement trajectory distribution information based on the on-site movement trajectory information; The operation supervision module compares the on-site movement trajectory distribution information with the corresponding target movement trajectory distribution information and calculates the on-site operation score of the operator; Get on-site movement trajectory distribution information. The specific method is as follows: Divide the work space into several sub-areas; Mark the spatial coordinates of each sub-region; Divide the on-site movement trajectory information of the operators into several sub-areas; Count the image capture time associated with the on-site movement trajectory information of the workers in each sub-area to calculate the total stay time in each sub-area; According to the total stay time in each sub-area, the stay time ratio between sub-areas is calculated; The on-site performance score of the operator is calculated as follows: Calculate the residence time in each sub-area in the on-site movement trajectory distribution information, and the part that exceeds the proportion range of the residence time between each sub-area in the target movement trajectory distribution information in real time. Accumulate the proportion of all exceeding parts, and subtract the proportion of all exceeding parts from 1 to calculate the on-site operation score of the operator.

2. The power digital intelligence safety monitoring system according to claim 1, characterized in that: The on-site trajectory acquisition module includes an on-site movement trajectory information calculation submodule, which divides the on-site operation space into several sub-areas and calculates the operator's stay time in each sub-area based on the operator's on-site movement trajectory information; The target movement trajectory distribution information and the on-site movement trajectory distribution information both include a time ratio of the stay time in each sub-area.

3. A method for supervising workers, characterized in that: The specific method is as follows: Determine the operator's work site and specific work type; Read the target movement trajectory distribution information corresponding to the work site and work type; Record the entire operation video of the operator and obtain the on-site movement trajectory information of the operator based on the entire operation video; Based on the on-site movement trajectory information, the proportion of the workers' stay time in each sub-area is counted, that is, the on-site movement trajectory distribution information; Compare the target movement trajectory distribution information with the on-site movement trajectory distribution information to calculate the on-site operation score of the operator; The target movement trajectory distribution information and the on-site movement trajectory distribution information both include the range of the proportion of the stay time between each sub-area of ​​the work space; Get on-site movement trajectory distribution information. The specific method is as follows: Divide the work space into several sub-areas; Mark the spatial coordinates of each sub-region; Divide the on-site movement trajectory information of the operators into several sub-areas; Count the image capture time associated with the on-site movement trajectory information of the workers in each sub-area to calculate the total stay time in each sub-area; According to the total stay time in each sub-area, the stay time ratio between sub-areas is calculated; The on-site performance score of the operator is calculated as follows: Calculate the residence time in each sub-area in the on-site movement trajectory distribution information, and the part that exceeds the proportion range of the residence time between each sub-area in the target movement trajectory distribution information in real time. Accumulate the proportion of all exceeding parts, and subtract the proportion of all exceeding parts from 1 to calculate the on-site operation score of the operator.

4. The operator supervision method according to claim 3, characterized in that: The on-site movement trajectory information of the operator is obtained based on the full operation video. The specific method is as follows: The shooting time is marked for the image data of the entire operation video at fixed time intervals; Identify operators in the entire operation video; The distance between the operator and the depth camera is measured by the depth camera; Combining the depth camera's coordinates, the shooting angle, the distance to the operator, and the spatial coordinate system, the operator's coordinates in the spatial coordinate system are calculated; The coordinates of the workers in the spatial coordinate system are associated with the image shooting time, arranged in chronological order, and the on-site movement trajectory information of the workers with the shooting time is constructed.

5. The operator supervision method according to claim 4, characterized in that: To identify the workers in the entire operation video, a human body recognition convolutional neural network model or a marker recognition convolutional neural network model is used.

6. The operator supervision method according to claim 3, characterized in that: The feedback of the operators on the on-site operation scores is collected. If the feedback of the operators on the on-site operation scores is lower than the preset value, the proportion range of the stay time between each sub-area in the target movement trajectory distribution information is updated.

7. The operator supervision method according to claim 6, characterized in that: Update the target movement trajectory distribution information for each sub-area's residence time ratio range. The specific method is as follows: Obtain the over-stay sub-area of ​​the feedback operator's over-stay time ratio range; Obtain feedback on the operator's overstay value in the overstay sub-area; The upper limit of the updated stay time ratio range of the super-stay sub-area is obtained by multiplying the super-stay value by the correction coefficient and adding the original upper limit range of the super-stay sub-area. The lower limit of the stay time ratio range remains unchanged. The correction factor is the predicted probability of an event causing the operator to exceed the stay time in the future.

Images