Substation virtual fence crossing prevention warning method and system based on the Internet of Things

Abstract

The present invention discloses a method and system for substation virtual fence crossing prevention and warning based on the Internet of Things, belonging to the technical field of virtual fence crossing prevention. The method establishes a three-dimensional coordinate grid of the substation through lidar scanning, defines the boundary coordinate set of the virtual fence, and deploys Internet of Things devices to collect the dynamic positions of moving objects; the perception time is normalized at equal time nodes to generate an over-crossing feature label including distance and detection status; a state perception sensitivity matrix is constructed to evaluate the over-crossing prevention fusion sensitivity and effectiveness between perception time segments, integrate the time range and analyze and lock the detection and protection range. The system includes a virtual fence construction module, an Internet of Things device management module, a data processing module and a warning evaluation module, realizing real-time monitoring of substation crossing behaviors, risk assessment and dynamic adjustment of the protection range, and improving the intelligent and precise level of substation safety protection.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual fence crossing protection, and specifically to an early warning method and system for virtual fence crossing protection of a substation based on the Internet of Things. Background Art

[0002] In the field of substation safety protection, traditional physical fences rely on physical barriers to divide safety areas, and there are significant technical bottlenecks:

[0003] Limited monitoring ability: Physical fences cannot cover complex spaces such as high-altitude operation areas and cable trenches, and it is difficult to track crossing behaviors in three-dimensional space in real time, resulting in monitoring blind spots;

[0004] Data isolation and lag: Dependent on manual inspections or single-point monitoring by independent sensors (such as infrared and cameras), the data lacks spatio-temporal correlation, and the abnormal response delay is relatively high, unable to meet the real-time protection requirements;

[0005] Insufficient dynamic adaptability: Traditional solutions are difficult to adapt to substation equipment layout adjustments or temporary operation scenarios. The fence boundaries are fixed and lack a self-optimization mechanism, and the protection strategy has poor flexibility;

[0006] Coarse risk assessment: The early warning rules based on a single threshold cannot integrate multi-source data (such as location, time, and equipment status), and it is difficult to accurately distinguish normal operations from crossing risks, resulting in a relatively high false alarm rate.

[0007] With the development of Internet of Things technology, it has become possible to monitor through multi-sensor collaboration, but existing solutions still have problems such as timestamp asynchronization and insufficient data fusion depth. For example, the fragmented sensor perception time leads to discontinuous trajectory analysis, and the data of multiple devices are not effectively associated, unable to form a three-dimensional monitoring network. Therefore, there is an urgent need for a virtual fence technology based on the Internet of Things to achieve precise early warning and protection of substation crossing risks through spatio-temporal data fusion and dynamic model optimization. Summary of the Invention

[0008] The purpose of the present invention is to provide an early warning method and system for virtual fence crossing protection of a substation based on the Internet of Things to solve the problems raised in the above background art.

[0009] To solve the above technical problems, the present invention provides the following technical solutions:

[0010] An early warning system for virtual fence crossing protection of a substation based on the Internet of Things, the system includes: a virtual fence construction module, an Internet of Things device management module, a data processing module, and an early warning evaluation module;

[0011] The virtual fence construction module is used to establish a virtual fence boundary coordinate set for the prohibited crossing area;

[0012] The Internet of Things device management module is used to deploy Internet of Things devices in the prohibited crossing area, construct a device archive library and bind the device installation location;

[0013] The data processing module is used to normalize the sensing time of the Internet of Things devices, generate crossing feature tags, construct a state sensing sensitivity matrix and evaluate the crossing protection fusion sensitivity;

[0014] The early warning evaluation module is used to evaluate the effectiveness of crossing protection based on the crossing protection fusion sensitivity, extract and integrate the sensing time segments, and analyze to obtain the locked detection protection range of the Internet of Things devices.

[0015] Further, the virtual fence construction module includes:

[0016] The coordinate generation unit is used to generate a three-dimensional coordinate grid of the substation through lidar scanning and generate a set of virtual fence boundary coordinates for the prohibited crossing area;

[0017] The boundary definition unit is used to define the virtual fence boundary of the prohibited crossing area based on the three-dimensional coordinate grid.

[0018] Further, the Internet of Things device management module includes:

[0019] The device deployment unit is used to deploy Internet of Things devices in the prohibited crossing area, and the Internet of Things devices are used to collect the dynamic positions of moving objects;

[0020] The archive binding unit is used to construct a device archive library and bind the installation location of the Internet of Things devices based on the coordinate grid.

[0021] Further, the data processing module includes:

[0022] The time normalization unit is used to normalize the sensing time of each Internet of Things device in the form of equal time nodes to obtain continuous sensing time segments;

[0023] The feature generation unit is used to track the dynamic positions of moving objects within the sensing time segment, establish crossing feature pairs and generate crossing feature tags;

[0024] The matrix construction unit is used to construct a state sensing sensitivity matrix based on the crossing feature tags and generate a matrix shard stream;

[0025] The sensitivity calculation unit is used to evaluate the crossing protection fusion sensitivity between sensing time segments based on the matrix shard stream.

[0026] Further, the early warning evaluation module includes:

[0027] The effectiveness evaluation unit is used to calculate the effectiveness of crossing protection based on the crossing protection fusion sensitivity;

[0028] A segment integration unit, configured to extract and integrate the sensed time segments according to an effectiveness threshold to obtain an integrated sensed time range;

[0029] A protection range analysis unit, configured to screen a warning center set centered on the Internet of Things device within the integrated sensed time range to constitute a locked detection protection range of the Internet of Things device.

[0030] A virtual fence crossing protection warning method based on the Internet of Things, the method comprising the following steps:

[0031] Step S1: Establish a virtual fence boundary coordinate set of a prohibited crossing area, deploy Internet of Things devices in the prohibited crossing area, construct a device archive, and bind the installation positions of the Internet of Things devices;

[0032] Step S2: Normalize the sensing time of each Internet of Things device in an equal time node manner to obtain a continuous sensed time segment set, establish a crossing feature pair, and generate a crossing feature label;

[0033] Step S3: Based on the crossing feature label, construct a state sensing sensitivity matrix, generate a matrix shard stream; based on the matrix shard stream, evaluate the crossing protection fusion sensitivity between the sensed time segments;

[0034] Step S4: Based on the crossing protection fusion sensitivity, evaluate the crossing protection effectiveness between the sensed time segments, and extract the sensed time segments to obtain an integrated sensed time range; based on the integrated sensed time range, analyze and obtain the locked detection protection range of the Internet of Things device and output it.

[0035] Further, the specific implementation process of step S1 includes:

[0036] Generate a three-dimensional coordinate grid of the substation through lidar scanning to generate a virtual fence boundary coordinate set of the prohibited crossing area , where represents the i-th coordinate grid, is the coordinate value in the three-dimensional direction of the coordinate grid , and I represents the total number of coordinate grids;

[0037] Deploy Internet of Things devices in the prohibited crossing area, the Internet of Things devices are used to collect the dynamic positions of moving objects, construct a device archive, and based on the coordinate grid, bind the installation positions of the Internet of Things devices. Denote the j-th Internet of Things device as , then denote the installation position of the Internet of Things device as , where is the coordinate value in the three-dimensional direction of the Internet of Things device .

[0038] Further, the specific implementation process of step S2 includes:

[0039] Normalize the sensing time of each Internet of Things device at equal time nodes to obtain a set of continuous sensing time segments , where represents the g-th sensing time segment, and G represents the total number of sensing time segments with a daily cycle period;

[0040] Internet of Things device During the sensing time segment , track the dynamic position of the moving object. If the moving object is tracked into the coordinate grid , establish an out-of-bounds feature pair , and generate an out-of-bounds feature label , where represents the fixed distance between the Internet of Things device and the coordinate grid , and , represents the detection status value, and is equal to 0 or 1. When it is 0, it means that no moving object is detected, and when it is 1, it means that a moving object is detected.

[0041] Further, the specific implementation process of step S3 includes:

[0042] Based on the out-of-bounds feature label, construct a state perception sensitivity matrix. The row number of the state perception sensitivity matrix is the coding number of the Internet of Things device, and the column number of the state perception sensitivity matrix is the coding number of the coordinate grid. Then, the matrix position at the j-th row and the i-th column of the state perception sensitivity matrix is , and the matrix element value at the matrix position is . Then, the state perception sensitivity matrix generated within the sensing time segment is denoted as , and the matrix shard stream is obtained;

[0043] Based on the matrix shard stream, evaluate the out-of-bounds protection fusion sensitivity between the sensing time segment , where in the formula, , represents the total number of matrix positions with a value of 1 after the Boolean logic AND between the state perception sensitivity matrix and the state perception sensitivity matrix , represents the state perception sensitivity matrix and the state perception sensitivity matrix The total number of matrix positions with a value of 1 after the Boolean logic union.

[0044] Further, the specific implementation process of step S4 includes:

[0045] Based on the cross - boundary protection fusion sensitivity, evaluate the perception time segment and the perception time segment The cross - boundary protection effectiveness between , where is the mean value of the cross - boundary protection fusion sensitivity, and , is the variance of the cross - boundary protection fusion sensitivity, and ;

[0046] Preset an effectiveness threshold. If the cross - boundary protection effectiveness is greater than or equal to the effectiveness threshold, extract the perception time segment and the perception time segment , otherwise do not extract the perception time segment and the perception time segment ;

[0047] Perform segment fusion on all the extracted perception time segments, set a time sliding window, and translate the time sliding window. If the perception time segments within the time sliding window are continuous, perform time - scale integration on the continuous perception time segments to obtain an integrated perception time range;

[0048] Select the integrated perception time range with the largest time span, and centered on the Internet of Things device collect the cross - boundary feature pairs corresponding to the matrix positions in each state perception sensitivity matrix , form the early - warning center set of the Internet of Things device , denoted as , and based on the cross - boundary feature label , screen the coordinate grid with the shortest distance between the early - warning center set and the Internet of Things device , and the locked detection and protection range of the Internet of Things device is composed of the screened coordinate grid to guide the Internet of Things device to perform detection and protection in the next cycle; In step S4, the determination logic of the locked detection and protection range:

[0049]

[0050] Early warning center set screening basis: The early warning center set is composed of the coordinate grids of the moving objects detected within each time segment. If no detection is made, it is 0. In step S3, a state perception sensitivity matrix is constructed, which reflects the detection states of different IoT devices for different coordinate grids, that is, the data is 1 or 0. If a certain coordinate grid is marked as "detected moving object" in multiple consecutive time segments, it indicates that a moving object has been detected in this area, and at the same time, there is a continuous risk of crossing the boundary in multiple consecutive time segments (quantified by the matrix statistics of the value 1);

[0051] Purpose of shortest distance screening: IoT devices The locked detection and protection range needs to dynamically focus on the area where crossing the boundary is most likely to occur, and at the same time select the grid point closest to the device because the perception accuracy of IoT devices decays with the increase of distance, and the detection reliability of the nearby grids is higher. At the same time, the shortest distance screening can reduce redundant calculations and optimize resource allocation, that is, reasonably allocate which IoT devices detect which coordinate grids, which constitutes the locked detection and protection range of IoT devices The physical meaning is: centered on the IoT device, in the historical effective detection area, priority is given to key monitoring of the nearest boundary crossing risk points, forming a "proximity first" dynamic protection strategy.

[0052] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: In the substation virtual fence crossing protection early warning method and system based on the Internet of Things provided by the present invention, the method establishes a three-dimensional coordinate grid of the substation through lidar scanning, defines the boundary coordinate set of the virtual fence, and deploys IoT devices to collect the dynamic positions of moving objects; normalizes the perception time at equal time nodes to generate an overstep feature label including distance and detection status; constructs a state perception sensitivity matrix to evaluate the overstep protection fusion sensitivity and effectiveness between perception time segments, integrates the time range and analyzes the locked detection and protection range. The system includes a virtual fence construction module, an IoT device management module, a data processing module, and an early warning evaluation module, realizing real-time monitoring of substation crossing behaviors, risk assessment, and dynamic adjustment of the protection range, and improving the intelligent and precise level of substation safety protection. Brief Description of the Drawings

[0053] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation to the present invention.

[0054] Figure 1 It is a schematic diagram of the steps of the substation virtual fence crossing protection early warning method based on the Internet of Things of the present invention. Detailed Embodiments

[0055] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0056] In the first embodiment: A substation virtual fence crossing prevention and warning system based on the Internet of Things is provided. The system includes: a virtual fence construction module, an Internet of Things device management module, a data processing module, and a warning evaluation module;

[0057] The virtual fence construction module is used to establish a set of virtual fence boundary coordinates for the prohibited crossing area;

[0058] Among them, the virtual fence construction module includes:

[0059] A coordinate generation unit, which is used to generate a three-dimensional coordinate grid of the substation through lidar scanning and generate a set of virtual fence boundary coordinates for the prohibited crossing area;

[0060] A boundary definition unit, which is used to define the virtual fence boundary of the prohibited crossing area based on the three-dimensional coordinate grid;

[0061] The Internet of Things device management module is used to deploy Internet of Things devices in the prohibited crossing area, construct a device archive and bind the device installation location;

[0062] Among them, the Internet of Things device management module includes:

[0063] A device deployment unit, which is used to deploy Internet of Things devices in the prohibited crossing area, and the Internet of Things devices are used to collect the dynamic positions of moving objects;

[0064] An archive binding unit, which is used to construct a device archive and bind the installation location of the Internet of Things device based on the coordinate grid;

[0065] The data processing module is used to regularize the perception time of the Internet of Things devices, generate crossing feature tags, construct a state perception sensitivity matrix and evaluate the crossing prevention fusion sensitivity;

[0066] Among them, the data processing module includes:

[0067] A time regularization unit, which is used to regularize the perception time of each Internet of Things device in the form of equal time nodes to obtain continuous perception time segments;

[0068] A feature generation unit, which is used to track the dynamic positions of moving objects within the perception time segment, establish crossing feature pairs and generate crossing feature tags;

[0069] A matrix construction unit, configured to construct a state perception sensitivity matrix based on out-of-bounds feature tags and generate a matrix shard stream;

[0070] A sensitivity calculation unit, configured to evaluate the out-of-bounds protection fusion sensitivity between perception time segments based on the matrix shard stream;

[0071] The early warning evaluation module is configured to evaluate the out-of-bounds protection effectiveness based on the out-of-bounds protection fusion sensitivity, extract and integrate the perception time segments, and analyze to obtain the locked detection protection range of the Internet of Things device;

[0072] Among them, the early warning evaluation module includes:

[0073] An effectiveness evaluation unit, configured to calculate the out-of-bounds protection effectiveness based on the out-of-bounds protection fusion sensitivity;

[0074] A segment integration unit, configured to extract and integrate the perception time segments according to the effectiveness threshold to obtain an integrated perception time range;

[0075] A protection range analysis unit, configured to screen an early warning center set centered on the Internet of Things device within the integrated perception time range to form the locked detection protection range of the Internet of Things device.

[0076] Please refer to Figure 1 , in the second embodiment: Provide a method for out-of-bounds protection warning of a substation virtual fence based on the Internet of Things, which is applicable to the first embodiment above. The method includes the following steps:

[0077] Step S1: Establish a virtual fence boundary coordinate set for the prohibited out-of-bounds area, deploy Internet of Things devices in the prohibited out-of-bounds area, construct a device archive, and bind the installation locations of the Internet of Things devices;

[0078] Exemplarily, generate a three-dimensional coordinate grid of the substation through lidar scanning to generate a virtual fence boundary coordinate set for the prohibited out-of-bounds area , where represents the i-th coordinate grid, is the coordinate value in the three-dimensional direction of the coordinate grid , and I represents the total number of coordinate grids;

[0079] Deploy Internet of Things devices in the prohibited out-of-bounds area. The Internet of Things devices are used to collect the dynamic positions of moving objects, construct a device archive, and based on the coordinate grid, bind the installation locations of the Internet of Things devices. Denote the j-th Internet of Things device as , then denote the installation location of the Internet of Things device as , where is the coordinate value in the three-dimensional direction of the Internet of Things device .

[0080] Step S2: Regularize the sensing times of each Internet of Things device at equal time nodes to obtain a set of continuous sensing time segments, establish an out-of-bounds feature pair, and generate an out-of-bounds feature label;

[0081] Exemplarily, regularize the sensing times of each Internet of Things device at equal time nodes to obtain a set of continuous sensing time segments , where represents the g-th sensing time segment, and G represents the total number of sensing time segments with a daily cycle period;

[0082] Internet of Things device tracks the dynamic position of a moving object within the sensing time segment . If it is tracked that the moving object enters the coordinate grid , establish an out-of-bounds feature pair and generate an out-of-bounds feature label , where represents the fixed distance between the Internet of Things device and the coordinate grid , and , represents the detection status value, and is equal to 0 or 1. When it is 0, it means that no moving object is detected, and when it is 1, it means that a moving object is detected.

[0083] Step S3: Based on the out-of-bounds feature label, construct a state perception sensitivity matrix and generate a matrix shard stream; based on the matrix shard stream, evaluate the out-of-bounds protection fusion sensitivity between sensing time segments;

[0084] Exemplarily, based on the out-of-bounds feature label, construct a state perception sensitivity matrix. The row number of the state perception sensitivity matrix is the coding number of the Internet of Things device, and the column number of the state perception sensitivity matrix is the coding number of the coordinate grid. Then the matrix position at the j-th row and the i-th column of the state perception sensitivity matrix is , and the matrix element value at the matrix position is . Then, the state perception sensitivity matrix generated within the sensing time segment is denoted as , and a matrix shard stream is obtained;

[0085] Based on the matrix shard stream, evaluate the out-of-bounds protection fusion sensitivity between the sensing time segment and the sensing time segment . In the formula, represents the state perception sensitivity matrix and the state perception sensitivity matrix The total number of matrix positions with a value of 1 after Boolean logic AND between them, represents the state perception sensitivity matrix and the state perception sensitivity matrix The total number of matrix positions with a value of 1 after Boolean logic OR between them.

[0086] Step S4: Based on the cross - boundary protection fusion sensitivity, evaluate the cross - boundary protection effectiveness between perception time segments, and extract the perception time segments to integrate the perception time range; Based on the integrated perception time range, analyze and obtain the locked - detection protection range of the Internet of Things device and output it;

[0087] Exemplarily, based on the cross - boundary protection fusion sensitivity, evaluate the perception time segment and the perception time segment The cross - boundary protection effectiveness between them , where is the mean of the cross - boundary protection fusion sensitivity, and , is the variance of the cross - boundary protection fusion sensitivity, and ;

[0088] Preset an effectiveness threshold. If the cross - boundary protection effectiveness is greater than or equal to the effectiveness threshold, then extract the perception time segment and the perception time segment , otherwise do not extract the perception time segment and the perception time segment ;

[0089] Perform segment fusion on all the extracted perception time segments, set a time sliding window, and translate the time sliding window. If the perception time segments within the time sliding window are continuous, then perform time - scale integration on the continuous perception time segments to obtain the integrated perception time range;

[0090] Select the integrated perception time range with the largest time span, and centered on the Internet of Things device collect the cross - boundary feature pairs corresponding to the matrix positions in each state perception sensitivity matrix , to form the early - warning center set of the Internet of Things device , denoted as , and based on the cross - boundary feature label , screen the coordinate grid with the shortest distance between the early - warning center set and the Internet of Things device , and the coordinate grid screened out forms the Internet of Things device ​The locking detection protection range to guide the Internet of Things devices to perform the detection protection in the next cycle period.

[0091] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0092] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A substation virtual fence crossing protection early warning method based on the Internet of Things is characterized by: The method comprises the following steps: Step S1: Establish a virtual fence boundary coordinate set for the prohibited crossing area, deploy IoT devices in the prohibited crossing area, build a device archive, and bind the installation location of the IoT devices; Step S2: Regularize the perception time of each IoT device in the form of equal time nodes to obtain a set of continuous perception time segments, establish an out-of-bounds feature pair, and generate an out-of-bounds feature label; Step S3: Based on the cross-border feature labels, a state perception sensitivity matrix is constructed to generate a matrix slice flow; based on the matrix slice flow, the cross-border protection fusion sensitivity between the perception time segments is evaluated; Step S4: Based on the cross-border protection fusion sensitivity, the cross-border protection effectiveness between the sensing time segments is evaluated, and the sensing time segments are extracted to integrate the sensing time range; based on the integrated sensing time range, the lock detection protection range of the IoT device is analyzed and output; The specific implementation process of step S2 includes: The perception time of each IoT device is regularized in the form of equal time nodes to obtain the continuous perception time segment set {t g |g∈[1,G]}, where t g represents the gth perception time segment, G represents the total number of perception time segments with a day cycle; The jth IoT device F j In the perception time segment t g Track the dynamic position of the moving object. If the moving object is tracked to enter the i-th coordinate grid R i When F is established, the cross-boundary feature pair F j |R i , and generate the out-of-bounds feature label t g (F j |R i ):[d(F j |R i ), S(t g )], where d(F j |R i ) represents IoT device F j with the coordinate grid R i A fixed distance between S(t g ) represents the detection state value, and S(t g ) is equal to 0 or 1. When it is 0, it means no moving object is detected, and when it is 1, it means a moving object is detected. (x i ,y i , z i ) is the coordinate grid R i The coordinate values in the three-dimensional direction, (x j ,y j , z j ) is the IoT device F j The coordinate values in the three-dimensional direction; The specific implementation process of step S3 includes: Based on the out-of-bounds feature label, a state perception sensitivity matrix is constructed. The row number of the state perception sensitivity matrix is the code number of the IoT device, and the column number of the state perception sensitivity matrix is the code number of the coordinate grid. Then the matrix position of the jth row and the ith column of the state perception sensitivity matrix is (F j , R i ), matrix position (F j , R i ) is the matrix element value S(t g ), then in the perception time segment t g The state perception sensitivity matrix generated internally is denoted as K(t g ), and obtain the matrix slice flow {K(t g )|g∈[1,G]}; Based on the matrix slice flow, the perception time segment t is evaluated g and the perceived time segment t g+1 Cross-border protection fusion sensitivity Where, NUM[K(t g )∩K(t g+1 )] represents the state perception sensitivity matrix K(t g ) and the state perception sensitivity matrix K(t g+1 ) contains the total number of matrix positions with a value of 1 after the Boolean logical AND between g )∪K(t g+1 )] represents the state perception sensitivity matrix K(t g ) and the state perception sensitivity matrix K(t g+1 ) contains the total number of matrix positions with a value of 1.

2. The method for early warning of substation virtual fence crossing protection based on the Internet of Things according to claim 1 is characterized in that: The specific implementation process of step S1 includes: The three-dimensional coordinate grid of the substation is generated by laser radar scanning, and the virtual fence boundary coordinate set {R i =(x i ,y i , z i )|i∈[1,I]}, where R i represents the i-th coordinate grid, and I represents the total number of coordinate grids; Deploy IoT devices in prohibited crossing-border areas. The IoT devices are used to collect the dynamic positions of moving objects, build a device archive, bind the installation positions of IoT devices based on the coordinate grid, and record the jth IoT device as F. j , then the IoT device F j The installation position is marked as F j =(x j ,y j , z j ).

3. The method for early warning of substation virtual fence crossing protection based on the Internet of Things according to claim 2 is characterized in that: The specific implementation process of step S4 includes: Based on the cross-border protection fusion sensitivity, the perception time segment t is evaluated g and the perceived time segment t g+1 Cross-border protection effectiveness Where μ is the mean value of the cross-border protection fusion sensitivity, and σ 2 is the variance of the cross-border protection fusion sensitivity, and Preset validity threshold, if the protection validity BE(t g →t g+1 ) is greater than or equal to the validity threshold, then the perception time segment t is extracted g and the perception time segment t g+1 , otherwise the perception time segment t is not extracted g and the perception time segment t g+1 ; Perform segment fusion on all extracted perception time segments, set a time sliding window, and translate the time sliding window. If the perception time segments within the time sliding window are continuous, then integrate the continuous perception time segments in time scale to obtain an integrated perception time range. Select the integrated perception time range with the maximum time span, and use IoT device F j As the center, collect the out-of-bounds feature pairs F corresponding to the matrix position in each state perception sensitivity matrix j |R i , forming the IoT device F j The early warning center set is denoted as W(F j )={R i |i∈[1,I]}, and based on the out-of-bounds feature label t g (F j |R i ), screening early warning center centralized with IoT devices F j The shortest distance between min{d(F j |R i )|R i ∈W(F j )} coordinate grid R i , the selected coordinate grid R i Composing IoT devices F j Lock detection protection range to guide IoT devices F j Carry out detection and protection for the next cycle.

4. A substation virtual fence crossing protection warning system based on the Internet of Things, which executes the substation virtual fence crossing protection warning method according to any one of claims 1 to 3, characterized in that: The system includes: a virtual fence construction module, an IoT device management module, a data processing module, and an early warning assessment module; The virtual fence construction module is used to establish a virtual fence boundary coordinate set for a prohibited crossing area; The IoT device management module is used to deploy IoT devices in prohibited crossing-border areas, build a device archive and bind device installation locations; The data processing module is used to time the perception time of IoT devices, generate cross-border feature labels, construct a state perception sensitivity matrix and evaluate the cross-border protection fusion sensitivity; The early warning assessment module is used to evaluate the effectiveness of cross-border protection based on the cross-border protection fusion sensitivity, extract and integrate the perception time segments, and analyze to obtain the locking detection protection range of the Internet of Things device.

5. The IoT-based substation virtual fence crossing protection warning system according to claim 4 is characterized in that: The virtual fence construction module includes: A coordinate generation unit, configured to generate a three-dimensional coordinate grid of the substation through laser radar scanning and generate a virtual fence boundary coordinate set of the prohibited crossing area; The boundary definition unit is used to define the virtual fence boundary of the prohibited crossing area based on the three-dimensional coordinate grid.

6. The IoT-based substation virtual fence crossing protection warning system according to claim 4 is characterized in that: The Internet of Things device management module includes: A device deployment unit, configured to deploy IoT devices in prohibited crossing areas, wherein the IoT devices are configured to collect dynamic positions of moving objects; The archive binding unit is used to build a device archive library and bind the installation locations of IoT devices based on the coordinate grid.

7. The IoT-based substation virtual fence crossing protection warning system according to claim 4 is characterized in that: The data processing module includes: The time timing unit is used to time the perception time of each IoT device in a manner of equal time nodes to obtain continuous perception time segments; A feature generation unit is used to track the dynamic position of the moving object within the sensing time segment, establish out-of-bounds feature pairs and generate out-of-bounds feature labels; A matrix construction unit, used to construct a state-aware sensitivity matrix based on out-of-bounds feature labels and generate a matrix slice flow; The sensitivity calculation unit is used to evaluate the cross-border protection fusion sensitivity between the perception time segments based on the matrix slice flow.

8. The IoT-based substation virtual fence crossing protection warning system according to claim 4 is characterized in that: The early warning assessment module includes: An effectiveness evaluation unit, configured to calculate the effectiveness of cross-border protection based on cross-border protection fusion sensitivity; a segment integration unit, configured to extract and integrate the perception time segments according to the validity threshold to obtain an integrated perception time range; The protection range analysis unit is used to screen the warning center set with the IoT device as the center within the integrated perception time range to form the locking detection protection range of the IoT device.

Images