A 3D Gaussian point cloud electronic fence alarm system for substations

Abstract

This invention provides a 3D Gaussian point cloud electronic fence alarm system for substations, comprising: an intelligent calibration module that calibrates the electronic fence in three-dimensional space based on a 3DGS model of the substation, determining the 3D Gaussian point cloud electronic fence of the substation; a spatial positioning module that combines multiple positioning methods to analyze the position of a work object, determining the position information of the work object; a boundary crossing analysis module that combines the position information of the work object with the 3D Gaussian point cloud electronic fence of the substation to perform spatial boundary crossing analysis and obtain the boundary crossing analysis result; and an alarm push module that performs multi-level linkage alarms and pushes based on the boundary crossing analysis result. This invention effectively improves anti-interference capability, enables accurate acquisition of the position information of the work object, thereby allowing for more accurate spatial boundary crossing analysis, avoiding false alarms or missed alarms, improving the accuracy of the 3D Gaussian point cloud electronic fence alarm system for substations, and reducing the false alarm rate.

Description

Technical Field

[0001] This invention relates to the field of security fence technology, and in particular to a 3D Gaussian point cloud electronic fence alarm system for substations. Background Technology

[0002] Personal safety, power grid safety, and equipment safety have always been core issues in power work. During substation renovation or expansion construction, the scope of construction activities is often difficult to control. Traditional protection solutions typically rely on physical fences for obstruction, but these lack real-time warnings, and their limited height and coverage area prevent comprehensive protection and thus fail to guarantee substation safety. To overcome the shortcomings of physical fence protection solutions, electronic fences have gradually been proposed.

[0003] Currently, existing electronic fence systems have poor anti-interference capabilities and are prone to false alarms or missed alarms due to interference factors, resulting in a high false alarm rate and failing to adequately guarantee the safety of substations. Therefore, there is an urgent need for a 3D Gaussian point cloud electronic fence alarm system for substations to solve the above-mentioned technical problems. Summary of the Invention

[0004] The purpose of this invention is to provide a 3D Gaussian point cloud electronic fence alarm system for substations, which improves anti-interference capabilities, enables more accurate location determination and spatial boundary analysis, improves the accuracy of the 3D Gaussian point cloud electronic fence alarm system for substations, avoids false alarms or missed alarms, and reduces the false alarm rate.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a 3D Gaussian point cloud electronic fence alarm system for substations, comprising: The intelligent calibration module is used to calibrate the three-dimensional spatial shape of the electronic fence for the 3DGS model of the substation, and to determine the 3D Gaussian point cloud electronic fence of the substation. The spatial positioning module is used to combine multiple positioning methods to perform position analysis on the work object and determine the position information of the work object. The boundary crossing analysis module is used to combine the location information of the work object with the 3D Gaussian point cloud electronic fence of the substation to perform spatial boundary crossing analysis and judgment, and obtain the boundary crossing analysis results. The alarm push module is used to generate multi-level linked alarms and push notifications based on the boundary analysis results.

[0006] Furthermore, when the intelligent calibration module calibrates the electronic fence for the 3DGS model of the substation, it includes: Based on the 3DGS model of the substation, the three-dimensional spatial safety zone is determined by analyzing the substation's operating procedures and safety specifications. Based on the 3DGS model of the substation, the hazard level and safety distance requirements of the power equipment are analyzed to determine the three-dimensional spatial hazard area; In the 3DGS model of the substation, the boundaries of the three-dimensional safe area and the three-dimensional dangerous area are determined, and electronic fences are established according to the boundaries to obtain the safe area electronic fence and the dangerous area electronic fence. Identifier configuration and attribute labeling are performed for electronic fences in safe areas and electronic fences in dangerous areas respectively.

[0007] Furthermore, the intelligent calibration module also performs dynamic calibration of the electronic fence based on the substation's maintenance and expansion construction, including: Determine the maintenance plan or expansion construction plan, and determine the target area based on the maintenance plan or expansion construction plan; Establish temporary electronic fences for the target area; The temporary electronic fence was updated and adjusted by combining it with the 3D Gaussian point cloud electronic fence of the substation.

[0008] Furthermore, the multiple positioning methods include indoor positioning and outdoor positioning. The spatial positioning module combines indoor and outdoor positioning methods to achieve indoor-outdoor fusion positioning. When the spatial positioning module analyzes the location of the work object using multiple positioning methods, it uses an intelligent switching algorithm to determine the indoor positioning method as the target positioning method when the work object enters the room and uses UWB positioning technology to determine the location of the work object. When the work object is outdoors, it determines the outdoor positioning method as the target positioning method and uses an RTK positioning terminal that supports the BeiDou satellite navigation system to obtain the location of the work object.

[0009] Furthermore, the spatial positioning module also performs spatial monitoring of the substation using an image acquisition device, including: The substation is monitored by acquiring images through an image acquisition device. Based on the substation monitoring images, identification and analysis are performed to determine whether there are living organisms or objects in the substation, and the identification and analysis results are obtained. Based on the identification and analysis results, when a living being or an object appears in the substation, the location of the living being or object is determined, and the target monitoring and identification location is locked.

[0010] Furthermore, the boundary crossing analysis module includes: an information processing unit and a data analysis unit; The information processing unit is used to receive the location information of the work object, and to perform fusion processing on the location information of the work object and the 3DGS model of the substation to obtain the location information of the work object in the 3DGS model of the substation. The data analysis unit is used to perform spatial geometric relationship analysis based on the location information of the working object in the 3DGS model of the substation and the electronic fence of the 3D Gaussian point cloud of the substation, and to obtain the boundary analysis results.

[0011] Furthermore, the data analysis unit performs spatial geometric relationship analysis based on the data processed by the work object location and combined with the 3D Gaussian point cloud electronic fence of the substation, including: Determine the boundary of the electronic fence area based on the 3D Gaussian point cloud of the substation; Based on the location information of the work object in the 3DGS model of the substation and the boundary of the electronic fence area, a positional relationship analysis is performed to determine the current area to which the work object belongs. Based on the current region to which the work object belongs, analyze whether the work object has entered the danger zone to obtain the boundary crossing analysis results.

[0012] Furthermore, spatial geometric relationship analysis is conducted based on the location information of the work object in the 3DGS model of the substation, combined with the boundary of the electronic fence area, including: Based on the boundaries of the electronic fence area, analyze the electronic fence area to determine the geometric type of the electronic fence area; When the geometry of the electronic fence area is a convex geometry, further analysis is performed on the convex geometry to determine whether it is a regular geometry. If the convex geometry is a regular geometry, the cross-sectional plane is determined based on the boundary of the electronic fence area. Based on the cross-sectional plane and the position information of the work object in the substation 3DGS model, the positional relationship between the work object and the electronic fence area is determined, and the current region to which the work object belongs is obtained. When the convex geometry is not a regular geometry, the SDF method is used to perform positional relationship analysis on the convex geometry to determine the electronic fence area where the work object is located, and the current region to which the work object belongs is obtained. When the geometry of the electronic fence area is concave, ray determination is performed based on the position information of the work object in the substation 3DGS model. The number of intersections between the ray analysis and the boundary of the electronic fence area is used to determine the positional relationship between the work object and the electronic fence area, thus obtaining the current area to which the work object belongs.

[0013] Furthermore, the alarm push module includes: an identification and monitoring unit and an alarm push unit. The identification and monitoring unit is used to identify and monitor the boundary analysis results of the boundary analysis module, and send a trigger signal to the alarm push unit according to the identification and monitoring results. The alarm push unit is used to trigger multi-level linkage alarms based on the trigger signal, and at the same time obtain related information for information organization and push.

[0014] Furthermore, when the alarm push unit performs multi-level linkage alarms based on the trigger signal, it issues warnings to work objects entering dangerous areas according to the linkage strategy, including: issuing alarm prompts through pop-up windows based on the trigger signal and sending alarm notifications to relevant work objects; simultaneously, performing location analysis on work objects entering dangerous areas to determine the target location, and issuing on-site warnings based on the target location through an audible and visual alarm device.

[0015] This invention enables the intelligent application of electronic fences in substations, allowing for the creation of electronic fences with complex three-dimensional spatial shapes. This ensures the safety of personnel while preventing damage to work vehicles or electrical equipment, effectively improving anti-interference capabilities, and enabling accurate acquisition of the location information of work objects. This allows for more accurate and rapid spatial boundary analysis, improving the accuracy of the substation 3D Gaussian point cloud electronic fence alarm system, avoiding false alarms or missed alarms, and ensuring that the substation 3D Gaussian point cloud electronic fence alarm system can better protect the safety of the substation.

[0016] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the application.

[0017] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0018] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the 3D Gaussian point cloud electronic fence alarm system for substations as described in this invention. Figure 2 This is a schematic diagram illustrating the steps of the intelligent calibration module in the substation 3D Gaussian point cloud electronic fence alarm system described in this invention. Figure 3 This is a schematic diagram of the boundary crossing analysis module in the substation 3D Gaussian point cloud electronic fence alarm system described in this invention. Figure 4 This is a schematic diagram of the alarm push module in the substation 3D Gaussian point cloud electronic fence alarm system described in this invention. Detailed Implementation

[0019] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0020] like Figure 1 As shown in the figure, this embodiment of the invention provides a 3D Gaussian point cloud electronic fence alarm system for substations, including: an intelligent calibration module, a spatial positioning module, a boundary crossing analysis module, and an alarm push module.

[0021] The intelligent calibration module is used to calibrate the three-dimensional spatial shape of the electronic fence for the 3DGS model of the substation, and to determine the 3D Gaussian point cloud electronic fence of the substation.

[0022] Among them, when the intelligent calibration module calibrates the electronic fence of the three-dimensional spatial shape of the 3DGS model of the substation, it defines safe areas / dangerous areas of complex three-dimensional spatial shapes in the 3DGS model of the substation, and then performs electronic fence calibration according to the defined safe areas / dangerous areas, and then uses the calibrated 3DGS model of the substation as the 3D Gaussian point cloud electronic fence of the substation.

[0023] The spatial positioning module is used to combine multiple positioning methods to perform position analysis on the work object and determine the position information of the work object. The positioning technologies include RTK positioning and UWB positioning. The work objects refer to personnel or vehicles working in the substation. When the spatial positioning module combines multiple positioning methods to analyze the location of the work objects, different positioning methods are used to determine the location information of the work objects. Typically, positioning devices are installed in safety helmets, wristbands, or other items that workers must wear while working in the substation, or deployed on work vehicles. The location information of the work objects is their position information in three-dimensional space.

[0024] The boundary crossing analysis module is used to combine the location information of the work object with the 3D Gaussian point cloud electronic fence of the substation to perform boundary crossing analysis and obtain the boundary crossing analysis results.

[0025] Among them, the boundary crossing analysis module integrates high-precision work object location information and the spatial computing capabilities of the substation 3DGS model. Based on the real-time work object location information and the substation 3D Gaussian point cloud electronic fence, it performs boundary crossing analysis and judgment to determine whether the work object has entered the danger zone, thereby obtaining the boundary crossing analysis result.

[0026] The alarm push module is used to generate multi-level linked alarms and push notifications based on the boundary analysis results.

[0027] Specifically, when the boundary analysis indicates that a work object has entered a dangerous area, the alarm push module combines multiple warning methods to issue an alarm and push notification. These warning methods include sound, light, and vibration.

[0028] The aforementioned intelligent application of electronic fences in substations enables the creation of complex three-dimensional spatial shapes for these fences, ensuring worker safety while preventing damage to work vehicles or electrical equipment. It effectively improves anti-interference capabilities, accurately acquires the location information of work objects, and allows for more accurate and rapid spatial boundary analysis. This enhances the accuracy of the substation's 3D Gaussian point cloud electronic fence alarm system, preventing false alarms or missed alarms, and ultimately providing better substation safety. The intelligent calibration module calibrates the electronic fence based on the substation's 3DGS model. Within this model, it can define electronic fences of arbitrarily complex three-dimensional shapes with clear boundaries, incorporating live equipment and hazardous areas. This improves the flexibility of the electronic fence, enabling precise spatial-level protection of live equipment and hazardous areas. The spatial positioning module combines multiple positioning methods to analyze the location of work objects, allowing for location analysis even in complex environments. This enhances anti-interference capabilities, ensures the accuracy of work object location information, and provides a basis for the boundary analysis module to perform spatial boundary analysis and judgment. By combining the location information of work objects with the 3D Gaussian point cloud electronic fence of the substation through the boundary analysis module, spatial boundary crossing analysis and judgment are achieved, realizing efficient boundary crossing judgment. This allows for timely alarms and push notifications based on the boundary crossing analysis results. The alarm push module enables multi-level linkage alarms and push notifications, effectively increasing the vigilance of relevant work objects and enabling them to take timely measures in response to alarms and push notifications, thereby improving the inherent safety level of the substation.

[0029] In one embodiment provided by the present invention, such as Figure 2 As shown, the intelligent calibration module, when calibrating the electronic fence for the 3DGS model of the substation, includes: A1. Based on the 3DGS model of the substation, the three-dimensional spatial safety zone is determined by analyzing the substation's operating procedures and safety specifications.

[0030] When analyzing a substation based on its 3DGS model according to its operating procedures and safety regulations, the safety space range is determined within the 3DGS model according to these regulations, thus obtaining a three-dimensional safety area. This safety space range includes equipment operating areas, personnel passageways, etc.

[0031] A2. Based on the 3DGS model of the substation, the hazard level and safety distance requirements of the power equipment are analyzed to determine the three-dimensional spatial hazard area.

[0032] When analyzing power equipment based on the hazard level and safety distance requirements of the substation 3DGS model, the hazardous space range is determined in the substation 3DGS model according to the hazard level and safety distance requirements of the equipment, thereby obtaining the three-dimensional hazardous area. The hazardous space range includes: the area around high-voltage equipment, hazardous chemical storage areas, etc.

[0033] A3. In the 3DGS model of the substation, the boundaries of the three-dimensional safe area and the three-dimensional dangerous area are determined, and electronic fences are established according to the boundaries to obtain the electronic fences of the safe area and the electronic fences of the dangerous area.

[0034] In the 3DGS model of the substation, when determining the boundaries of the three-dimensional safe area and the three-dimensional dangerous area, boundary feature point analysis is performed on the three-dimensional safe area and the three-dimensional dangerous area respectively. The point cloud density change points on the boundary of the three-dimensional safe area and the three-dimensional dangerous area in the 3DGS model are taken as boundary feature points. The geometric connection relationship between the straight line segments or curve segments between adjacent boundary feature points is analyzed to determine the boundary shape of the electronic fence, and then the electronic fence of the safe area and the electronic fence of the dangerous area are established.

[0035] A4. Configure identifiers and label attributes for electronic fences in safe areas and electronic fences in dangerous areas respectively.

[0036] Specifically, when configuring identifiers and labeling attributes for safe and hazardous electronic fences, identifiers are assigned to each safe and hazardous area based on the respective electronic fences. These identifiers are then matched and labeled in the substation 3DGS model within the safe or hazardous electronic fences. Simultaneously, attribute analysis is performed on both safe and hazardous electronic fences to determine their regional attributes and the types of work objects allowed to enter them, thus obtaining regional attribute data. Furthermore, spatial analysis is conducted to determine the height, width, and other spatial data of both safe and hazardous electronic fences, obtaining regional spatial data. Finally, the regional attribute data and spatial parameters are matched and labeled in the substation 3DGS model within the safe or hazardous electronic fences.

[0037] The above-mentioned approach, through safety area analysis based on substation operating procedures and safety specifications, and hazardous area analysis based on the hazard level and safety distance requirements of electrical equipment, precisely defines safe / hazardous areas in complex three-dimensional spatial shapes. This clearly identifies which areas within the substation are safe and which are hazardous, achieving precise spatial-level protection for energized equipment and hazardous areas. It also ensures timely detection when work objects cross into hazardous areas, guaranteeing substation safety. Furthermore, by configuring identifiers and labeling attributes for safe and hazardous electronic fences respectively, the characteristics of the corresponding areas are clearly understood, providing assurance for spatial boundary crossing analysis and improving the accuracy of the analysis results.

[0038] In one embodiment of the present invention, the intelligent calibration module further performs dynamic calibration of the electronic fence based on the maintenance and expansion construction of the substation, including: Determine the maintenance plan or expansion construction plan, and determine the target area based on the maintenance plan or expansion construction plan.

[0039] The determination of the target area based on the maintenance plan or expansion construction plan is divided into two categories: determining the target area based on the maintenance plan and determining the target area based on the expansion construction plan. When determining the target area based on the maintenance plan, the maintenance object is identified according to the plan, and the impact range of maintenance on the maintenance object is determined based on the maintenance object and the maintenance operation analysis. This impact range is then defined as the target area. When determining the target area based on the expansion construction plan, the expansion construction location and scale are determined according to the expansion construction plan. Then, the impact range of the expansion construction is determined based on the expansion construction location and the expansion construction scale, and this impact range is then defined as the target area.

[0040] Establish temporary electronic fences for the target area.

[0041] When establishing a temporary electronic fence for a target area, a preliminary overall temporary electronic fence is established. Then, spatial analysis is performed within this overall temporary electronic fence, incorporating maintenance and expansion plans, to determine safe and hazardous areas and define their spatial data. Finally, based on this spatial data, the overall temporary electronic fence is further subdivided and its area attributes are labeled to obtain the final electronic fence. Here, spatial data includes height, depth, and width.

[0042] The temporary electronic fence was updated and adjusted by combining it with the 3D Gaussian point cloud electronic fence of the substation.

[0043] When updating and adjusting temporary electronic fences in conjunction with the 3D Gaussian point cloud electronic fence of the substation, the overlapping area of ​​the temporary electronic fence within the 3D Gaussian point cloud electronic fence is analyzed to determine the overlapping area and obtain its attributes. Then, it is analyzed whether there are attribute conflicts between the attributes of the temporary electronic fence and the attributes of the overlapping area. If there is an attribute conflict, the overlapping area is temporarily disabled in the 3D Gaussian point cloud electronic fence of the substation, and the temporary electronic fence is superimposed onto the overlapping area and merged with the electronic fences in other areas to form a new 3D Gaussian point cloud electronic fence for the substation. If there is no attribute conflict between the attributes of the temporary electronic fence and the attributes of the overlapping area, the overlapping area is used as a marked area, and no adjustment is needed for the 3D Gaussian point cloud electronic fence of the substation.

[0044] Furthermore, after the maintenance or expansion work is completed, the temporary electronic fence is removed, and the electronic fence for the overlay area or the marked area is reconfirmed and restored. Specifically, during the electronic fence confirmation and restoration process, if the overlay area or marked area has not undergone attribute changes after the maintenance or expansion work, the original 3D Gaussian point cloud electronic fence of the substation is directly restored. If the overlay area or marked area has undergone attribute changes after the maintenance or expansion work, the electronic fence for the overlay area or marked area is recreated, and then the original 3D Gaussian point cloud electronic fence of the substation is updated based on the recreated electronic fence.

[0045] The aforementioned dynamic calibration of the electronic fence based on substation maintenance and expansion construction allows for flexible adjustment of the 3D Gaussian point cloud electronic fence during substation maintenance or expansion, improving its dynamic flexibility. This not only avoids safety hazards during maintenance or expansion, ensuring substation safety, but also prevents accidental activation of the electronic fence during maintenance or expansion, thus avoiding confusion in multi-level alarm and push notifications and ensuring the accuracy and stability of the substation's 3D Gaussian point cloud electronic fence alarm system. Furthermore, by updating and adjusting temporary electronic fences in conjunction with the substation's 3D Gaussian point cloud electronic fence, conflicts between temporary electronic fences and safe / dangerous area electronic fences are avoided, preventing them from affecting spatial boundary analysis and judgment. This provides assurance for boundary analysis results and improves the accuracy of the substation's 3D Gaussian point cloud electronic fence alarm system.

[0046] In one embodiment of the present invention, the multiple positioning methods include: indoor positioning and outdoor positioning. The spatial positioning module combines indoor and outdoor positioning methods to achieve indoor-outdoor fusion positioning. When the spatial positioning module performs position analysis on the work object using multiple positioning methods, it determines the indoor positioning method as the target positioning method when the work object enters the room using an intelligent switching algorithm, and uses UWB positioning technology to determine the position of the work object. When the work object is outdoors, it determines the outdoor positioning method as the target positioning method, and uses an RTK positioning terminal that supports the BeiDou satellite navigation system to obtain the position of the work object.

[0047] The indoor positioning method is UWB (Ultra-Wideband) positioning technology, while the outdoor positioning method is RTK (Real-Time Kinematic Differential) positioning technology based on the BeiDou Navigation Satellite System. The intelligent switching algorithm automatically switches between indoor and outdoor positioning methods depending on whether the work object is indoors. Specifically, the substation's 3DGS model is pre-divided into indoor and outdoor areas. The preliminary position of the work object is acquired in real time and compared with the indoor and outdoor areas. If the preliminary position falls within the indoor area, the work object is determined to be indoors, and the indoor positioning method is set as the target positioning method. If the preliminary position falls within the outdoor area, the work object is determined to be outdoors, and the outdoor positioning method is set as the target positioning method. When using UWB positioning technology to determine the work object's location, the UWB tag worn by the work object communicates with the base station, and 3D positioning analysis is performed based on the communication data to obtain the work object's location information. When the location of the work object is obtained using an RTK positioning terminal supporting the BeiDou satellite navigation system, the BeiDou satellite signal received by the positioning terminal worn by the work object is combined with RTK differential technology for signal analysis. Using the carrier phase differential observation values ​​between the base station and the rover, the three-dimensional coordinates of the rover are calculated in real time to obtain the work object's location information. Here, UWB tags and positioning terminals are installed in wearable devices such as safety helmets and wristbands.

[0048] The aforementioned spatial positioning module can acquire high-precision three-dimensional spatial positions in real time, ensuring accurate positioning of work objects in complex scenarios such as substations. Indoor and outdoor fusion positioning, utilizing both indoor and outdoor positioning methods, achieves seamless switching and high-precision positioning, enabling accurate acquisition of work object location information regardless of whether it is indoors or outdoors. Furthermore, UWB positioning technology effectively improves anti-interference capabilities, achieving high-precision work object location determination. Additionally, the use of a BeiDou high-precision positioning terminal combined with RTK technology to eliminate satellite signal errors using differential GPS signals further enhances the accuracy of the determined work object location. This allows the boundary crossing analysis module to receive high-precision three-dimensional spatial positions of work objects in real time, ensuring accurate boundary crossing analysis and judgment.

[0049] In one embodiment of the present invention, the spatial positioning module further performs spatial monitoring of the substation via an image acquisition device, including: The substation is monitored by acquiring images through an image acquisition device.

[0050] The image acquisition device can be a camera or other device capable of continuous image acquisition. When acquiring images of a substation using this device, comprehensive monitoring and image acquisition are performed to obtain substation monitoring images.

[0051] Based on the substation monitoring images, identification and analysis are performed to determine whether there are living organisms or objects in the substation, and the identification and analysis results are obtained.

[0052] Among them, living organisms include both humans and animals. When performing identification and analysis based on substation monitoring images, the system analyzes and determines whether humans or animals are present in the substation, and obtains the identification and analysis results based on the analyzed data. Objects refer to inanimate objects, such as plastic bags, dust covers, and tree branches.

[0053] Based on the identification and analysis results, when a living being or an object appears in the substation, the location of the living being or object is determined, and the target monitoring and identification location is locked.

[0054] If the identification and analysis results indicate that no living organisms were found in the substation, this step is unnecessary. When determining the location of any living organisms or objects that are found, the relative positions of the organisms and surrounding power equipment in the substation monitoring images are used to determine their location, thus locking the target monitoring and identification position and obtaining the target location information.

[0055] Furthermore, when the boundary crossing analysis module performs spatial boundary crossing analysis based on the target organism's location and the substation's 3D Gaussian point cloud electronic fence, it first determines whether the target location information is the location of a target organism or a target object. If the target location information is the location of a target organism, it further determines whether the target organism is a human or an animal. If the target organism is a human, it combines the location information of the work object to determine whether it is a work object. If the target organism is not a work object, it combines the target organism's location information with the substation's 3D Gaussian point cloud electronic fence to perform spatial boundary crossing analysis and obtain the boundary crossing analysis result. If the target organism is an animal, it combines the animal's target organism location information with the substation's 3D Gaussian point cloud electronic fence to perform spatial boundary crossing analysis and obtain the boundary crossing analysis result. Here, the work object refers to a wearable device equipped with an installed UWB tag that communicates with the base station.

[0056] The aforementioned spatial monitoring of the substation using image acquisition devices enables the spatial positioning module to promptly detect the presence of living organisms within the substation, thereby preventing safety issues and damage to electrical equipment and ensuring the substation's safety. Furthermore, when the target organism is a human, it determines whether they are a work object, preventing the boundary analysis module from repeatedly analyzing the same work object and effectively improving the efficiency of spatial boundary analysis judgment.

[0057] In one embodiment provided by the present invention, such as Figure 3 As shown, the boundary crossing analysis module includes: an information processing unit and a data analysis unit; The information processing unit is used to receive the location information of the work object, and to perform fusion processing on the location information of the work object and the 3DGS model of the substation to obtain the location information of the work object in the 3DGS model of the substation.

[0058] The process of fusing the work object location information with the substation 3DGS model includes: determining the positioning coordinate system for the work object location information and simultaneously determining the model coordinate system for the substation 3DGS model; analyzing the correlation between the positioning coordinate system and the model coordinate system; performing coordinate transformation on the work object location information according to the correlation to obtain the work object location processing data; and using the point cloud data in the substation 3DGS model, combined with spatial geometric algorithms, calculating the precise position of the work object in the model space to obtain the work object's position information in the substation 3DGS model.

[0059] The data analysis unit is used to perform spatial geometric relationship analysis based on the location information of the working object in the 3DGS model of the substation and the electronic fence of the 3D Gaussian point cloud of the substation, and to obtain the boundary analysis results.

[0060] In particular, when performing spatial geometric relationship analysis based on the location information of the work object in the 3DGS model of the substation and the electronic fence of the 3D Gaussian point cloud of the substation, the attributes of the area where the work object is located are clarified, and the relationship between the work object and the electronic fence of the dangerous area is clarified, thereby determining whether the work object has crossed the boundary and obtaining the boundary crossing analysis results.

[0061] The aforementioned information processing unit receives and processes the location information of the work object, enabling the location information to be adapted to the 3DGS model of the substation during spatial boundary crossing analysis. This ensures the consistency of the spatial boundary crossing analysis data, avoids deviations in the boundary crossing analysis results due to inconsistent coordinate systems, and prevents the alarm push module from affecting the feedback. This ensures the accuracy of multi-level linkage alarms and push notifications. Furthermore, the data analysis unit can efficiently perform location calculations and boundary judgments, improving the efficiency of spatial boundary crossing analysis and ensuring the accuracy of the boundary crossing analysis results. Consequently, the alarm push module can promptly perform multi-level linkage alarms and push notifications when the work object crosses the boundary, ensuring the safety of the substation.

[0062] In one embodiment of the present invention, the data analysis unit performs spatial geometric relationship analysis based on the location information of the working object in the 3DGS model of the substation and the electronic fence of the 3D Gaussian point cloud of the substation, including: The boundaries of the electronic fence area are determined based on the 3D Gaussian point cloud of the substation.

[0063] When obtaining the boundary of the electronic fence area, the boundary analysis of the 3D Gaussian point cloud electronic fence of the substation is performed sequentially according to the area blocks to determine the boundary of the electronic fence area.

[0064] The location relationship analysis was performed on the location information of the work object in the 3DGS model of the substation and the boundary of the electronic fence area to determine the current area to which the work object belongs.

[0065] Specifically, when analyzing the positional relationship between the work object and the electronic fence area, the positional information of the work object in the 3DGS model of the substation is combined with the boundary analysis of the corresponding electronic fence area to determine the positional relationship between the work object and the electronic fence area. Thus, the current area to which the work object belongs is determined based on the positional relationship between the work object and the electronic fence area.

[0066] Based on the current region to which the work object belongs, analyze whether the work object has entered the danger zone to obtain the boundary crossing analysis results.

[0067] Specifically, when determining the boundary crossing analysis result based on the current region to which the work object belongs, the corresponding annotation information is retrieved for the current region to which the work object belongs. Based on the annotation information, it is determined whether the current region to which the work object belongs is a dangerous region. If the current region to which the work object belongs is a dangerous region, the boundary crossing analysis result is that the work object has entered the dangerous region; otherwise, the boundary crossing analysis result is that the work object is in the safe region.

[0068] Furthermore, after identifying the safe zone and the dangerous zone electronic fences, a positional relationship analysis is performed on them to determine adjacent safe and dangerous zones and obtain the interface between them. A warning range is then determined for adjacent safe zones based on the interface, with a preset range near the dangerous zone designated as the warning zone. When the boundary analysis result indicates that the work object is within a safe zone, it is analyzed whether a warning zone exists within the work object's current area. If a warning zone exists, further analysis is performed to determine if the work object is within the warning zone, resulting in a secondary analysis result. Based on this secondary analysis result, if the work object is within the warning zone, a hazard warning is issued via an alarm push module.

[0069] The aforementioned data analysis unit performs relational analysis within a spatial region, integrating BeiDou high-precision positioning and 3DGS spatial computing capabilities. This enables precise real-time perception of the work object's position within a complex three-dimensional space and efficient boundary crossing judgment, improving the accuracy of boundary crossing analysis results and reducing false alarms and missed alarms. Furthermore, by determining the boundary of the electronic fence area based on the substation's 3D Gaussian point cloud electronic fence, spatial geometric relationship analysis based on the electronic fence area can be performed using quantitative information, improving the accuracy of positional relationship analysis and ensuring the accuracy of boundary crossing analysis results. Moreover, when determining the current region to which the work object belongs by combining its position information in the substation's 3DGS model with the boundary of the electronic fence area, positional relationship analysis can be performed simultaneously for different electronic fence areas. This not only improves the efficiency of spatial geometric relationship analysis and reduces the delay in determining boundary crossing analysis results, allowing the alarm push module to promptly perform multi-level linkage alarms and push notifications based on the boundary crossing analysis results, but also reduces the error of spatial geometric relationship analysis data, avoiding false alarms and missed alarms. Furthermore, by analyzing the positional relationship between electronic fences in safe and dangerous areas, a warning zone can be set up in the safe area when the safe and dangerous areas are adjacent. This allows for timely warnings to be given to workers when they enter the warning zone, reducing the likelihood of them entering the dangerous area and ensuring their personal safety.

[0070] In one embodiment of the present invention, spatial geometric relationship analysis is performed based on the location information of the working object in the 3DGS model of the substation and the boundary of the electronic fence area, including: Based on the boundaries of the electronic fence area, perform electronic fence area analysis to determine the geometric type of the electronic fence area.

[0071] The geometric types include convex and concave geometry. During electronic fence area analysis, any two points on the boundary of the electronic fence area are designated as detection points. The analysis checks whether the line connecting the detection points lies inside the electronic fence area. If all lines between detection points are inside the electronic fence area, the geometric type of the electronic fence area is convex geometry; if some lines between detection points are not inside the electronic fence area, the geometric type of the electronic fence area is concave geometry.

[0072] When the geometry of the electronic fence area is a convex geometry, further analysis is performed on the convex geometry to determine whether it is a regular geometry. If the convex geometry is a regular geometry, the cross-sectional plane is determined based on the boundary of the electronic fence area. Based on the cross-sectional plane and the position information of the work object in the substation 3DGS model, the positional relationship between the work object and the electronic fence area is determined, and the current region to which the work object belongs is obtained. When the convex geometry is not a regular geometry, the SDF method is used to perform positional relationship analysis on the convex geometry to determine the electronic fence area where the work object is located, and the current region to which the work object belongs is obtained.

[0073] In determining the positional relationship between the work object and the electronic fence area based on the cross-sectional plane and the position information of the work object in the 3DGS model of the substation, the corresponding coordinate information of the work object in the 3DGS model of the substation is substituted into the expression of the cross-sectional plane to obtain the calculated value. Based on the calculated value and the height of the electronic fence, it is determined whether the position of the work object in the 3DGS model of the substation is within the electronic fence area, thus determining the current region to which the work object belongs. When using the SDF (Signed Distance Function) method to analyze the positional relationship of convex geometry, a discrete SDF field is pre-calculated for the electronic fence area according to the boundary of the electronic fence area, obtaining the discrete SDF field of the electronic fence area. Based on the discrete SDF field of the electronic fence area and the position information of the work object in the 3DGS model of the substation, trilinear interpolation is used to calculate the analytical data. Based on the analytical data and tolerance, it is determined whether the position of the work object in the 3DGS model of the substation is within the electronic fence area, thus determining the current region to which the work object belongs.

[0074] When the geometry of the electronic fence area is concave, ray determination is performed based on the position information of the work object in the substation 3DGS model. The number of intersections between the ray analysis and the boundary of the electronic fence area is used to determine the positional relationship between the work object and the electronic fence area, thus obtaining the current area to which the work object belongs.

[0075] In the process of determining rays based on the location information of the work object in the 3DGS model of the substation, multiple test rays are projected in different directions from the location of the work object in the 3DGS model of the substation as the endpoint, and the intersection points of the test rays with the boundary of the electronic fence area are obtained. This results in a set of test ray intersection points, which determines the test ray intersection point information. Boundary feature point analysis is performed based on the boundary of the electronic fence area to obtain the boundary feature point information of the electronic fence area. Based on the boundary feature point information of the electronic fence area, the test ray intersection point information is filtered and compared with the boundary feature point information of the electronic fence area. Points with the same test ray intersection point information as the boundary feature point information of the electronic fence area are removed from the test ray intersection point set, resulting in a filtered test ray intersection point set. The number of intersection points in the filtered test ray intersection point set is obtained, and the parity of the number of intersection points in the filtered test ray intersection point set is analyzed to determine the positional relationship between the work object and the electronic fence area, and to obtain the current area to which the work object belongs.

[0076] The above-mentioned method not only allows for parallel spatial geometric relationship analysis of the electronic fence area, efficiently determining the current region to which the work object belongs, thus obtaining boundary violation analysis results in a short time and achieving efficient boundary violation judgment, but also employs different methods for analysis according to the geometric type of the electronic fence area. This ensures accurate spatial geometric relationship analysis for electronic fence areas with any three-dimensional spatial shape, reducing analysis errors, ensuring the stability of spatial geometric relationship analysis, and making the current region to which the work object belongs more accurate. Moreover, when the geometric type of the electronic fence area is concave, by combining boundary feature point analysis of the electronic fence area boundary with the intersection point analysis of the test ray, accidental artifacts caused by intersection with concave points are avoided, ensuring the accuracy of the positional relationship between the work object and the electronic fence area and preventing deviations in the current region to which the work object belongs.

[0077] In one embodiment provided by the present invention, such as Figure 4 As shown, the alarm push module includes: an identification and monitoring unit and an alarm push unit. The identification and monitoring unit is used to identify and monitor the boundary analysis results of the boundary analysis module, and send a trigger signal to the alarm push unit based on the identification and monitoring results.

[0078] The identification and monitoring unit monitors the boundary crossing analysis module in real time to determine whether the boundary crossing analysis module has obtained a boundary crossing analysis result. When the boundary crossing analysis module obtains a boundary crossing analysis result, it identifies the boundary crossing analysis result to determine whether the boundary crossing analysis result indicates that the work object has entered the dangerous area. When the identification and monitoring result indicates that the boundary crossing analysis result indicates that the work object has entered the dangerous area, it sends a trigger signal to the alarm push unit.

[0079] The alarm push unit is used to trigger multi-level linkage alarms based on the trigger signal, and at the same time obtain related information for information organization and push.

[0080] Multi-level linkage refers to the coordinated warning system between the substation site and the remote platform, including on-site warnings and system warnings. On-site warnings involve alerting the substation, while system warnings are issued remotely via the substation's 3D Gaussian point cloud electronic fence alarm system. Both on-site and system warnings utilize sound, light, and vibration as warning methods. The process involves acquiring, organizing, and pushing related information, including: identifying work objects entering the hazardous area based on boundary crossing analysis results, locking onto the target work object, and obtaining its information; determining how the target work object entered its current area based on its location in the substation's 3DGS model, its movement trajectory within the substation, and its current area of ​​residence, identifying the boundary crossing point, and obtaining its location; and filtering the virtual images presented by the substation's 3DGS model, tracking the target work object, determining if it exists in the virtual images, extracting the virtual images containing the target work object, and identifying the associated video images to obtain the target work object tracking video. Key information is determined based on the target work object information and the location of the boundary violation, and a first push notification is generated based on this key information. Video processing and compression are performed on the target work object tracking video to obtain a second push notification. The first and second push notifications are matched to obtain a final push notification, which is then sent to the appropriate user.

[0081] The aforementioned identification and monitoring unit enables timely identification of boundary violation analysis results when the boundary violation analysis module determines the results. This allows for timely alarms and notifications when the analysis indicates that a work object has entered a dangerous area. This facilitates efficient emergency response to hazardous situations. Furthermore, the alarm notification unit provides both on-site and system alerts, ensuring that work objects entering dangerous areas are aware of their situation and prevent further harm from misbehavior. It also increases the vigilance of relevant supervisory personnel, enabling them to take timely countermeasures when work objects enter dangerous areas. Simultaneously, information processing and notification allow supervisory personnel to issue alarms and notifications more accurately, significantly reducing emergency response time and improving the inherent safety level of the substation. Moreover, during information processing and notification, targeting specific work objects allows for the determination of notification information based on those objects, enabling simultaneous on-site and system alerts. The first and second notifications enable effective countermeasures to be taken regarding dangerous areas, reducing damage to electrical equipment and minimizing the impact on work objects entering dangerous areas, thus ensuring the safety of the substation.

[0082] In one embodiment of the present invention, when the alarm push unit performs multi-level linkage alarms based on the trigger signal, it issues a warning to the work object entering the dangerous area according to the linkage strategy, including: issuing an alarm prompt through a pop-up window based on the trigger signal and sending an alarm notification to the relevant work object; simultaneously performing location analysis on the work object entering the dangerous area, determining the target location, and issuing an on-site warning based on the target location through an audible and visual alarm device.

[0083] The linkage strategy refers to the simultaneous implementation of on-site and system alerts through mutual cooperation. Based on trigger signals, alarm prompts are sent via pop-up windows, and alarm notifications are sent to relevant work objects. This is a remote alert conducted on a remote platform. When sending alarm notifications to relevant work objects, alarm SMS messages or app push notifications can be sent. When on-site alerts are issued based on the target location using audible and visual alarm devices, audible and visual alarms and verbal broadcasts are conducted in the corresponding danger zone at the target location.

[0084] The aforementioned alarm push unit implements multi-level linkage alarms, enabling simultaneous on-site and system alerts. This ensures that relevant personnel are promptly informed of their entry into dangerous areas, allowing for timely countermeasures. It also helps personnel already in dangerous areas understand their situation and evacuate promptly, preventing further approach and potential personal injury. Furthermore, remote alerts via pop-up windows on the remote platform enhance the attention of relevant personnel, preventing them from ignoring the alarms. Targeted alarm notifications to specific personnel further increase their vigilance. On-site alerts are conducted simultaneously using audible and visual alarms and verbal broadcasts, further raising awareness and preventing personnel from remaining or moving within dangerous areas.

[0085] Those skilled in the art should understand that the terms "first" and "second" in this invention merely refer to different application stages.

[0086] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0087] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A 3D Gaussian point cloud electronic fence alarm system for substations, characterized in that, include: The intelligent calibration module is used to calibrate the three-dimensional spatial shape of the electronic fence for the 3DGS model of the substation, and to determine the 3D Gaussian point cloud electronic fence of the substation. The spatial positioning module is used to combine multiple positioning methods to perform position analysis on the work object and determine the position information of the work object. The boundary crossing analysis module is used to combine the location information of the work object with the 3D Gaussian point cloud electronic fence of the substation to perform spatial boundary crossing analysis and judgment, and obtain the boundary crossing analysis results. The alarm push module is used to generate multi-level linked alarms and push notifications based on the boundary analysis results.

2. The substation 3D Gaussian point cloud electronic fence alarm system according to claim 1, characterized in that, The intelligent calibration module, when calibrating the electronic fence for the 3DGS model of the substation, includes: Based on the 3DGS model of the substation, the three-dimensional spatial safety zone is determined by analyzing the substation's operating procedures and safety specifications. Based on the 3DGS model of the substation, the hazard level and safety distance requirements of the power equipment are analyzed to determine the three-dimensional spatial hazard area; In the 3DGS model of the substation, the boundaries of the three-dimensional safe area and the three-dimensional dangerous area are determined, and electronic fences are established according to the boundaries to obtain the safe area electronic fence and the dangerous area electronic fence. Identifier configuration and attribute labeling are performed for electronic fences in safe areas and electronic fences in dangerous areas respectively.

3. The substation 3D Gaussian point cloud electronic fence alarm system according to claim 2, characterized in that, The intelligent calibration module also performs dynamic calibration of the electronic fence based on the substation's maintenance and expansion construction, including: Determine the maintenance plan or expansion construction plan, and determine the target area based on the maintenance plan or expansion construction plan; Establish temporary electronic fences for the target area; The temporary electronic fence was updated and adjusted by combining it with the 3D Gaussian point cloud electronic fence of the substation.

4. The substation 3D Gaussian point cloud electronic fence alarm system according to claim 1, characterized in that, The multiple positioning methods include indoor positioning and outdoor positioning. The spatial positioning module combines indoor and outdoor positioning methods to achieve indoor-outdoor fusion positioning. When the spatial positioning module analyzes the location of the work object using multiple positioning methods, it uses an intelligent switching algorithm to determine the indoor positioning method as the target positioning method when the work object enters the room and uses UWB positioning technology to determine the location of the work object. When the work object is outdoors, it determines the outdoor positioning method as the target positioning method and uses an RTK positioning terminal that supports the BeiDou satellite navigation system to obtain the location of the work object.

5. The substation 3D Gaussian point cloud electronic fence alarm system according to claim 4, characterized in that, The spatial positioning module also performs spatial monitoring of the substation via an image acquisition device, including: The substation is monitored by acquiring images through an image acquisition device. Based on the substation monitoring images, identification and analysis are performed to determine whether there are living organisms or objects in the substation, and the identification and analysis results are obtained. Based on the identification and analysis results, when a living being or an object appears in the substation, the location of the living being or object is determined, and the target monitoring and identification location is locked.

6. The substation 3D Gaussian point cloud electronic fence alarm system according to claim 1, characterized in that, The boundary crossing analysis module includes: an information processing unit and a data analysis unit; The information processing unit is used to receive the location information of the work object, and to perform fusion processing on the location information of the work object and the 3DGS model of the substation to obtain the location information of the work object in the 3DGS model of the substation. The data analysis unit is used to perform spatial geometric relationship analysis based on the location information of the working object in the 3DGS model of the substation and the electronic fence of the 3D Gaussian point cloud of the substation, and to obtain the boundary analysis results.

7. The substation 3D Gaussian point cloud electronic fence alarm system according to claim 6, characterized in that, The data analysis unit performs spatial geometric relationship analysis based on the data processed by the location of the work object and combined with the 3D Gaussian point cloud electronic fence of the substation, including: Determine the boundary of the electronic fence area based on the 3D Gaussian point cloud of the substation; Based on the location information of the work object in the 3DGS model of the substation and the boundary of the electronic fence area, a positional relationship analysis was performed to determine the current area to which the work object belongs. Based on the current region to which the work object belongs, analyze whether the work object has entered the danger zone to obtain the boundary crossing analysis result.

8. The substation 3D Gaussian point cloud electronic fence alarm system according to claim 7, characterized in that, Spatial geometric relationship analysis was performed based on the location information of the working object in the 3DGS model of the substation, combined with the boundary of the electronic fence area, including: Based on the boundaries of the electronic fence area, perform electronic fence area analysis to determine the geometric type of the electronic fence area; When the geometry of the electronic fence area is a convex geometry, further analysis is performed on the convex geometry to determine whether it is a regular geometry. If the convex geometry is a regular geometry, the cross-sectional plane is determined based on the boundary of the electronic fence area. Based on the cross-sectional plane and the position information of the work object in the substation 3DGS model, the positional relationship between the work object and the electronic fence area is determined, and the current region to which the work object belongs is obtained. When the convex geometry is not a regular geometry, the SDF method is used to perform positional relationship analysis on the convex geometry to determine the electronic fence area where the work object is located, and the current region to which the work object belongs is obtained. When the geometry of the electronic fence area is concave, ray determination is performed based on the position information of the work object in the substation 3DGS model. The number of intersections between the ray analysis and the boundary of the electronic fence area is used to determine the positional relationship between the work object and the electronic fence area, thus obtaining the current area to which the work object belongs.

9. The 3D Gaussian point cloud electronic fence alarm system for substations according to claim 1, characterized in that, The alarm push module includes: an identification and monitoring unit and an alarm push unit. The identification and monitoring unit is used to identify and monitor the boundary analysis results of the boundary analysis module, and send a trigger signal to the alarm push unit according to the identification and monitoring results. The alarm push unit is used to trigger multi-level linkage alarms based on the trigger signal, and at the same time obtain related information for information organization and push.

10. The 3D Gaussian point cloud electronic fence alarm system for substations according to claim 9, characterized in that, When the alarm push unit performs multi-level linkage alarms based on the trigger signal, it will issue a warning to the work object entering the dangerous area according to the linkage strategy, including: issuing an alarm prompt through a pop-up window based on the trigger signal, sending an alarm notification to the relevant work object, performing location analysis on the work object entering the dangerous area, determining the target location, and issuing an on-site warning through an audible and visual alarm device based on the target location.

Images