Multi-target environment edge monitoring and early warning system and method based on artificial intelligence

Abstract

The present application relates to the technical field of environmental monitoring, and discloses a multi-target environmental edge monitoring and early warning system and method based on artificial intelligence, which comprises the following steps: obtaining environmental monitoring records based on ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, work-based monitoring equipment and water-based monitoring equipment, and constructing an equipment space correlation framework; the multi-target fusion is constrained by space distribution and remains stable; dynamic time sequence correlation features are extracted and an equipment unified time sequence reference is established under the framework; time sequence offset patterns are identified according to the unified time sequence reference; a time sequence correction algorithm containing an offset adaptive correction mechanism is used to correct the environmental monitoring records, compress the time offset across targets, and finally generate cross-equipment risk evolution features from the corrected environmental monitoring records, and the equipment space correlation framework is dynamically updated based on the features, so that the multi-target data fusion and time sequence alignment form a closed loop, thereby improving the cross-target time sequence alignment accuracy and ensuring the continuity of environmental risk trend identification.

Description

Technical Field

[0001] This invention relates to the field of environmental monitoring technology, and more specifically, to a multi-target environmental edge monitoring and early warning system and method based on artificial intelligence. Background Technology

[0002] In the scenario of bank slope environmental monitoring, various types of monitoring devices are usually deployed in complex terrains where water and land meet. The data they collect differ in format, sampling rhythm and spatial reference. Existing technologies mostly rely on abstract methods such as local data fusion, time segment splicing, static change threshold and regional rule judgment. These methods are usually applicable under the premise of stable monitoring points, high synchronization of the acquisition link and weak environmental disturbance.

[0003] In actual operation, there are two objective unstable factors. First, the data update rhythm of the multi-target monitoring device in the riverbank area fluctuates continuously with hydrological, meteorological and surface deformation fluctuations. Second, the time offset and jitter generated by the data transmission link in the mixed water and land environment. The above factors will weaken the stability of multi-target data fusion along the same processing chain and further amplify the temporal deviation of cross-target data, resulting in the loss of continuity of risk trend identification. Therefore, the technical problem that needs to be solved is how to ensure the stability of multi-target monitoring data fusion, and on this basis improve the cross-target temporal alignment accuracy to ensure the continuity of environmental risk trend identification.

[0004] In view of this, the present invention proposes a multi-target environmental edge monitoring and early warning system and method based on artificial intelligence to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, the present invention provides a multi-target environmental edge monitoring and early warning system and method based on artificial intelligence.

[0006] To achieve the above objectives, the present invention provides the following technical solution: Firstly, it provides an artificial intelligence-based multi-target environmental edge monitoring and early warning method, including: Environmental monitoring records are obtained using ground-based, air-based, space-based, industrial-based, and water-based monitoring equipment in the riverbank area. A spatial association framework for the equipment is constructed based on the spatial distribution relationship of these monitoring devices. Based on the equipment spatial association framework, dynamic temporal association features are extracted among ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment, and a unified temporal reference for the equipment is established based on the dynamic temporal association features. Based on the unified timing reference of the equipment, the timing offset pattern of the environmental monitoring records is identified, a timing correction algorithm including an adaptive offset correction mechanism is constructed, and the timing offset pattern in the environmental monitoring records is corrected using the timing correction algorithm including the adaptive offset correction mechanism. Based on the revised environmental monitoring records, cross-device risk evolution characteristics are generated, and the device spatial association framework is dynamically updated based on these characteristics.

[0007] In some embodiments, constructing a device space association framework includes: Based on the installation location and monitoring coverage of each monitoring device, the boundary of the monitoring coverage area is extracted, and the extracted boundary area is mapped according to the spatial structure of the bank slope area. The spatial coverage overlap area formed between the coverage areas of each monitoring device is identified through the mapped area data. Based on the spatial coverage overlap area, the overlap pattern of the coverage areas of any two monitoring devices is compared item by item to obtain the comparison results. Based on the comparison results, the spatial intersection features are extracted from the monitoring coverage data, and then the spatial intersection relationship between the two monitoring devices is qualitatively analyzed based on the spatial intersection features. Based on the spatial cross relationships obtained from the qualitative analysis, equipment combinations with stable cross relationships are collected as equipment spatial association combinations, and an equipment spatial association framework is constructed based on the equipment spatial association combinations.

[0008] In some embodiments, based on the spatial coverage overlap area, the overlap pattern of the coverage areas of any two monitoring devices is compared item by item to obtain the comparison result, including: Based on the overlapping area of ​​the spatial coverage of the two monitoring devices, the spatial boundaries of the coverage areas of each monitoring device are divided one by one, and the spatial location correspondence processing is performed on the divided spatial boundaries to obtain the coverage correspondence relationship between the boundary locations of the two monitoring devices. Based on the coverage correspondence of the boundary positions, the shape characteristics of the coverage areas of the two monitoring devices in the overlapping areas are structurally compared to obtain coverage comparison records; Based on the coverage comparison records, a comparison result is formed of the spatial coverage overlap pattern between the coverage areas of the two monitoring devices.

[0009] In some embodiments, based on the coverage correspondence of the boundary locations, the shape features of the coverage areas of the two monitoring devices in the overlapping area are structurally compared to obtain a coverage comparison record, including: Based on the coverage correspondence of the overlapping areas and boundary positions of the spatial coverage, the boundary structure of the coverage areas of the two monitoring devices in the overlapping area is extracted as boundary structure segments. The spatial position and orientation characteristics of each boundary structure segment are analyzed segment by segment to obtain boundary comparison segment data. Based on the boundary comparison segment data, the shape and geometric features of the coverage areas of the two monitoring devices in the overlapping area are compared segment by segment to obtain shape feature comparison records; Based on the shape feature comparison records, the shape correspondence between the coverage areas of the two monitoring devices in the overlapping area is compiled into a coverage comparison record.

[0010] In some embodiments, based on the equipment spatial association framework, dynamic temporal association features among ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-grade monitoring equipment, and water-based monitoring equipment are extracted, including: Based on the coverage comparison records and environmental monitoring records, the boundary structure of the coverage areas of the two monitoring devices in the spatial coverage overlap area was analyzed item by item. The response change characteristics of the boundary structure of the coverage area in the time dimension were extracted one by one. Based on the extracted boundary structure response change characteristics, the starting time of the response change of the coverage area when the two monitoring devices responded to environmental disturbances was clearly recorded. Based on the start time of the response changes in the coverage areas of the two monitoring devices, the start time of the response changes of the monitoring devices is compared one by one, the time sequence of the response changes of the two monitoring devices is recorded in detail, and based on the detailed record of the time sequence, a clear observation response order relationship between the two monitoring devices is established. Based on the established clear observation response sequence relationship, the response start times of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment in multiple environmental disturbance events are sorted into response time sequence relationship links, and based on the response time sequence relationship links, the specific observation response propagation paths between equipment are clarified; Based on the clearly defined observation response propagation paths between devices, dynamic temporal correlation features are extracted among ground-based monitoring devices, air-based monitoring devices, space-based monitoring devices, industrial-based monitoring devices, and water-based monitoring devices.

[0011] In some embodiments, a clear order of observation responses between two monitoring devices is established based on detailed records in chronological order, including: Based on the detailed records in chronological order, the observation start time sequences of the two monitoring devices in multiple environmental disturbance events were extracted from the environmental monitoring records, and the observation start time sequences were arranged in the order of the events to form a table of observation time changes for comparison. Based on the observation time change table, the order of observation times of the two monitoring devices in each disturbance event is compared event by event, and the observation sequence fragment records of the two monitoring devices in each environmental disturbance event are established based on the comparison results. Based on the observation sequence fragment records, the observation sequence fragments of the two monitoring devices in all disturbance events are summarized and organized, and the summarized and organized observation sequence fragments are integrated into the observation response order relationship between the two monitoring devices.

[0012] In some embodiments, a timing correction algorithm incorporating an adaptive offset correction mechanism is constructed, including: Based on the order of observation responses, environmental monitoring records are traced back, and the time difference between the observation response times of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment is calculated one by one. For each pair of monitoring equipment, a time series offset assessment is performed to obtain equipment response time series offset data. Based on the device response timing offset data, identify the stable response offset pattern in each device response offset data, and generate an offset adaptive correction mechanism for correcting the timing of device response observation data based on the stable response offset pattern. By embedding the offset adaptive correction mechanism used to correct the timing of the device response observation data into the timing correction algorithm, a timing correction algorithm containing the offset adaptive correction mechanism is obtained.

[0013] In some embodiments, an adaptive offset correction mechanism for correcting the timing of device response observation data is generated based on a stable response offset mode, including: Based on the stable response migration pattern, response migration numerical sequences are extracted one by one from the observation response time series data of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment; and the response migration numerical sequences of each equipment are arranged event by event to form a response migration numerical vector that can be compared one by one. Based on the above response offset numerical vectors, calculate the relative difference value of each device's response offset numerical vector, and construct an offset correction numerical table for device response timing based on the relative difference value. Based on the offset correction value table of device response timing, an adaptive offset correction mechanism is generated to correct the timing of device response observation data.

[0014] In some embodiments, generating cross-device risk evolution characteristics based on modified environmental monitoring records includes: Based on the corrected environmental monitoring records, response change data of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment to various environmental disturbance events were extracted, and spatial response difference analysis was performed on the response change data on an event-by-event basis to obtain the response change difference characteristics between each device. Based on the differences in response changes among devices, the transmission direction of risk changes among monitoring devices is determined on an event-by-event basis, and a clear risk propagation path is constructed between devices; Based on the constructed risk propagation path, the cross-device risk evolution characteristics in multiple environmental disturbance events among ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment are identified.

[0015] Secondly, an AI-based multi-target environmental edge monitoring and early warning system is provided, which is used to implement the aforementioned AI-based multi-target environmental edge monitoring and early warning method, including: Spatial Framework Module: Used to acquire environmental monitoring records based on ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment in the riverbank area, and to construct a spatial association framework for the equipment based on the spatial distribution relationship of the ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment; The temporal correlation module is used to extract dynamic temporal correlation features between ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment based on the equipment spatial correlation framework, and to establish a unified temporal reference for the equipment based on the dynamic temporal correlation features. Timing correction module: Used to identify the timing offset pattern of environmental monitoring records based on the unified timing reference of the equipment, construct a timing correction algorithm including an adaptive offset correction mechanism, and use the timing correction algorithm including an adaptive offset correction mechanism to correct the timing offset pattern in the environmental monitoring records. Evolution Update Module: Used to generate cross-device risk evolution characteristics based on the corrected environmental monitoring records, and dynamically update the device spatial association framework based on the cross-device risk evolution characteristics.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention acquires environmental monitoring records from ground-based, airborne, space-based, industrial, and water-based monitoring equipment, and constructs a spatial association framework based on their spatial distribution. This framework constrains multi-target data at the spatial level, reducing fusion instability caused by distribution differences. Dynamic temporal correlation features are then extracted within this framework to establish a unified temporal benchmark for the equipment, unifying different update rhythms and transmission jitter to the same time caliber. Based on this unified temporal benchmark, temporal offset patterns in environmental monitoring records are identified. A temporal correction algorithm incorporating an adaptive offset correction mechanism is used to correct the offsets, gradually converging the time difference across targets and reducing jitter propagation introduced by the link. After temporal correction, cross-equipment risk evolution features are generated based on the corrected environmental monitoring records, and the spatial association framework is dynamically updated accordingly. This creates a closed-loop iteration between spatial constraints and temporal alignment. Thus, on the one hand, the stability of multi-target monitoring data fusion is maintained within the spatial association framework; on the other hand, the accuracy of cross-target temporal alignment is improved during the temporal correction process, ultimately ensuring the continuity of environmental risk trend identification. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the multi-target environmental edge monitoring and early warning method based on artificial intelligence in this invention. Figure 2 This is a schematic diagram of the structure of the multi-target environmental edge monitoring and early warning system based on artificial intelligence in this invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments and the accompanying drawings. In the following detailed description, many specific details are set forth to provide a thorough understanding of the exemplary embodiments described. However, it will be apparent to those skilled in the art that the described embodiments may be practiced without some or all of these specific details. In other exemplary embodiments, well-known structures have not been described in detail to avoid unnecessarily obscuring the concepts of this disclosure. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention. Furthermore, the various aspects described in the embodiments may be combined arbitrarily without conflict.

[0019] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0020] Example 1 Please see Figure 1 As shown, this embodiment discloses a multi-target environmental edge monitoring and early warning method based on artificial intelligence, including: S10: Environmental monitoring records are obtained based on ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment in the riverbank area, and a spatial association framework for the equipment is constructed based on the spatial distribution relationship of the ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment; In this embodiment, the monitoring devices installed in the bank slope area include ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-grade monitoring equipment, and water-based monitoring equipment. Specifically, the ground-based monitoring equipment may include a GNSS displacement monitor and a surface crack monitoring sensor deployed on the bank slope surface; the air-based monitoring equipment may include an unmanned aerial vehicle (UAV) platform equipped with a high-resolution optical remote sensing sensor; the water-based monitoring equipment may include a water level monitoring buoy installed in the water surface area; and the space-based monitoring equipment may include an Earth observation satellite equipped with a synthetic aperture radar (SAR) sensor or a multispectral remote sensing sensor, used to periodically acquire large-scale surface deformation information, surface humidity distribution information, or topographic relief change information of the bank slope area. The monitoring equipment for the engineering foundation can specifically include tilt sensors, stress monitoring sensors, or structural vibration monitoring units fixedly deployed on the engineering structures around the bank slope. These devices are used to continuously collect information on the stress changes, tilt changes, or vibration response of the bank slope support structure or adjacent engineering facilities under environmental disturbances. The aforementioned devices respectively collect environmental monitoring records of bank slope displacement changes, bank slope crack propagation trends, and dynamic changes in water level. To construct a spatial correlation framework for the equipment, the specific coverage boundaries are first extracted based on the installation location and monitoring coverage area of ​​the aforementioned monitoring equipment. These coverage boundaries are then accurately mapped and compared according to spatial location to determine the clear spatial coverage overlap areas between the coverage areas of different monitoring equipment.

[0021] In this embodiment, the spatial coverage of each monitoring device can be represented by specific coordinates, as shown in Table 1: Table 1 After determining the coverage area coordinates of each site, the coverage area coordinates of each device are spatially mapped and compared to identify the spatial coverage overlap areas between devices, as shown in Table 2: Table 2 The aforementioned overlapping spatial coverage areas clarify the spatial relationships between different monitoring devices, which are used to subsequently construct a spatial association framework for the devices, thereby establishing stable spatial associations between the devices.

[0022] The construction of the device space association framework includes: Based on the installation location and monitoring coverage of each monitoring device, the boundary of the monitoring coverage area is extracted, and the extracted boundary area is mapped according to the spatial structure of the bank slope area. The spatial coverage overlap area formed between the coverage areas of each monitoring device is identified through the mapped area data. Based on the spatial coverage overlap area, the overlap pattern of the coverage areas of any two monitoring devices is compared item by item to obtain the comparison results. Based on the comparison results, the spatial intersection features are extracted from the monitoring coverage data, and then the spatial intersection relationship between the two monitoring devices is qualitatively analyzed based on the spatial intersection features. Based on the spatial cross relationships obtained from the qualitative analysis, equipment combinations with stable cross relationships are collected as equipment spatial association combinations, and an equipment spatial association framework is constructed based on the equipment spatial association combinations.

[0023] In this embodiment, the equipment spatial association framework refers to a data structure that clearly describes the spatial deployment relationship of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment in the bank slope area. This structure enables spatial data fusion between different types of equipment, improving the accuracy and continuity of risk trend identification.

[0024] In practical bank slope environmental monitoring scenarios, ground-based monitoring equipment (such as GNSS displacement monitors) is typically deployed on the bank slope surface for long-term, continuous monitoring of bank slope displacement changes; airborne monitoring equipment (such as UAV platforms equipped with optical remote sensing sensors) acquires information on bank slope surface morphology and its changes through periodic inspections; water-based monitoring equipment (such as water level monitoring buoys) is used to collect water level change data in real time; space-based monitoring equipment (such as Earth observation satellites equipped with synthetic aperture radar or multispectral sensors) is used to acquire overall deformation trends and surface change information of the bank slope area on a larger spatial scale; and engineering-based monitoring equipment (such as tilt sensors deployed on bank slope support structures or adjacent engineering facilities)... Force monitoring sensors are used to monitor changes in force or attitude of engineering structures under environmental disturbances. Since these devices are distributed in different locations on the bank slope and the data formats vary greatly, it is necessary to clearly construct a spatial association framework for the devices in order to accurately determine the data fusion relationship between the devices. In order to construct the spatial association framework for the devices, this embodiment first clearly extracts the spatial coverage area of ​​each device, and performs boundary determination and spatial mapping processing based on the actual installation location and observation range of the devices to clearly identify the overlapping areas of spatial coverage between the devices. This process is a conventional spatial boundary extraction and GIS mapping process, and will not be described in detail here.

[0025] It is necessary to explain in detail the process of comparing the overlapping patterns of the coverage areas of any two monitoring devices item by item, and extracting spatial intersection features based on the comparison results. This detailed processing is necessary because the spatial intersection and variation patterns of monitoring data in the riverbank area are complex. It is essential to clearly define the spatial boundary relationships and spatial intersection features of each coverage area, and then conduct qualitative analysis based on these specific spatial intersection features to accurately identify the stable spatial intersection relationship between the two monitoring devices. Optionally, for example, when the monitoring coverage areas of the ground-based monitoring device and the airborne monitoring device have a clear overlap in a specific area, the spatial intersection features specifically include differences in boundary positions, the shape variation pattern of the overlapping coverage areas, or the stability of the spatial overlap. These features can be represented as specific numerical pairs or spatial positional correspondences to support the accurate construction of the spatial association framework of the devices.

[0026] It should be noted that the clear and stable spatial cross relationship enables this embodiment to form a more accurate combination of device spatial associations, ensuring the clarity and stability of spatial relationships when fusion of cross-device data, thereby effectively solving the problems of insufficient stability of multi-device data fusion and poor continuity of risk trend identification in the prior art.

[0027] For example, when there is a clear overlap in the spatial coverage areas of a ground-based monitoring device and an airborne monitoring device, the spatial boundary coordinate set of the device coverage areas can be specifically represented as: The coordinate set of the boundary of the coverage area of ​​the ground monitoring equipment: {(x1,y1),(x2,y2),(x3,y3)}; The coordinate set of the boundary of the coverage area of ​​the airborne monitoring equipment: {(x4,y4),(x5,y5),(x6,y6)}; By explicitly mapping and comparing each segment, the specific and clear coordinates of the overlapping area are identified as {(x10,y10),(x11,y11),(x12,y12)}, and clearly recorded as the spatial overlapping area, thus forming a data foundation that can be used to further clarify the spatial intersection relationship.

[0028] It should be further explained that this embodiment clearly constructs a spatial association framework for equipment based on the above method, which can clearly and effectively solve the problem of insufficient accuracy of data fusion of different types of monitoring equipment in the riverbank environmental monitoring scenario, and ensure that the identification of environmental risk trends is more stable and reliable.

[0029] Based on the spatial coverage overlap area, the overlap pattern of the coverage areas of any two monitoring devices is compared item by item to obtain the comparison results, including: Based on the overlapping area of ​​the spatial coverage of the two monitoring devices, the spatial boundaries of the coverage areas of each monitoring device are divided one by one, and the spatial location correspondence processing is performed on the divided spatial boundaries to obtain the coverage correspondence relationship between the boundary locations of the two monitoring devices. Based on the coverage correspondence of the boundary positions, the shape characteristics of the coverage areas of the two monitoring devices in the overlapping areas are structurally compared to obtain coverage comparison records; Based on the coverage comparison records, a comparison result is formed of the spatial coverage overlap pattern between the coverage areas of the two monitoring devices.

[0030] Based on the coverage correspondence at the boundary locations, the shape characteristics of the coverage areas of the two monitoring devices within the overlapping area are structurally compared to obtain a coverage comparison record, including: Based on the coverage correspondence between the overlapping areas and boundary locations of the spatial coverage, the boundary structures of the coverage areas of the two monitoring devices within the overlapping areas are extracted as boundary structure segments. The spatial location and orientation characteristics of each boundary structure segment are analyzed segment by segment to obtain boundary comparison segment data.

[0031] Based on the boundary comparison segment data, the shape and geometric features of the overlapping areas of the two monitoring devices are compared segment by segment to obtain shape feature comparison records.

[0032] Based on the shape feature comparison records, the shape correspondence between the coverage areas of the two monitoring devices in the overlapping area is compiled into a coverage comparison record.

[0033] In this embodiment, the spatial coverage overlap area refers to the area where the monitoring coverage areas of different monitoring devices clearly overlap in spatial location. The coverage comparison record specifically refers to the result of clearly recording the comparison of shape features between the coverage areas of two monitoring devices, which is used to further analyze the observation response relationship between the devices. In order to obtain the above-mentioned coverage comparison record, it is first necessary to perform more detailed analysis and processing on the clear spatial coverage overlap area. Specifically, the coverage area of ​​each monitoring device is clearly divided into multiple clear spatial boundary structure segments. These boundary structure segments represent the different specific locations and directional features of the coverage area boundaries in space.

[0034] For example, in the spatial overlap area between the two clearly defined monitoring coverage areas of ground-based monitoring equipment and airborne monitoring equipment, this embodiment clearly delineates the spatial boundary structure segments as follows: The boundary structure segment of the coverage area of ​​the ground-based monitoring equipment is: {(X1,Y1), (X2,Y2), (X3,Y3)}; The boundary structure segment of the coverage area of ​​the airborne monitoring equipment is: {(X4,Y4), (X5,Y5), (X6,Y6)}; The spatial location correspondence of the above-mentioned spatial boundary structure segments is clearly processed segment by segment to obtain a clear boundary location coverage correspondence between the coverage areas of the two monitoring devices. For example, the following can be clearly formed: {(X1,Y1)→(X4,Y4), (X2,Y2)→(X5,Y5), (X3,Y3)→(X6,Y6)}. Based on the clear coverage correspondence, a structured comparison process is carried out between the coverage areas.

[0035] In this embodiment, the structured comparison process specifically involves comparing the geometric shape change characteristics of the coverage areas of the two monitoring devices in the overlapping area segment by segment, clearly analyzing the spatial consistency and differences of the shape features, thereby clearly obtaining the shape feature comparison record between the coverage areas. This comparison record can clearly reflect the clear spatial shape relationship between the two monitoring devices.

[0036] For example, the geometric morphological characteristics of the area covered by the ground-based monitoring equipment are defined as {curved boundary, convex shape}, and the geometric morphological characteristics of the area covered by the air-based monitoring equipment are defined as {curved boundary, concave shape}. Through clear segment-by-segment structured comparison processing, the shape feature comparison record is clearly formed as: {(convex, concave)}.

[0037] Based on the aforementioned clear shape feature comparison records, the comparison results of the clear spatial overlap patterns between the coverage areas of the two monitoring devices are clearly compiled, clearly indicating the spatial relationship between the coverage areas of the two monitoring devices, such as partial overlap, complete overlap, or complementary connection. The clear comparison results can be clearly used for subsequent analysis of the temporal relationship and risk evolution characteristics between the devices, thereby improving the reliability and accuracy of environmental monitoring data fusion and trend identification.

[0038] S20: Based on the equipment spatial association framework, extract the dynamic temporal association features between ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment, and establish a unified temporal reference for the equipment based on the dynamic temporal association features; Based on the equipment spatial association framework, dynamic temporal association features are extracted among ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-grade monitoring equipment, and water-based monitoring equipment, including: Based on the coverage comparison records and environmental monitoring records, the boundary structure of the coverage areas of the two monitoring devices in the spatial coverage overlap area was analyzed item by item. The response change characteristics of the boundary structure of the coverage area in the time dimension were extracted one by one. Based on the extracted boundary structure response change characteristics, the starting time of the response change of the coverage area when the two monitoring devices responded to environmental disturbances was clearly recorded. Based on the start time of the response changes in the coverage areas of the two monitoring devices, the start time of the response changes of the monitoring devices is compared one by one, the time sequence of the response changes of the two monitoring devices is recorded in detail, and based on the detailed record of the time sequence, a clear observation response order relationship between the two monitoring devices is established. Based on the established clear observation response sequence relationship, the response start times of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment in multiple environmental disturbance events are sorted into response time sequence relationship links, and based on the response time sequence relationship links, the specific observation response propagation paths between equipment are clarified; Based on the clearly defined observation response propagation paths between devices, dynamic temporal correlation features are extracted among ground-based monitoring devices, air-based monitoring devices, space-based monitoring devices, industrial-based monitoring devices, and water-based monitoring devices.

[0039] In this embodiment, dynamic temporal correlation features are used to clearly represent the temporal correlation between different types of monitoring equipment in response to environmental disturbances within the bank slope area. By establishing dynamic temporal correlation features, the fusion and coordination of observation data from different equipment in time can be effectively achieved, thereby further improving the reliability and continuity of environmental risk trend identification. The unified time series benchmark for equipment is a unified response time series benchmark that clearly represents different types of monitoring equipment. By using a unified time series benchmark, the clear correspondence between observation data from different monitoring equipment in the time dimension can be ensured, thereby significantly improving the accuracy of subsequent risk analysis.

[0040] Specifically, based on the coverage comparison records and environmental monitoring records, the boundary structure of the coverage areas of the two monitoring devices in the spatial coverage overlap area is analyzed item by item. The purpose of this process is to clearly analyze the response changes of the boundary structure at different time points, thereby clearly identifying the dynamic temporal correlation characteristics between the devices.

[0041] In this embodiment, it is assumed that the clear coverage comparison records show that the ground-based monitoring equipment and the airborne monitoring equipment have clear boundary structure overlap features within the clear spatial coverage overlap area {(X10,Y10),(X11,Y11),(X12,Y12)}. Through item-by-item structure analysis, the response changes of the two devices at different times are clearly extracted from the environmental monitoring records to obtain the clear response change start time data, as shown in Table 3: Table 3 The aforementioned data clearly demonstrates the exact temporal differences in the observed response changes between the two devices, which can be used for further detailed time-series correlation analysis.

[0042] Based on this, according to the start time of the response changes of the two monitoring devices, the time sequence of the device response changes in each environmental disturbance event is compared and analyzed in detail, and the response sequence in each event is clearly recorded, thereby clearly establishing the observation response sequence relationship between the two monitoring devices.

[0043] Furthermore, based on the clearly established observation response sequence relationship, the response start times of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment in multiple environmental disturbance events are clearly organized into specific response timing relationship links, and specific observation response propagation paths between each device are clearly formed.

[0044] For example, in a specific environmental disturbance event, the propagation path of the device response is defined as follows: Space-based monitoring equipment (2024-01-01 09:55) → Ground-based monitoring equipment (2024-01-01 10:01) → Air-based monitoring equipment (2024-01-01 10:03) → Industrial-based monitoring equipment (2024-01-01 10:06) → Water-based monitoring equipment (2024-01-01 10:08); The clearly defined response propagation path clearly indicates the explicit order of observation responses between devices, and lays the foundation for the next step of explicitly extracting dynamic temporal correlation features.

[0045] Finally, based on the clearly defined observation response propagation paths among the aforementioned devices, dynamic temporal correlation features among ground-based monitoring devices, air-based monitoring devices, space-based monitoring devices, industrial-based monitoring devices, and water-based monitoring devices are extracted. These features clearly reflect the explicit response temporal relationships among the various monitoring devices, enabling the subsequent construction of a unified temporal benchmark for the devices. This further enhances the accuracy and reliability of environmental monitoring and early warning analysis. Compared to existing technologies, this approach exhibits higher spatial monitoring data fusion stability and cross-device response temporal alignment accuracy, significantly improving the continuity and accuracy of risk trend identification.

[0046] Based on detailed chronological records, a clear order of observation responses between the two monitoring devices is established, including: Based on the detailed records in chronological order, the observation start time sequences of the two monitoring devices in multiple environmental disturbance events were extracted from the environmental monitoring records, and the observation start time sequences were arranged in the order of the events to form a table of observation time changes for comparison. Based on the observation time change table, the order of observation times of the two monitoring devices in each disturbance event is compared event by event, and the observation sequence fragment records of the two monitoring devices in each environmental disturbance event are established based on the comparison results. Based on the observation sequence fragment records, the observation sequence fragments of the two monitoring devices in all disturbance events are summarized and organized, and the summarized and organized observation sequence fragments are integrated into the observation response order relationship between the two monitoring devices.

[0047] In this embodiment, the observation response order relationship is a data structure that describes the specific order of observation responses of multiple monitoring devices in different environmental disturbance events. The observation response order relationship can accurately reflect the clear observation response propagation path between devices, thus providing a clear basis for device timing calibration. First, based on the clear recorded time sequence, the observation start time sequence of ground-based monitoring devices and airborne monitoring devices in multiple environmental disturbance events is clearly extracted from the environmental monitoring records, and then clearly arranged according to the actual order of events to form a clear observation time change table. This processing logic aims to clearly reveal the clear response start situation of monitoring devices at different times, avoid the confusion of device response timing, and thus improve the reliability of response order analysis.

[0048] For example, as shown in Table 4: Table 4 The aforementioned observation time change table clearly shows the exact start time of each monitoring device's response to each environmental disturbance event, providing clear data support for the next step of comparative analysis. After obtaining the aforementioned observation time change table, this embodiment clearly compares the order of observation times of the two monitoring devices in each specific environmental disturbance event, and clearly establishes observation sequence fragment records. Each fragment record specifically represents the response order of the device in a certain environmental disturbance event. This clear observation sequence fragment record can describe in detail the specific temporal characteristics of the device response.

[0049] After obtaining a clear sequence of observation records, this embodiment summarizes and organizes all the records to form a clear, stable, and distinct observation response order relationship. This clear observation response order relationship not only clearly reveals the propagation path of the observation responses of multiple monitoring devices in the bank slope area, but also provides a clear and distinct data foundation for subsequent time series correlation analysis and environmental monitoring data fusion. Compared with the general time series recording and sorting methods in the prior art, this clear, detailed, and stable response order relationship significantly improves the accuracy of cross-device data fusion and the stability and continuity of risk trend identification.

[0050] S30: Identify the time-series offset pattern of environmental monitoring records based on the unified time-series reference of the equipment, construct a time-series correction algorithm including an offset adaptive correction mechanism, and use the time-series correction algorithm including an offset adaptive correction mechanism to correct the time-series offset pattern in the environmental monitoring records. Construct a timing correction algorithm that includes an adaptive offset correction mechanism, including: Based on the order of observation responses, environmental monitoring records are traced back, and the time difference between the observation response times of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment is calculated one by one. For each pair of monitoring equipment, a time series offset assessment is performed to obtain equipment response time series offset data. Based on the device response timing offset data, identify the stable response offset pattern in each device response offset data, and generate an offset adaptive correction mechanism for correcting the timing of device response observation data based on the stable response offset pattern. By embedding the offset adaptive correction mechanism used to correct the timing of the device response observation data into the timing correction algorithm, a timing correction algorithm containing the offset adaptive correction mechanism is obtained.

[0051] In this embodiment, the offset adaptive correction mechanism is a technical mechanism specifically used to correct the timing deviation of the observation response of different monitoring devices. By embedding this mechanism into the timing correction algorithm, the timing accuracy of the device response data is significantly improved, thereby ensuring the reliability and accuracy of environmental monitoring data fusion and risk trend identification. First, based on the aforementioned clearly established observation response order relationship, the environmental monitoring records are traced back to calculate the specific time difference between the observation response times of the ground-based monitoring devices, air-based monitoring devices, space-based monitoring devices, industrial-based monitoring devices, and water-based monitoring devices. This processing step aims to clearly identify the differences in the actual observation timing of the monitoring devices, providing clear data support for subsequent offset correction.

[0052] For example, given a specific environmental disturbance event, the response start time for space-based monitoring equipment is 09:55 on January 1, 2024; the response start time for ground-based monitoring equipment is 10:01 on January 1, 2024; the response start time for airborne monitoring equipment is 10:03 on January 1, 2024; the response start time for industrial monitoring equipment is 10:06 on January 1, 2024; and the response start time for water-based monitoring equipment is 10:05 on January 1, 2024. The time difference between the different monitoring devices can then be calculated as follows: Time difference between space-based and ground-based monitoring equipment: 6 minutes; Time difference between ground-based and air-based monitoring equipment: 2 minutes; Time difference between air-based and water-based monitoring equipment: 2 minutes; Time difference between air-based and industrial-based monitoring equipment: 3 minutes; Time difference between ground-based and water-based monitoring equipment: 4 minutes.

[0053] Based on this, further time series offset assessment of the above time difference can clearly identify the stable response offset patterns of different monitoring devices in multiple environmental disturbance events. For example, it can be clearly identified that the response of space-based monitoring devices is usually about 6 minutes earlier than that of ground-based monitoring devices, the response of ground-based monitoring devices is usually about 2 minutes earlier than that of air-based monitoring devices, and the response of industrial-based monitoring devices is usually about 3 minutes later than that of air-based monitoring devices.

[0054] Furthermore, based on the clearly identified stable response offset pattern, this embodiment generates an adaptive offset correction mechanism for correcting the timing of device response observation data. This adaptive offset correction mechanism can be explicitly represented as a set of offset correction rules related to the device type, for example: When the response start time of the ground-based monitoring equipment is T, the corrected response start time of the air-based monitoring equipment is T+2 minutes. The calibrated response start time of the water-based monitoring equipment is T+4 minutes. The response start time of the calibrated industrial foundation monitoring equipment is T+5 minutes or T+6 minutes. The corrected response start time of the space-based monitoring equipment is T−6 minutes or the response start time back to the corresponding remote sensing observation time window.

[0055] By using the aforementioned offset correction rules, the observation response data of different types of monitoring equipment can be aligned under a unified time reference, thereby effectively eliminating the time deviations introduced by differences in sampling rhythm, transmission link jitter, and observation scale.

[0056] Finally, this embodiment explicitly embeds the aforementioned specific offset adaptive correction mechanism into the time series correction algorithm, forming a time series correction algorithm that includes an offset adaptive correction mechanism. Through this explicit correction algorithm, the time series alignment accuracy of observation data from ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment can be significantly improved. This clearly enhances the stability and accuracy of environmental monitoring data fusion and risk trend identification. Compared with the general static time series correction methods in the prior art, the explicit offset adaptive correction mechanism in this embodiment can more accurately and dynamically adjust the equipment response observation time series, significantly improving the accuracy and reliability of monitoring data processing.

[0057] An adaptive offset correction mechanism is generated based on the stable response offset pattern to correct the timing of device response observation data, including: Based on the stable response migration pattern, response migration numerical sequences are extracted one by one from the observation response time series data of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment; and the response migration numerical sequences of each equipment are arranged event by event to form a response migration numerical vector that can be compared one by one. Based on the above response offset numerical vectors, calculate the relative difference value of each device's response offset numerical vector, and construct an offset correction numerical table for device response timing based on the relative difference value. Based on the offset correction value table of device response timing, an adaptive offset correction mechanism is generated to correct the timing of device response observation data.

[0058] In this embodiment, the offset adaptive correction mechanism is specifically a technical method for clearly correcting the timing of response observation data of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment based on the stable response offset mode. This mechanism can significantly improve the alignment accuracy of observation data from different monitoring equipment in the time dimension, thereby further enhancing the reliability of risk trend identification.

[0059] Specifically, to clarify the adaptive correction mechanism for the generated offset, we first explicitly start from the stable response offset mode, and then extract a clear response offset numerical sequence from the observation response time series data of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment. This response offset numerical sequence clearly represents the difference in the observation response time of each equipment in a specific event. Furthermore, we explicitly arrange the above response offset numerical sequences in the order of events to form a response offset numerical vector for clear comparison one by one.

[0060] For example, Table 5 shows an example of a specific response offset numerical vector: Table 5 Based on the clearly obtained response offset numerical vectors, this embodiment specifically calculates the relative difference values ​​between the response offset numerical vectors of each device. The relative difference values ​​clearly indicate the degree of consistency of the response offset of each device in multiple events. Specifically, it can be obtained by calculating the element-by-element difference or mean deviation of the numerical vectors. Based on the clear relative difference values, a clear offset correction numerical table for the device response timing is specifically constructed.

[0061] For example, Table 6 shows a clear example of offset correction numerical representation: Table 6 Finally, based on the aforementioned offset correction value table for the clearly defined device response timing, this embodiment specifically generates an offset adaptive correction mechanism. This correction mechanism is specifically represented by a set of explicit correction rules, for example, explicitly stated as follows: when the response time of the ground-based monitoring device is T, the corrected response time of the air-based monitoring device is T+2 minutes, while the corrected response time of the water-based monitoring device is T+4 minutes. Through this explicit offset adaptive correction mechanism, the timing of the device response observation data can be effectively corrected, and the timing accuracy and stability of environmental monitoring data fusion and risk trend identification can be significantly improved. Compared with the traditional static correction method, it can more clearly and accurately reflect the true timing relationship between device responses.

[0062] S40: Generate cross-device risk evolution characteristics based on the corrected environmental monitoring records, and dynamically update the device spatial association framework based on the cross-device risk evolution characteristics.

[0063] Based on the revised environmental monitoring records, cross-device risk evolution characteristics are generated, including: Based on the corrected environmental monitoring records, response change data of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment to various environmental disturbance events were extracted, and spatial response difference analysis was performed on the response change data on an event-by-event basis to obtain the response change difference characteristics between each device. Based on the differences in response changes among devices, the transmission direction of risk changes among monitoring devices is determined on an event-by-event basis, and a clear risk propagation path is constructed between devices; Based on the constructed risk propagation path, the cross-device risk evolution characteristics in multiple environmental disturbance events among ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment are identified.

[0064] It is understandable that "cross-device risk evolution characteristics" specifically refers to a clear data feature that describes the spatial transmission relationship of risk changes of multiple monitoring devices in multiple environmental disturbance events. By having clear cross-device risk evolution characteristics, the reliability of environmental risk trend identification can be further improved, and specific data support can be provided for the subsequent dynamic updating of the device spatial correlation framework.

[0065] Furthermore, to clearly generate the aforementioned cross-device risk evolution characteristics, it is first necessary to clearly extract the response change data of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment to each specific environmental disturbance event based on the corrected environmental monitoring records. The specific manifestations of the response change data are clarified, including specific characteristics such as changes in the intensity of the equipment response, changes in coverage, or changes in the response time. On this basis, a clear spatial response difference analysis is conducted on an event-by-event basis to obtain clear and detailed response change difference characteristics between each device.

[0066] For example, this embodiment clarifies the response change data of three types of monitoring devices in a specific environmental disturbance event, as shown in Table 7: Table 7 Based on the aforementioned clearly defined response change data, an event-by-event spatial response difference analysis is conducted on the response changes between each device. For example, the specific differences in the intensity of device response changes are clearly analyzed, and it is clearly identified that the response of ground-based monitoring equipment is significantly stronger than that of air-based monitoring equipment, and that the response of air-based monitoring equipment is significantly stronger than that of water-based monitoring equipment, thereby clearly obtaining the distinct difference characteristics of response changes between devices.

[0067] Based on this, and according to the differences in response changes among the devices, each environmental disturbance event is analyzed individually to determine the specific spatial transmission direction of risk changes among different monitoring devices. For example, in a single environmental disturbance event, the propagation path of risk changes can be identified, which is transmitted sequentially from ground-based monitoring devices to air-based monitoring devices and further extends to water-based monitoring devices. In events involving both space-based and engineering-based monitoring devices, the transmission of risk changes from the macroscopic deformation trends perceived by space-based monitoring devices to the local responses of ground-based and air-based monitoring devices can also be identified, further affecting the engineering structural response of engineering-based monitoring devices and the hydrological response of water-based monitoring devices, thus forming a complete cross-device risk propagation path.

[0068] Based on comparative analysis of multiple environmental disturbance events, the aforementioned risk propagation paths are summarized and organized, further identifying the stable cross-device risk evolution characteristics among different types of monitoring equipment. For example, in multiple events, space-based monitoring equipment exhibits the characteristic of prior perception of macroscopic risk changes, while ground-based monitoring equipment generally responds to changes more strongly than space-based monitoring equipment, which in turn generally responds more strongly than water-based monitoring equipment. The response of engineering-based monitoring equipment, on the other hand, shows a structurally delayed response after the aforementioned changes occur. This clearly demonstrates the spatial evolution law of risk propagation within the bank slope area from macroscopic perception and confirmation of local deformation to the stepwise transmission of engineering structures and hydrological responses.

[0069] This embodiment, based on the aforementioned stable cross-equipment risk evolution characteristics, dynamically updates the equipment spatial association framework. This allows the spatial relationships between equipment to be continuously adjusted and corrected as the actual risk evolution process unfolds, thereby improving the adaptability and accuracy of the equipment spatial association framework to changes in the real slope environment. Compared to existing methods that use static spatial relationships, this embodiment can more accurately reflect the risk evolution trend involving multiple types of monitoring equipment, significantly improving the reliability and effectiveness of environmental risk trend identification and early warning analysis.

[0070] Example 2 Please see Figure 2 As shown, based on the same inventive concept, this embodiment discloses a multi-target environmental edge monitoring and early warning system based on artificial intelligence. For details not covered in this embodiment, please refer to the relevant parts of Embodiment 1. The system includes: Spatial Framework Module: Used to acquire environmental monitoring records based on ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment in the riverbank area, and to construct a spatial association framework for the equipment based on the spatial distribution relationship of the ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment; The temporal correlation module is used to extract dynamic temporal correlation features between ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment based on the equipment spatial correlation framework, and to establish a unified temporal reference for the equipment based on the dynamic temporal correlation features. Timing correction module: Used to identify the timing offset pattern of environmental monitoring records based on the unified timing reference of the equipment, construct a timing correction algorithm including an adaptive offset correction mechanism, and use the timing correction algorithm including an adaptive offset correction mechanism to correct the timing offset pattern in the environmental monitoring records. Evolution Update Module: Used to generate cross-device risk evolution characteristics based on the corrected environmental monitoring records, and dynamically update the device spatial association framework based on the cross-device risk evolution characteristics.

[0071] The detailed description above, in conjunction with the accompanying drawings, describes examples but does not represent all examples that can be implemented or fall within the scope of the claims. The terms “example” and “exemplary” are used in this specification to mean “serving as an example, instance or illustration” and do not mean “superior to or better than other examples”.

[0072] Throughout this specification, the phrase "an embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with that embodiment is included in at least one embodiment of the invention. Therefore, the use of these phrases may refer to more than one embodiment. Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0073] It should also be noted that these embodiments may be described as processes depicted as flowcharts, structural diagrams, or block diagrams. Although a flowchart may describe the operations as sequential processes, many of these operations can be performed in parallel or concurrently, and the order of these operations may be rearranged.

Claims

1. A multi-target environmental edge monitoring and early warning method based on artificial intelligence, characterized in that, include: Environmental monitoring records are obtained using ground-based, air-based, space-based, industrial-based, and water-based monitoring equipment in the riverbank area. A spatial association framework for the equipment is constructed based on the spatial distribution relationship of these monitoring devices. Based on the equipment spatial association framework, dynamic temporal association features are extracted among ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment, and a unified temporal reference for the equipment is established based on the dynamic temporal association features. Based on the unified timing reference of the equipment, the timing offset pattern of the environmental monitoring records is identified, a timing correction algorithm including an adaptive offset correction mechanism is constructed, and the timing offset pattern in the environmental monitoring records is corrected using the timing correction algorithm including the adaptive offset correction mechanism. Based on the revised environmental monitoring records, cross-device risk evolution characteristics are generated, and the device spatial association framework is dynamically updated based on these characteristics.

2. The multi-target environmental edge monitoring and early warning method based on artificial intelligence according to claim 1, characterized in that, Constructing a device space association framework includes: Based on the installation location and monitoring coverage of each monitoring device, the boundary of the monitoring coverage area is extracted, and the extracted boundary area is mapped according to the spatial structure of the bank slope area. The spatial coverage overlap area formed between the coverage areas of each monitoring device is identified through the mapped area data. Based on the spatial coverage overlap area, the overlap pattern of the coverage areas of any two monitoring devices is compared item by item to obtain the comparison results. Based on the comparison results, the spatial intersection features are extracted from the monitoring coverage data, and then the spatial intersection relationship between the two monitoring devices is qualitatively analyzed based on the spatial intersection features. Based on the spatial cross relationships obtained from the qualitative analysis, equipment combinations with stable cross relationships are collected as equipment spatial association combinations, and an equipment spatial association framework is constructed based on the equipment spatial association combinations.

3. The multi-target environmental edge monitoring and early warning method based on artificial intelligence according to claim 2, characterized in that, Based on the spatial coverage overlap area, the overlap pattern of the coverage areas of any two monitoring devices is compared item by item to obtain the comparison results, including: Based on the overlapping area of ​​the spatial coverage of the two monitoring devices, the spatial boundaries of the coverage areas of each monitoring device are divided one by one, and the spatial location correspondence processing is performed on the divided spatial boundaries to obtain the coverage correspondence relationship between the boundary locations of the two monitoring devices. Based on the coverage correspondence of the boundary positions, the shape characteristics of the coverage areas of the two monitoring devices in the overlapping areas are structurally compared to obtain coverage comparison records; Based on the coverage comparison records, a comparison result is formed of the spatial coverage overlap pattern between the coverage areas of the two monitoring devices.

4. The multi-target environmental edge monitoring and early warning method based on artificial intelligence according to claim 3, characterized in that, Based on the coverage correspondence at the boundary locations, the shape characteristics of the coverage areas of the two monitoring devices within the overlapping area are structurally compared to obtain a coverage comparison record, including: Based on the coverage correspondence of the overlapping areas and boundary positions of the spatial coverage, the boundary structure of the coverage areas of the two monitoring devices in the overlapping area is extracted as boundary structure segments. The spatial position and orientation characteristics of each boundary structure segment are analyzed segment by segment to obtain boundary comparison segment data. Based on the boundary comparison segment data, the shape and geometric features of the coverage areas of the two monitoring devices in the overlapping area are compared segment by segment to obtain shape feature comparison records; Based on the shape feature comparison records, the shape correspondence between the coverage areas of the two monitoring devices in the overlapping area is compiled into a coverage comparison record.

5. The multi-target environmental edge monitoring and early warning method based on artificial intelligence according to claim 3, characterized in that, Based on the equipment spatial association framework, dynamic temporal association features are extracted among ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-grade monitoring equipment, and water-based monitoring equipment, including: Based on the coverage comparison records and environmental monitoring records, the boundary structure of the coverage areas of the two monitoring devices in the spatial coverage overlap area was analyzed item by item. The response change characteristics of the boundary structure of the coverage area in the time dimension were extracted one by one. Based on the extracted boundary structure response change characteristics, the starting time of the response change of the coverage area when the two monitoring devices responded to environmental disturbances was clearly recorded. Based on the start time of the response changes in the coverage areas of the two monitoring devices, the start time of the response changes of the monitoring devices is compared one by one, the time sequence of the response changes of the two monitoring devices is recorded in detail, and based on the detailed record of the time sequence, a clear observation response order relationship between the two monitoring devices is established. Based on the established clear observation response sequence relationship, the response start times of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment in multiple environmental disturbance events are sorted into response time sequence relationship links, and based on the response time sequence relationship links, the specific observation response propagation paths between equipment are clarified; Based on the clearly defined observation response propagation paths between devices, dynamic temporal correlation features are extracted among ground-based monitoring devices, air-based monitoring devices, space-based monitoring devices, industrial-based monitoring devices, and water-based monitoring devices.

6. The multi-target environmental edge monitoring and early warning method based on artificial intelligence according to claim 5, characterized in that, Based on detailed chronological records, a clear order of observation responses between the two monitoring devices is established, including: Based on the detailed records in chronological order, the observation start time sequences of the two monitoring devices in multiple environmental disturbance events were extracted from the environmental monitoring records, and the observation start time sequences were arranged in the order of the events to form a table of observation time changes for comparison. Based on the observation time change table, the order of observation times of the two monitoring devices in each disturbance event is compared event by event, and the observation sequence fragment records of the two monitoring devices in each environmental disturbance event are established based on the comparison results. Based on the observation sequence fragment records, the observation sequence fragments of the two monitoring devices in all disturbance events are summarized and organized, and the summarized and organized observation sequence fragments are integrated into the observation response order relationship between the two monitoring devices.

7. The multi-target environmental edge monitoring and early warning method based on artificial intelligence according to claim 6, characterized in that, Construct a timing correction algorithm that includes an adaptive offset correction mechanism, including: Based on the order of observation responses, environmental monitoring records are traced back, and the time difference between the observation response times of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment is calculated one by one. For each pair of monitoring equipment, a time series offset assessment is performed to obtain equipment response time series offset data. Based on the device response timing offset data, identify the stable response offset pattern in each device response offset data, and generate an offset adaptive correction mechanism for correcting the timing of device response observation data based on the stable response offset pattern. By embedding the offset adaptive correction mechanism used to correct the timing of the device response observation data into the timing correction algorithm, a timing correction algorithm containing the offset adaptive correction mechanism is obtained.

8. The multi-target environmental edge monitoring and early warning method based on artificial intelligence according to claim 7, characterized in that, An adaptive offset correction mechanism is generated based on the stable response offset pattern to correct the timing of device response observation data, including: Based on the stable response migration pattern, response migration numerical sequences are extracted one by one from the observation response time series data of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment; and the response migration numerical sequences of each equipment are arranged event by event to form a response migration numerical vector that can be compared one by one. Based on the above response offset numerical vectors, calculate the relative difference value of each device's response offset numerical vector, and construct an offset correction numerical table for device response timing based on the relative difference value. Based on the offset correction value table of device response timing, an adaptive offset correction mechanism is generated to correct the timing of device response observation data.

9. The multi-target environmental edge monitoring and early warning method based on artificial intelligence according to claim 1, characterized in that, Based on the revised environmental monitoring records, cross-device risk evolution characteristics are generated, including: Based on the corrected environmental monitoring records, response change data of ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment to various environmental disturbance events were extracted, and spatial response difference analysis was performed on the response change data on an event-by-event basis to obtain the response change difference characteristics between each device. Based on the differences in response changes among devices, the transmission direction of risk changes among monitoring devices is determined on an event-by-event basis, and a clear risk propagation path is constructed between devices; Based on the constructed risk propagation path, the cross-device risk evolution characteristics in multiple environmental disturbance events among ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment are identified.

10. An AI-based multi-target environmental edge monitoring and early warning system, used to implement the AI-based multi-target environmental edge monitoring and early warning method as described in any one of claims 1-9, characterized in that, include: Spatial Framework Module: Used to acquire environmental monitoring records based on ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment in the riverbank area, and to construct a spatial association framework for the equipment based on the spatial distribution relationship of the ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment and water-based monitoring equipment; The temporal correlation module is used to extract dynamic temporal correlation features between ground-based monitoring equipment, air-based monitoring equipment, space-based monitoring equipment, industrial-based monitoring equipment, and water-based monitoring equipment based on the equipment spatial correlation framework, and to establish a unified temporal reference for the equipment based on the dynamic temporal correlation features. Timing correction module: Used to identify the timing offset pattern of environmental monitoring records based on the unified timing reference of the equipment, construct a timing correction algorithm including an adaptive offset correction mechanism, and use the timing correction algorithm including an adaptive offset correction mechanism to correct the timing offset pattern in the environmental monitoring records. Evolution Update Module: Used to generate cross-device risk evolution characteristics based on the corrected environmental monitoring records, and dynamically update the device spatial association framework based on the cross-device risk evolution characteristics.

Images