Meteorological environment intelligent sensing and real-time monitoring method and system

Abstract

The invention provides a meteorological environment intelligent sensing and real-time monitoring method and system, and relates to the technical field of gas phase monitoring, and the method comprises the steps: setting a plurality of meteorological environment sensing nodes in a target monitoring region, obtaining the original data of a meteorological environment, and generating a multi-dimensional meteorological state vector according to a preset time window; according to the multi-dimensional meteorological state vectors of the adjacent time windows, meteorological state change vectors are constructed, the meteorological state change vectors of different meteorological environment sensing nodes are compared on the regional scale, and a regional change reference is formed; carrying out consistency judgment on the meteorological state change vector of each meteorological environment sensing node and a regional change reference, identifying a local meteorological anomaly evolution process, and generating a corresponding identifier; and for the identified local abnormal region, dynamically adjusting the acquisition time window of the original data of the meteorological environment, and updating the meteorological state change vector according to the adjusted time window, thereby realizing continuous sensing and real-time monitoring of the change process of the local meteorological environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of gas phase monitoring, in particular to a meteorological environment intelligent sensing and real-time monitoring method and system. BACKGROUND

[0002] The existing meteorological environment monitoring technology usually arranges fixed meteorological monitoring stations in the target area to realize the collection of meteorological elements such as temperature, humidity, air pressure, wind speed and precipitation. Various sensors sample data at preset time intervals and upload the collected meteorological data to the center server through wired or wireless communication mode, and the server completes data storage, statistical analysis and visualization display to provide basic data support for weather forecasting or environmental assessment.

[0003] However, in the application scene of local strong convective weather monitoring in mountainous areas, the above existing technology has obvious deficiencies. Because the spatial distribution density of fixed meteorological stations is limited, when strong rainfall or gusts are formed in local areas such as valleys and slopes in a short time, the meteorological data collected by the adjacent monitoring stations may still be normal or change lag, which leads to that the real-time monitoring result made by the server based on the data cannot reflect the actual meteorological change, and further affects the accurate judgment of the warning time of sudden geological disasters. SUMMARY

[0004] The purpose of the present application is to provide a meteorological environment intelligent sensing and real-time monitoring method and system, which aims to solve the problems mentioned in the background.

[0005] To solve the above technical problems, the technical scheme of the present application is as follows: In a first aspect, a meteorological environment intelligent sensing and real-time monitoring method, the method comprising: Obtaining meteorological environment original data collected by a plurality of meteorological environment sensing nodes in a target monitoring area at a plurality of continuous collection time points, the meteorological environment original data including temperature, humidity, wind speed and precipitation intensity; According to the meteorological environment original data, the corresponding meteorological environment original data of the same meteorological environment sensing node at the plurality of continuous collection time points is combined and processed according to a preset time window, to generate a multi-dimensional meteorological state vector; According to the multi-dimensional meteorological state vector, the multi-dimensional meteorological state vectors in adjacent time windows are differentially processed to generate a meteorological state change vector representing the change of meteorological state with time; According to the meteorological state change vector, the meteorological state change vectors corresponding to different meteorological environment sensing nodes in the same time window are compared and processed to calculate the change difference degree, and a regional change reference is generated; Based on the regional change benchmark, the consistency of the meteorological state change vectors corresponding to each meteorological environment sensing node is determined. When the meteorological state change vector corresponding to a certain meteorological environment sensing node deviates from the regional change benchmark, a local meteorological anomaly evolution identifier corresponding to that meteorological environment sensing node is generated. Based on the local meteorological anomaly evolution indicators, the time window for collecting the original meteorological environment data in the corresponding area is adjusted, and the meteorological state change vector is regenerated based on the adjusted time window. Based on the regenerated meteorological state change vector, the local meteorological anomaly evolution markers are updated, and the real-time meteorological environment monitoring results of the local meteorological anomaly evolution area are output.

[0006] Preferably, based on the meteorological state change vector, the meteorological state change vectors corresponding to different meteorological environment sensing nodes within the same time window are compared and processed to calculate the degree of change difference and generate a regional change benchmark, including: Based on the meteorological state change vector, the change information corresponding to temperature, humidity, wind speed and precipitation intensity contained therein is broken down and processed to generate element change components corresponding to each meteorological element. Based on the element change components, the element change components from different meteorological environment sensing nodes under the same meteorological element type are aligned to generate an aligned set of element change components. Based on the aligned set of element change components, the stability of the element change components corresponding to each meteorological environment sensing node is evaluated, and abnormal element change components whose change characteristics are inconsistent with those of most nodes are identified. The element change components identified as abnormal are removed, and the remaining element change components are assigned corresponding weights according to the historical stability of the nodes and then summarized to generate a summary result of element change that represents the overall change status of the meteorological element within the time window. The regional change benchmark is generated by combining the summary results of element changes corresponding to different meteorological elements. The regional change benchmark is then updated in subsequent time windows based on the newly acquired summary results of element changes, so that the regional change benchmark can evolve dynamically over time.

[0007] Preferably, based on the regional change benchmark, the consistency of the meteorological state change vectors corresponding to each meteorological environment sensing node is determined. When the meteorological state change vector corresponding to a certain meteorological environment sensing node deviates from the regional change benchmark, a local meteorological anomaly evolution indicator corresponding to that meteorological environment sensing node is generated, including: Based on the meteorological state change vectors corresponding to each meteorological environment sensing node, the corresponding meteorological state change trajectories are generated by performing correlation processing in chronological order. According to the regional change benchmark, the change state of the meteorological state change trajectory in each time window is matched one by one to generate a matching result corresponding to each time window; According to the matching result, the change amplitude, change direction and change trend of the meteorological state change trajectory in the continuous multiple time windows are comprehensively judged with the corresponding relationship of the regional change benchmark to generate a trajectory consistency judgment result; When the trajectory consistency judgment result shows that the meteorological state change trajectory continuously deviates from the regional change benchmark in the continuous multiple time windows, and the deviation direction remains the same, it is determined that the meteorological state change process of the meteorological environment perception node is an abnormal evolution process; According to the abnormal evolution process, a local meteorological abnormal evolution identifier associated with the corresponding meteorological environment perception node is generated to represent the systematic separation between the meteorological state evolution path of the node in the region and the overall evolution path of the target monitoring region.

[0008] Preferably, according to the local meteorological abnormal evolution identifier, the meteorological environment original data collection time window of the corresponding region is adjusted, and the meteorological state change vector is regenerated based on the adjusted collection time window, including: According to the local meteorological abnormal evolution identifier, the corresponding meteorological environment perception node is obtained, and the abnormal evolution duration and abnormal evolution intensity corresponding to the meteorological environment perception node are determined according to the local meteorological abnormal evolution identifier; According to the abnormal evolution duration and abnormal evolution intensity, the meteorological environment original data collection time window of the meteorological environment perception node is adjusted hierarchically to generate an adjusted collection time window corresponding to the abnormal evolution degree; According to the adjusted collection time window, the meteorological environment original data collected by the meteorological environment perception node at the continuous multiple collection time points is reacquired; According to the reacquired meteorological environment original data, the corresponding multi-dimensional meteorological state vector and meteorological state change vector are regenerated to reflect the meteorological state evolution after the abnormal evolution degree changes.

[0009] Preferably, the identified abnormal element change component is removed, and the remaining element change components are summarized after being given corresponding weights according to the historical stability degree of the node to generate an element change summary result representing the overall change state of the meteorological element in the time window, including: The element change components corresponding to each meteorological environment perception node in the predetermined historical time range and multiple historical time windows are obtained to form historical element change data; According to the historical element change data, the element change component of each meteorological environment perception node in each historical time window is compared with the regional change reference in the corresponding time window, and a historical consistency record representing the consistency result in each historical time window is generated; According to the historical consistency record, the length of time that each meteorological environment perception node maintains a consistent change state in the historical time range is accumulated, and a historical stability degree evaluation result corresponding to each meteorological environment perception node is generated; According to the historical stability degree evaluation result, different historical stability degrees are divided into multiple stability levels, and each stability level is mapped to a preset weight interval, and a weight value corresponding to each meteorological environment perception node is generated; After eliminating the element change component identified as abnormal, the remaining element change components are weighted and summarized according to the weight value, and an element change summary result representing the overall change state of the meteorological element in the current time window is generated; In the subsequent time window, the element change summary result is used as input data to update the historical consistency record, so that the weight value participates in the continuous updating process of the regional change reference.

[0010] Preferably, according to the matching result, the change amplitude, change direction and change trend of the meteorological state change trajectory in the continuous multiple time windows are comprehensively judged with the corresponding relationship of the regional change reference, and a trajectory consistency judgment result is generated, including: Obtain the matching result corresponding to the meteorological state change trajectory in the continuous multiple time windows as trajectory matching data; According to the trajectory matching data, the change direction of the meteorological state change trajectory in each time window is compared with the change direction of the regional change reference, and a direction consistency judgment result is generated; In the case where the direction consistency judgment result indicates that the change direction remains consistent, the change amplitude of the meteorological state change trajectory in the continuous multiple time windows is accumulated according to the trajectory matching data, and an amplitude accumulation result is generated; When the amplitude accumulation result meets the preset accumulation condition, the change trend of the meteorological state change trajectory in the continuous multiple time windows is compared with the change trend of the regional change reference according to the trajectory matching data, and a trend consistency judgment result is generated; According to the direction consistency judgment result, the amplitude accumulation result and the trend consistency judgment result, a trajectory consistency judgment result is generated to represent whether the meteorological state change trajectory constitutes an abnormal evolution process.

[0011] Preferably, according to the abnormal evolution duration and the abnormal evolution intensity, the meteorological environment original data collection time window of the meteorological environment perception node is adjusted in stages to generate an adjusted collection time window corresponding to the abnormal evolution degree, including: Obtain the meteorological environment perception node corresponding to the local meteorological abnormal evolution identifier, and obtain the abnormal evolution duration information and the abnormal evolution intensity information associated with the local meteorological abnormal evolution identifier; According to the abnormal evolution duration information, it is judged whether the abnormal evolution duration reaches the preset duration trigger condition, and when the duration trigger condition is reached, a duration-dominated window adjustment instruction is generated; According to the abnormal evolution intensity information, it is judged whether the abnormal evolution intensity reaches the preset intensity trigger condition, and when the intensity trigger condition is reached, an intensity-dominated window adjustment instruction is generated; When the duration-dominated window adjustment instruction and the intensity-dominated window adjustment instruction are generated at the same time, the intensity-dominated window adjustment instruction is selected as the effective instruction; According to the effective window adjustment instruction, the corresponding adjusted collection time window is determined from the preset multiple collection time window levels, and the different collection time window levels are switched when the abnormal evolution duration changes.

[0012] In a second aspect, a meteorological environment intelligent perception and real-time monitoring system, the system comprises: A data acquisition module is used to acquire meteorological environment original data collected by a plurality of meteorological environment perception nodes in a target monitoring area at a plurality of continuous collection time points, and the meteorological environment original data includes temperature, humidity, wind speed and precipitation intensity; A state vector generation module is used to combine and process the meteorological environment original data corresponding to the same meteorological environment perception node at a plurality of continuous collection time points according to a preset time window based on the meteorological environment original data, to generate a multi-dimensional meteorological state vector; A change vector generation module is used to difference process the multi-dimensional meteorological state vectors in adjacent time windows based on the multi-dimensional meteorological state vectors, to generate a meteorological state change vector representing the change of the meteorological state over time; A regional change reference generation module is used to compare and process the meteorological state change vectors corresponding to different meteorological environment perception nodes in the same time window based on the meteorological state change vector, to calculate a change difference degree, and to generate a regional change reference; An abnormal evolution determination module is used to determine the consistency of the meteorological state change vectors corresponding to each meteorological environment perception node based on the regional change reference, and when the meteorological state change vector corresponding to a certain meteorological environment perception node deviates from the regional change reference, a local meteorological abnormal evolution identifier corresponding to the meteorological environment perception node is generated; The collection time window adjusting module is configured to adjust the meteorological environment original data collection time window of the corresponding region according to the local meteorological anomaly evolution identification, and regenerate the meteorological state change vector based on the adjusted collection time window. The monitoring result output module is configured to update the local meteorological anomaly evolution identification according to the regenerated meteorological state change vector, and output the real-time meteorological environment monitoring result of the local meteorological anomaly evolution region.

[0013] The above scheme of the present application at least has the following beneficial effects: By setting a plurality of meteorological environment sensing nodes in the target monitoring region, obtaining meteorological environment original data at a plurality of continuous collection time points, and combining and processing the data of the same node according to a preset time window, the discrete single-point sampling results can be converted into meteorological state descriptions with time continuity, so that the meteorological environment monitoring is no longer limited to the instantaneous value at a single time point, but can reflect the overall change state of meteorological elements within a certain time range, thereby providing a more stable and complete data basis for subsequent analysis.

[0014] On this basis, by differentiating the meteorological state in adjacent time windows to generate a meteorological state change vector, and comparing the change of different meteorological environment sensing nodes within the same time window to form a regional change reference, the overall change structure of the meteorological environment can be described on a regional scale, avoiding the judgment deviation caused by relying on individual monitoring point data, so that the meteorological change in the region has a unified reference standard.

[0015] Further, the meteorological state change vector corresponding to each meteorological environment sensing node is consistent with the regional change reference, and when the change process of a node deviates from the overall change structure of the region, a corresponding local meteorological anomaly evolution identification is generated, so that the basis for anomaly identification is improved from single numerical anomaly to deviation of the evolution process with time, which is beneficial to identifying the abnormal development trend of the local meteorological environment when the meteorological value has not reached a significant abnormal level.

[0016] After identifying the local meteorological anomaly evolution process, the meteorological environment original data collection time window of the related region is adjusted according to the corresponding anomaly evolution identification, and the meteorological state change vector is regenerated based on the adjusted time window, so that the time resolution can be improved for the abnormal region, while the original collection rhythm is maintained for the non-abnormal region, thereby realizing the targeted intensified monitoring of the key region without increasing the overall monitoring burden.

[0017] By comprehensively using the above technical means, the abnormal evolution process of the local meteorological environment can be reflected earlier and more accurately in the application scene of complex spatial distribution and obvious local meteorological change, such as mountainous areas, to provide more timely and reliable monitoring basis for subsequent disaster warning and risk judgment, and to make up for the deficiency of the existing fixed meteorological monitoring technology in monitoring local sudden meteorological change. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 is a flow chart of a meteorological environment intelligent perception and real-time monitoring method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0019] Exemplary embodiments of the present disclosure will be described in greater detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be accurately conveyed to those skilled in the art.

[0020] As Figure 1 shown, an embodiment of the present application proposes a meteorological environment intelligent perception and real-time monitoring method, which comprises: acquiring meteorological environment original data collected by a plurality of meteorological environment perception nodes in a target monitoring area at a plurality of continuous collection time points, the meteorological environment original data including temperature, humidity, wind speed and precipitation intensity; combining and processing the meteorological environment original data corresponding to the same meteorological environment perception node at the plurality of continuous collection time points according to a preset time window based on the meteorological environment original data, to generate a multi-dimensional meteorological state vector; differentially processing the multi-dimensional meteorological state vectors in adjacent time windows based on the multi-dimensional meteorological state vector, to generate a meteorological state change vector representing the change of the meteorological state over time; comparing and processing the meteorological state change vectors corresponding to different meteorological environment perception nodes in the same time window based on the meteorological state change vector, calculating the change difference, and generating a regional change reference; conducting consistency determination on the meteorological state change vectors corresponding to each meteorological environment perception node based on the regional change reference, and generating a local meteorological abnormal evolution identifier corresponding to the meteorological environment perception node when the meteorological state change vector corresponding to the meteorological environment perception node deviates from the regional change reference; adjusting the meteorological environment original data collection time window of the corresponding region based on the local meteorological abnormal evolution identifier, and regenerating the meteorological state change vector based on the adjusted collection time window; According to the regenerated meteorological state change vector, a local meteorological anomaly evolution mark is updated, and a real-time meteorological environment monitoring result of a local meteorological anomaly evolution area is output.

[0021] In the embodiment of the application, by setting a plurality of meteorological environment sensing nodes in the target monitoring area, and acquiring meteorological environment original data such as temperature, humidity, wind speed and precipitation intensity at a plurality of continuous acquisition time points, and then combining and processing the meteorological environment original data corresponding to the same meteorological environment sensing node according to a preset time window, the discrete data can be converted into a multi-dimensional meteorological state vector with time correlation, so that the meteorological environment state is changed from single-point numerical description to overall state description within the time window, which is beneficial to reflect the continuous change characteristics of the meteorological environment in the time dimension.

[0022] By differentiating the multi-dimensional meteorological state vectors in adjacent time windows to generate meteorological state change vectors, and comparing the meteorological state change vectors corresponding to different meteorological environment sensing nodes within the same time window to form a regional change reference, the overall change structure of the meteorological state can be described in the regional scale, avoiding the local deviation caused by relying on single node data only, so that the regional meteorological environment change has a unified reference basis.

[0023] After obtaining the regional change reference, the meteorological state change vectors corresponding to each meteorological environment sensing node are consistent with the regional change reference, and when the meteorological state change vector of a certain meteorological environment sensing node deviates from the regional change reference, the corresponding local meteorological anomaly evolution mark is generated, so that the abnormality judgment is transferred from single numerical fluctuation to the deviation of the meteorological state evolution process with time, which is helpful to identify the case that the local meteorological environment change path in the region is inconsistent with the overall change path.

[0024] After generating the local meteorological anomaly evolution mark, the meteorological environment original data acquisition time window of the corresponding area is adjusted according to the mark, and the meteorological state change vector is regenerated based on the adjusted acquisition time window, so that the meteorological environment change process of the abnormal area can be described in a finer granularity, while maintaining the original acquisition rhythm of the non-abnormal area, realizing the differential monitoring of the meteorological environment change in different areas, and continuously correcting the local meteorological anomaly evolution mark based on the updated meteorological state change vector, so as to ensure that the monitoring result is consistent with the actual meteorological environment change process.

[0025] For example, in a mountain weather monitoring scenario, multiple weather environment perception nodes are distributed in different positions such as valleys, slopes and ridges. When the precipitation intensity and humidity change trend of a valley region is inconsistent with that of the surrounding region in a short time, the local weather abnormal evolution process of the region can be identified by comparing the multi-dimensional weather state vector and the regional change benchmark, and the collection time window of the region is adjusted accordingly to obtain more intensive weather environment data, so as to continuously track the development of the local weather change and facilitate real-time monitoring and understanding of the weather environment state of the region.

[0026] In a preferred embodiment of the present application, the weather environment original data collected by the multiple weather environment perception nodes in the target monitoring region at the continuous collection time points is obtained, including: A plurality of weather environment perception nodes are arranged according to a predetermined spatial distribution in the target monitoring region, and each weather environment perception node is configured with a sensing unit for collecting temperature, humidity, wind speed and precipitation intensity; According to a unified collection time sequence control strategy, the data collection operation of each weather environment perception node is driven to be executed synchronously at the continuous collection time points to form a weather environment original data sequence with time continuity; The temperature, humidity, wind speed and precipitation intensity data collected by each weather environment perception node at each collection time point are associatedly stored with the corresponding node identifier and collection time information to form a weather environment original data set required for subsequent processing.

[0027] In a preferred embodiment of the present application, according to the weather environment original data, the corresponding weather environment original data of the same weather environment perception node at the continuous collection time points is combined and processed according to a preset time window to generate a multi-dimensional weather state vector, including: The length of the time window for describing the weather state is preset so that each time window covers multiple continuous collection time points; For the same weather environment perception node, the original data of temperature, humidity, wind speed and precipitation intensity corresponding to the node in each time window are selected and arranged in the order of collection time; The arranged data are combined and processed, and temperature, humidity, wind speed and precipitation intensity in the same time window are taken as a whole state description, so as to generate a multi-dimensional weather state vector for representing the weather state of the weather environment perception node in the time window.

[0028] In a preferred embodiment of the present application, according to the multi-dimensional weather state vector, the multi-dimensional weather state vectors in adjacent time windows are differentially processed to generate a weather state change vector representing the change of the weather state with time, including: According to the time sequence, the multi-dimensional meteorological state vectors corresponding to two adjacent time windows are selected for the same meteorological environment perception node; The state information corresponding to the temperature, humidity, wind speed and precipitation intensity in the two time windows is compared one by one respectively to determine the change of each meteorological element between the adjacent time windows; The change of each meteorological element in the adjacent time windows is summarized to form a meteorological state change vector which can reflect the change range, change direction and change trend, and is used to describe the change characteristics of the meteorological state of the meteorological environment perception node over time.

[0029] In a preferred embodiment of the present application, according to the regenerated meteorological state change vector, the local meteorological anomaly evolution identification is updated, and the real-time meteorological environment monitoring result of the local meteorological anomaly evolution area is output, including: After completing the collection time window adjustment, the corresponding meteorological state change vector is regenerated based on the new time window, and the meteorological state change vector is taken as the state change input at the current time; The currently generated meteorological state change vector is compared with the meteorological state change vector used to generate the local meteorological anomaly evolution identification previously to determine whether the evolution direction and evolution intensity of the meteorological state change have changed; According to the comparison result, the existing local meteorological anomaly evolution identification is updated, so that the identification can continuously reflect the latest state of the local meteorological environment change process, and the updated result is taken as the real-time meteorological environment monitoring output of the local meteorological anomaly evolution area.

[0030] In a preferred embodiment of the present application, according to the meteorological state change vector, the meteorological state change vectors corresponding to different meteorological environment perception nodes in the same time window are compared and processed, the change difference degree is calculated, and the regional change reference is generated, including: According to the meteorological state change vector, the change information corresponding to the temperature, humidity, wind speed and precipitation intensity contained therein is split and processed to generate element change components corresponding to each meteorological element respectively; According to the element change components, the element change components from different meteorological environment perception nodes of the same meteorological element type are aligned to generate an aligned element change component set; According to the aligned element change component set, the element change components corresponding to each meteorological environment perception node are stably evaluated to identify abnormal element change components whose change characteristics are inconsistent with the majority of nodes; The element change components identified as abnormal are excluded, and the remaining element change components are summarized after being given corresponding weights according to the historical stability degree of the nodes to generate an element change summary result representing the overall change state of the meteorological element in the time window. The element change summary results corresponding to different meteorological elements are combined to generate a regional change reference, and the regional change reference is updated according to newly acquired element change summary results in a subsequent time window, so that the regional change reference dynamically evolves over time.

[0031] In the embodiment of the present application, by splitting the meteorological state change vector into element change components corresponding to different meteorological elements, and aligning and evaluating the change components from different meteorological environment perception nodes within the same meteorological element dimension, abnormal change sources inconsistent with the overall change structure can be filtered out at the regional scale. At the same time, the historical stability degree of the node is introduced to weight the element change components, so that the meteorological environment perception nodes with long-term change behavior stability occupy a higher proportion of influence in the regional change reference, so that the formed regional change reference can continuously reflect the dominant change characteristics of the meteorological state in the target monitoring region, and dynamically update over time, reducing the interference of individual abnormal nodes on the regional judgment result.

[0032] In a preferred embodiment of the present application, according to the set of aligned element change components, the element change components corresponding to each meteorological environment perception node are evaluated for stability, and abnormal element change components whose change characteristics are inconsistent with those of most nodes are identified, including: After completing the alignment of the element change components corresponding to each meteorological environment perception node under the same meteorological element type, each element change component is taken as an evaluation object; The change direction and change amplitude embodied by the element change components of most meteorological environment perception nodes under the same meteorological element type are taken as reference change characteristics, and the element change components corresponding to each meteorological environment perception node are compared one by one; When the element change component corresponding to a meteorological environment perception node is inconsistent with the reference change characteristics in change direction or change amplitude, the element change component is marked as an abnormal candidate component; The change performance of the element change component marked as an abnormal candidate component within the current time window is confirmed, and finally the abnormal element change component whose change characteristics are inconsistent with those of most nodes is identified.

[0033] In a preferred embodiment of the present application, the element change summary results corresponding to different meteorological elements are combined to generate a regional change reference, and the regional change reference is updated according to newly acquired element change summary results in a subsequent time window, so that the regional change reference dynamically evolves over time, including: In the current time window, the element change summary results corresponding to temperature, humidity, wind speed and precipitation intensity are respectively acquired, and each element change summary result is taken as input data at the regional level; According to the preset element combination rule, the element change summary results corresponding to each meteorological element are uniformly integrated to form a regional change benchmark capable of representing the overall meteorological state change structure of the target monitoring area; After entering the subsequent time window, the element change summary results corresponding to each meteorological element generated in the new time window are acquired; The newly acquired element change summary results are associated with the previously formed regional change benchmark, and the regional change benchmark is updated accordingly, so that the regional change benchmark can continuously reflect the change of the meteorological state of the target monitoring area over time.

[0034] In a preferred embodiment of the present application, according to the regional change benchmark, the meteorological state change vector corresponding to each meteorological environment perception node is determined for consistency, and when the meteorological state change vector corresponding to a certain meteorological environment perception node deviates from the regional change benchmark, a local meteorological abnormal evolution identifier corresponding to the meteorological environment perception node is generated, comprising: According to the meteorological state change vector corresponding to each meteorological environment perception node, the associated processing is performed in time sequence to generate the corresponding meteorological state change trajectory; According to the regional change benchmark, the change state of the meteorological state change trajectory in each time window is matched one by one to generate the matching result corresponding to each time window; According to the matching result, the corresponding relationship between the change amplitude, change direction and change trend of the meteorological state change trajectory in the continuous multiple time windows and the regional change benchmark is comprehensively determined to generate the trajectory consistency determination result; When the trajectory consistency determination result indicates that the meteorological state change trajectory continuously deviates from the regional change benchmark in the continuous multiple time windows, and the deviation direction remains the same, it is determined that the meteorological state change process of the meteorological environment perception node is an abnormal evolution process; According to the abnormal evolution process, a local meteorological abnormal evolution identifier associated with the corresponding meteorological environment perception node is generated to represent the systematic separation between the meteorological state evolution path of the node and the overall evolution path of the target monitoring area.

[0035] In the embodiment of the present application, the meteorological state change vector is associated in time dimension to form the meteorological state change trajectory, and the change state of the trajectory in the continuous multiple time windows is matched in combination with the regional change benchmark, so that the abnormal recognition is converted from the numerical deviation at a single time point to the overall judgment of the change path. On this basis, the corresponding relationship between the change direction, change amplitude and change trend is comprehensively determined, so that the abnormal determination has continuity and direction consistency constraints, which helps to distinguish short-term fluctuations and continuous evolution processes, thereby improving the accuracy of identifying local meteorological environment abnormal evolution processes.

[0036] In a preferred embodiment of the present application, the meteorological state change vectors corresponding to each meteorological environment sensing node are sequentially associated to generate corresponding meteorological state change trajectories, including: For each meteorological environment sensing node, the meteorological state change vectors generated in the continuous multiple time windows are sequentially obtained; The meteorological state change vectors of the same meteorological environment sensing node in different time windows are arranged and associated in time sequence to form a data sequence reflecting the continuous process of the meteorological state change of the node; The data sequence is taken as the meteorological state change trajectory to describe the evolution path of the meteorological state change of the meteorological environment sensing node in multiple time windows.

[0037] In a preferred embodiment of the present application, the change state of the meteorological state change trajectory in each time window is matched one by one according to the regional change reference to generate the matching result corresponding to each time window, including: After obtaining the meteorological state change trajectory, the change state of the meteorological state change trajectory in each time window is taken as the matching object; The change state of the meteorological state change trajectory in the time window is compared with the regional change reference by taking the regional change reference in the corresponding time window as the matching reference; According to the comparison result, it is judged whether the change direction, change amplitude and change trend of the meteorological state change trajectory in the time window are consistent with the regional change reference; The judgment result in each time window is recorded respectively to generate the matching result corresponding to each time window, which is used for subsequent consistency determination of the meteorological state change trajectory.

[0038] In a preferred embodiment of the present application, the meteorological environment original data collection time window of the corresponding region is adjusted according to the local meteorological anomaly evolution identifier, and the meteorological state change vector is regenerated based on the adjusted collection time window, including: According to the local meteorological anomaly evolution identifier, the corresponding meteorological environment sensing node is obtained, and the abnormal evolution duration and abnormal evolution intensity corresponding to the meteorological environment sensing node are determined according to the local meteorological anomaly evolution identifier; According to the abnormal evolution duration and abnormal evolution intensity, the meteorological environment original data collection time window of the meteorological environment sensing node is adjusted to generate the adjusted collection time window corresponding to the abnormal evolution degree; According to the adjusted collection time window, the meteorological environment original data collected by the meteorological environment sensing node at continuous multiple collection time points is reacquired; According to the reacquired meteorological environment original data, the corresponding multi-dimensional meteorological state vector and meteorological state change vector are regenerated, which are used to reflect the meteorological state evolution after the abnormal evolution degree is changed.

[0039] In the embodiment of the present application, by acquiring the abnormal evolution duration and abnormal evolution intensity according to the local meteorological abnormal evolution identification, and adjusting the collection time window accordingly, the collection frequency and the meteorological environment change degree form a clear corresponding relationship. When the abnormal evolution is dominated by duration, the collection time window is adjusted step by step to enhance the time resolution; when the abnormal evolution is dominated by intensity change, the collection density is quickly improved through cross-level adjustment, so that the targeted strengthening monitoring of the local abnormal area is realized without changing the overall monitoring architecture, and when the abnormal evolution state changes, the monitoring process is switched between different collection time window levels to keep the synchronization between the monitoring process and the meteorological environment evolution state.

[0040] In a preferred embodiment of the present application, according to the local meteorological abnormal evolution identification, the corresponding meteorological environment perception node is acquired, and the abnormal evolution duration and abnormal evolution intensity corresponding to the meteorological environment perception node are determined according to the local meteorological abnormal evolution identification, including: After the local meteorological abnormal evolution identification is generated, the meteorological environment perception node identification information associated in the identification is read to determine the specific meteorological environment perception node that has abnormal evolution; For the determined meteorological environment perception node, the meteorological state change vector of the node in the continuous multiple time windows is traced back to determine the time range in which the meteorological state change vector of the node is continuously determined as abnormal evolution; The number of time windows in which the abnormal evolution is continuously determined in the time range is taken as the description result of the abnormal evolution duration; At the same time, according to the deviation degree of the meteorological state change vector of the meteorological environment perception node during the abnormal evolution relative to the regional change reference, the change amplitude of the abnormal evolution process is inductively described to form the description result of the abnormal evolution intensity, which is used to represent the degree level of the abnormal evolution of the meteorological state of the node.

[0041] In a preferred embodiment of the present application, the identified abnormal element change component is removed, and the remaining element change components are summarized after being given corresponding weights according to the historical stability degree of the node to generate an element change summary result representing the overall change state of the meteorological element in the time window, including: The element change components corresponding to each meteorological environment perception node in a predetermined historical time range and multiple historical time windows are acquired to form historical element change data; According to the historical element change data, the element change component of each meteorological environment sensing node in each historical time window is compared with the regional change reference in the corresponding time window, and historical consistency records representing consistency results in each historical time window are generated; According to the historical consistency records, the length of time during which each meteorological environment sensing node maintains consistent change in the historical time range is accumulated, and historical stability degree evaluation results corresponding to each meteorological environment sensing node are generated; According to the historical stability degree evaluation results, different historical stability degrees are divided into multiple stability levels, and each stability level is mapped to a preset weight interval, and weight values corresponding to each meteorological environment sensing node are generated; After eliminating the element change component identified as abnormal, the remaining element change components are weighted and summarized according to the weight values, and an element change summary result representing the overall change state of the meteorological element in the current time window is generated; In the subsequent time window, the element change summary result is used as input data to update the historical consistency records, so that the weight values participate in the continuous updating process of the regional change reference.

[0042] In the embodiments of the present application, by continuously recording the element change components of each meteorological environment sensing node in a predetermined historical time range, and forming historical consistency records in combination with the consistency with the regional change reference in the historical time window, the change stability of the node in the long-term running process can be explicitly described. On this basis, the historical stability degree is divided into different levels and mapped to the corresponding weight, so that the element change component no longer participates in the equal weight in the summary process, but is distinguished according to the historical performance of the node, so that the generated element change summary result more reflects the change characteristics of stable nodes. At the same time, the summary result is used to update the historical consistency records in reverse, so that the weight is dynamically adjusted over time, which helps the regional change reference to gradually fit the real regional meteorological change structure in the long-term running.

[0043] In a preferred embodiment of the present application, according to the historical consistency records, the length of time during which each meteorological environment sensing node maintains consistent change in the historical time range is accumulated, and historical stability degree evaluation results corresponding to each meteorological environment sensing node are generated, including: For each meteorological environment sensing node, the number of time windows in which it is determined to maintain consistency with the regional change reference in each historical time window in the historical time range is counted; The number of time windows that continuously or cumulatively maintain consistent change state is taken as the basis for describing the stable running time of the meteorological environment sensing node; According to the proportion of the stable running time in the historical time range, the long-term change stability of each meteorological environment sensing node is classified and described, so as to generate the evaluation result reflecting the historical stability degree of each meteorological environment sensing node.

[0044] In a preferred embodiment of the present application, according to the historical stability degree evaluation result, different historical stability degrees are divided into multiple stability levels, and each stability level is mapped to a preset weight interval to generate the weight value corresponding to each meteorological environment sensing node, including: According to the historical stability degree evaluation result, the meteorological environment sensing nodes are divided into multiple stability levels according to the length of the stable running time, so that there is a clear boundary between different stability levels; For each stability level, a corresponding weight interval is preset, so that the weight interval corresponding to the meteorological environment sensing node with a higher stability level is higher than that of the meteorological environment sensing node with a lower stability level; The stability level to which the meteorological environment sensing node belongs is matched with the corresponding weight interval, and the specific weight value is determined in the weight interval, so as to generate the weight value corresponding to each meteorological environment sensing node.

[0045] In a preferred embodiment of the present application, the preset weight interval setting method includes: In the system initialization stage, according to the number and historical running characteristics of the meteorological environment sensing nodes in the target monitoring area, the number of weight intervals is determined, so that the number of weight intervals is consistent with the number of stability levels; A plurality of continuous weight intervals are set in order from low to high, so that there is no overlap between adjacent weight intervals, and each weight interval corresponds to a unique stability level; When setting the weight interval, the weight interval in a higher stability level is higher than that in a lower stability level in the overall numerical range, so as to ensure that the historical stability degree can play a distinguishing role in the weighted aggregation process; In the system running process, when the historical stability degree evaluation result changes, the corresponding weight interval is matched again according to the updated stability level, so that the weight interval setting is consistent with the long-term change characteristics of the meteorological environment sensing node.

[0046] In a preferred embodiment of the present application, according to the matching result, the change amplitude, change direction and change trend of the meteorological state change trajectory in the continuous multiple time windows are comprehensively judged with the corresponding relationship with the regional change reference, to generate the trajectory consistency judgment result, including: The matching result corresponding to the meteorological state change trajectory in the continuous multiple time windows is obtained as the trajectory matching data; According to the trajectory matching data, the change direction of the meteorological state change trajectory in each time window is compared with the change direction of the regional change reference, and a direction consistency determination result is generated; In the case where the direction consistency determination result indicates that the change direction remains consistent, the change amplitude of the meteorological state change trajectory in the continuous multiple time windows is accumulated according to the trajectory matching data, and an amplitude accumulation result is generated; In the case where the amplitude accumulation result meets the preset accumulation condition, the change trend of the meteorological state change trajectory in the continuous multiple time windows is compared with the change trend of the regional change reference according to the trajectory matching data, and a trend consistency determination result is generated; The direction consistency determination result, the amplitude accumulation result and the trend consistency determination result are jointly processed to generate a trajectory consistency determination result for representing whether the meteorological state change trajectory constitutes an abnormal evolution process.

[0047] In the embodiments of the present application, the change direction is first judged, then the change amplitude is accumulated under the premise of consistent direction, and the comparison of change trend is further introduced after meeting the accumulation condition, so that the judgment of abnormal evolution has a clear judgment sequence and trigger condition. This processing method can identify the evolution process that the single change amplitude does not reach the abnormal level but is continuously accumulated in time, and can distinguish the transition state with consistent direction but not yet formed trend, thereby avoiding misjudgment of short-time disturbance as abnormal evolution, and making the generated trajectory consistency determination result more consistent with the actual law of meteorological state evolution over time.

[0048] In a preferred embodiment of the present application, the setting method of the preset accumulation condition comprises: In the system initialization stage, according to the conventional time scale of meteorological environment change in the target monitoring area, the number of time windows used to describe the accumulation process of meteorological state change is preset, so that the accumulation process covers multiple continuous time windows; For the change amplitude information corresponding to each time window in the meteorological state change trajectory, the change amplitudes are recorded in a window-by-window manner in time sequence, and a continuous change record sequence is formed; When the change amplitude records in the continuous multiple time windows present a continuous growth or continuous weakening state under the premise that the change direction remains consistent, it is determined that the change amplitude meets the accumulation requirement, thereby serving as a basis for judging that the preset accumulation condition is met; In the above manner, the preset accumulation condition can reflect the continuous accumulation characteristics of meteorological state change in the time dimension, rather than the instantaneous change in a single time window.

[0049] In a preferred embodiment of the present application, the meteorological environment raw data collection time window of the meteorological environment perception node is adjusted according to the abnormal evolution duration and the abnormal evolution intensity, and an adjusted collection time window corresponding to the abnormal evolution degree is generated, including: Obtaining the meteorological environment perception node corresponding to the local meteorological abnormal evolution identifier, and obtaining the abnormal evolution duration information and the abnormal evolution intensity information associated with the local meteorological abnormal evolution identifier; According to the abnormal evolution duration information, it is judged whether the abnormal evolution duration reaches a preset duration trigger condition, and when the duration trigger condition is reached, a duration-dominated window adjustment instruction is generated; According to the abnormal evolution intensity information, it is judged whether the abnormal evolution intensity reaches a preset intensity trigger condition, and when the intensity trigger condition is reached, an intensity-dominated window adjustment instruction is generated; When the duration-dominated window adjustment instruction and the intensity-dominated window adjustment instruction are generated at the same time, the intensity-dominated window adjustment instruction is selected as the effective instruction; According to the effective window adjustment instruction, the corresponding adjusted collection time window is determined from the preset multiple collection time window levels, and the different collection time window levels are switched when the abnormal evolution duration changes.

[0050] In the embodiment of the present application, the abnormal evolution duration and the abnormal evolution intensity are obtained respectively, and different adjustment paths are triggered according to different information, so that the change of the collection time window has a clear decision basis. When the abnormal evolution mainly shows the extension of the duration, the collection time window is adjusted step by step according to the level, which is conducive to stable tracking of the change process; when the abnormal evolution mainly shows the intensity mutation, the collection rhythm is quickly changed through the cross-level adjustment method, which helps to capture the key change node. When the abnormal evolution state changes, the different collection time window levels are switched, which can ensure that the collection frequency always matches the abnormal evolution degree, so as to ensure the data continuity while improving the description ability of the abnormal area meteorological change process.

[0051] In a preferred embodiment of the present application, the setting method of the duration trigger condition includes: In the initial stage of system operation, the common duration range of the meteorological abnormal evolution process in the target monitoring area is counted according to the historical meteorological monitoring data; In the common duration range, the time length for distinguishing short fluctuations from continuous evolution processes is selected as the duration trigger condition, so that the time length covers multiple continuous time windows; When the abnormal evolution state corresponding to a certain meteorological environment sensing node persists in continuous time windows and the duration reaches the duration trigger condition during system operation, duration-dominant collection time window adjustment logic is triggered; Through this setting mode, the duration trigger condition is used to identify the case that the meteorological state deviates from the overall evolution path for a long time, avoiding frequent adjustment of the collection rhythm due to short-term disturbances.

[0052] In a preferred embodiment of the present application, the setting method of the intensity trigger condition comprises: During the system initialization phase, the reference change level for describing the abnormal evolution intensity is determined according to the historical change range of different meteorological elements in the target monitoring area; During system operation, the deviation degree of the meteorological state change vector of the meteorological environment sensing node during the abnormal evolution period relative to the regional change reference is compared with the reference change level; When the deviation degree reaches or exceeds the pre-set intensity trigger condition in a single time window, it is determined that the abnormal evolution intensity reaches the trigger requirement, thereby triggering the intensity-dominant collection time window adjustment logic; Through the above setting mode, the intensity trigger condition can quickly respond to the sudden change of the meteorological state.

[0053] In a preferred embodiment of the present application, the setting method of the multiple collection time window levels comprises: During the system configuration phase, multiple collection time window levels of different lengths are pre-set, so that the collection time window levels are distributed in a stepped manner in terms of time span; The collection time window levels are sorted in descending order of time span, so that the collection time window with a larger time span corresponds to a lower collection frequency, and the collection time window with a smaller time span corresponds to a higher collection frequency; Each collection time window level is configured with a corresponding switching condition, so that the system can switch between different collection time window levels according to the duration of abnormal evolution and the intensity of abnormal evolution; During system operation, the corresponding collection time window level is selected according to the effective window adjustment instruction, and the collection time window level is re-evaluated when the abnormal evolution state changes, thereby realizing dynamic adjustment of the collection time window between multiple levels.

[0054] Embodiments of the present application also provide a meteorological environment intelligent sensing and real-time monitoring system, which comprises: A data collection module is configured to acquire meteorological environment raw data collected by multiple meteorological environment sensing nodes in a target monitoring area at continuous collection time points, wherein the meteorological environment raw data includes temperature, humidity, wind speed, and precipitation intensity. a state vector generation module configured to combine meteorological environment raw data corresponding to a same meteorological environment perception node at continuous multiple collection time points according to a preset time window to generate a multi-dimensional meteorological state vector based on the meteorological environment raw data; a change vector generation module configured to perform differential processing on the multi-dimensional meteorological state vectors in adjacent time windows to generate a meteorological state change vector representing changes in the meteorological state over time based on the multi-dimensional meteorological state vectors; a regional change reference generation module configured to perform comparison processing on the meteorological state change vectors corresponding to different meteorological environment perception nodes in a same time window, calculate a change difference degree, and generate a regional change reference based on the meteorological state change vectors; an abnormal evolution determination module configured to perform consistency determination on the meteorological state change vectors corresponding to each meteorological environment perception node based on the regional change reference, and generate a local meteorological abnormal evolution identifier corresponding to a meteorological environment perception node when the meteorological state change vector corresponding to the meteorological environment perception node deviates from the regional change reference; a collection time window adjustment module configured to adjust a meteorological environment raw data collection time window of a corresponding region based on the local meteorological abnormal evolution identifier, and regenerate the meteorological state change vector based on the adjusted collection time window; a monitoring result output module configured to update the local meteorological abnormal evolution identifier based on the regenerated meteorological state change vector, and output a real-time meteorological environment monitoring result of a local meteorological abnormal evolution region.

[0055] It should be noted that the system corresponds to the above method, and all implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0056] Embodiments of the present application also provide a computing device, comprising a processor and a memory storing a computer program, wherein the computer program is executed by the processor to perform the above method. All implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0057] Embodiments of the present application also provide a computer readable storage medium storing instructions, wherein the instructions are executed on a computer to make the computer perform the above method. All implementation manners in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0058] The above is the preferred embodiment of the present application. It should be noted that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which should also be considered within the scope of protection of the present application.

Claims

1. A method for intelligent sensing and real-time monitoring of meteorological environment, characterized in that, The method includes: Acquire raw meteorological and environmental data collected by multiple meteorological and environmental sensing nodes within the target monitoring area at multiple consecutive collection times. The raw meteorological and environmental data includes temperature, humidity, wind speed, and precipitation intensity. Based on the raw meteorological and environmental data, according to the preset time window, the raw meteorological and environmental data corresponding to the same meteorological and environmental sensing node at multiple consecutive collection times are combined and processed to generate a multi-dimensional meteorological state vector. Based on the multidimensional meteorological state vector, the multidimensional meteorological state vectors within adjacent time windows are differentially processed to generate a meteorological state change vector that characterizes the change of meteorological state over time. Based on the meteorological state change vector, the meteorological state change vectors corresponding to different meteorological environment sensing nodes within the same time window are compared and processed to calculate the degree of change difference and generate a regional change benchmark. Based on the regional change benchmark, the consistency of the meteorological state change vectors corresponding to each meteorological environment sensing node is determined. When the meteorological state change vector corresponding to a certain meteorological environment sensing node deviates from the regional change benchmark, a local meteorological anomaly evolution identifier corresponding to that meteorological environment sensing node is generated. Based on the local meteorological anomaly evolution indicators, the time window for collecting the original meteorological environment data in the corresponding area is adjusted, and the meteorological state change vector is regenerated based on the adjusted time window. Based on the regenerated meteorological state change vector, the local meteorological anomaly evolution markers are updated, and the real-time meteorological environment monitoring results of the local meteorological anomaly evolution area are output.

2. The method for intelligent sensing and real-time monitoring of meteorological environment according to claim 1, characterized in that, Based on the meteorological state change vector, the meteorological state change vectors corresponding to different meteorological environment sensing nodes within the same time window are compared and processed to calculate the degree of change difference, generating a regional change benchmark, including: Based on the meteorological state change vector, the change information corresponding to temperature, humidity, wind speed and precipitation intensity contained therein is broken down and processed to generate element change components corresponding to each meteorological element. Based on the element change components, the element change components from different meteorological environment sensing nodes under the same meteorological element type are aligned to generate an aligned set of element change components. Based on the aligned set of element change components, the stability of the element change components corresponding to each meteorological environment sensing node is evaluated, and abnormal element change components whose change characteristics are inconsistent with those of most nodes are identified. The element change components identified as abnormal are removed, and the remaining element change components are assigned corresponding weights according to the historical stability of the nodes and then summarized to generate a summary result of element change that represents the overall change status of the meteorological element within the time window. The regional change benchmark is generated by combining the summary results of element changes corresponding to different meteorological elements. The regional change benchmark is then updated in subsequent time windows based on the newly acquired summary results of element changes, so that the regional change benchmark can evolve dynamically over time.

3. The method for intelligent sensing and real-time monitoring of meteorological environment according to claim 1, characterized in that, Based on the regional change benchmark, the consistency of meteorological state change vectors corresponding to each meteorological environment sensing node is determined. When the meteorological state change vector corresponding to a certain meteorological environment sensing node deviates from the regional change benchmark, a local meteorological anomaly evolution indicator corresponding to that meteorological environment sensing node is generated, including: Based on the meteorological state change vectors corresponding to each meteorological environment sensing node, the corresponding meteorological state change trajectories are generated by performing correlation processing in chronological order. Based on the regional change benchmark, the change status of meteorological state change trajectory in each time window is matched one by one to generate the matching result corresponding to each time window. Based on the matching results, a comprehensive judgment is made on the correlation between the magnitude, direction, and trend of the meteorological state change trajectory within multiple consecutive time windows and the regional change benchmark, generating a trajectory consistency judgment result. When the trajectory consistency determination result shows that the meteorological state change trajectory continuously deviates from the regional change benchmark within multiple consecutive time windows, and the direction of deviation remains consistent, the meteorological state change process of the meteorological environment sensing node is determined to be an abnormal evolution process. Based on the abnormal evolution process, a local meteorological abnormal evolution identifier associated with the corresponding meteorological environment sensing node is generated to characterize the systematic separation between the meteorological state evolution path of the node's location and the overall evolution path of the target monitoring area.

4. The method for intelligent sensing and real-time monitoring of meteorological environment according to claim 1, characterized in that, Based on the local meteorological anomaly evolution indicators, the time window for collecting raw meteorological environmental data in the corresponding area is adjusted, and a new meteorological state change vector is generated based on the adjusted time window, including: Based on the local meteorological anomaly evolution markers, the corresponding meteorological environment sensing nodes are obtained, and the duration and intensity of the anomaly evolution corresponding to the meteorological environment sensing node are determined based on the local meteorological anomaly evolution markers. Based on the duration and intensity of the abnormal evolution, the time window for collecting raw meteorological and environmental data at the meteorological and environmental sensing nodes is adjusted in stages to generate an adjusted collection time window corresponding to the degree of abnormal evolution. Based on the adjusted collection time window, the original meteorological and environmental data collected by the meteorological and environmental sensing node at multiple consecutive collection times are reacquired. Based on the newly acquired meteorological environment data, corresponding multidimensional meteorological state vectors and meteorological state change vectors are regenerated to reflect the evolution of meteorological state after changes in the degree of anomalous evolution.

5. The method for intelligent sensing and real-time monitoring of meteorological environment according to claim 2, characterized in that, Anomalous element change components are removed, and the remaining element change components are weighted according to their historical stability before being aggregated to generate a summary result representing the overall change status of the meteorological element within the time window, including: Acquire the element change components corresponding to each meteorological environment sensing node within a predetermined historical time range and multiple historical time windows to form historical element change data; Based on historical data on element changes, the element change components of each meteorological environment sensing node within each historical time window are compared with the regional change benchmark within the corresponding time window to generate a historical consistency record that represents the consistency results within each historical time window. Based on historical consistency records, the duration of time during which each meteorological environment sensing node maintains a consistent state of change within a historical time range is accumulated to generate historical stability evaluation results for each meteorological environment sensing node. Based on the historical stability evaluation results, different historical stability levels are divided into multiple stability levels, and each stability level is mapped to a preset weight range to generate the weight value corresponding to each meteorological environment sensing node. After removing the element change components identified as abnormal, the remaining element change components are weighted and summarized according to their weight values ​​to generate a summary result of element change that represents the overall change status of the meteorological element within the current time window. In subsequent time windows, the summary results of element changes are used as input data to update historical consistency records, so that the weight values ​​participate in the continuous updating process of regional change benchmarks.

6. The method for intelligent sensing and real-time monitoring of meteorological environment according to claim 3, characterized in that, Based on the matching results, a comprehensive assessment is made of the correlation between the magnitude, direction, and trend of meteorological state change trajectories across multiple consecutive time windows and the regional change benchmark, generating trajectory consistency assessment results, including: Obtain the matching results of meteorological state change trajectories within multiple consecutive time windows, and use them as trajectory matching data; Based on trajectory matching data, the direction of change of meteorological state change trajectory within each time window is compared with the direction of change of regional change benchmark to generate a direction consistency judgment result. If the direction consistency determination results show that the direction of change is consistent, the amplitude of the change of the meteorological state trajectory in multiple consecutive time windows is accumulated based on the trajectory matching data to generate the amplitude accumulation result. When the amplitude accumulation result meets the preset accumulation conditions, the trend of meteorological state change trajectory in multiple consecutive time windows is compared with the trend of regional change benchmark based on trajectory matching data to generate trend consistency judgment result. The trajectory consistency determination results are jointly processed based on the directional consistency determination results, the amplitude accumulation results, and the trend consistency determination results to generate trajectory consistency determination results that characterize whether the trajectory of meteorological state change constitutes an abnormal evolution process.

7. The method for intelligent sensing and real-time monitoring of meteorological environment according to claim 4, characterized in that, Based on the duration and intensity of the anomaly evolution, the time window for collecting raw meteorological and environmental data at meteorological and environmental sensing nodes is adjusted in stages to generate adjusted collection time windows corresponding to the degree of anomaly evolution, including: Obtain the meteorological environment sensing node corresponding to the local meteorological anomaly evolution marker, and obtain the anomaly evolution duration and anomaly evolution intensity information associated with the local meteorological anomaly evolution marker; Based on the abnormal evolution duration information, determine whether the abnormal evolution duration has reached the preset duration trigger condition. When the duration trigger condition is reached, generate a duration-driven window adjustment command. Based on the abnormal evolution intensity information, determine whether the abnormal evolution intensity has reached the preset intensity triggering condition. When the intensity triggering condition is reached, generate an intensity-driven window adjustment command. When generating both duration-driven and intensity-driven window adjustment commands simultaneously, the intensity-driven window adjustment command shall be selected as the effective command. Based on the effective window adjustment instruction, the corresponding adjusted acquisition time window is determined from multiple preset acquisition time window levels, and the process switches between different acquisition time window levels when the abnormal evolution continues to change.

8. A meteorological environment intelligent sensing and real-time monitoring system, characterized in that, The system, used in the method of any one of claims 1 to 7, comprises: The data acquisition module is used to acquire raw meteorological and environmental data collected by multiple meteorological and environmental sensing nodes within the target monitoring area at multiple consecutive acquisition times. The raw meteorological and environmental data includes temperature, humidity, wind speed, and precipitation intensity. The state vector generation module is used to combine and process the raw meteorological and environmental data corresponding to multiple consecutive collection times of the same meteorological and environmental sensing node according to a preset time window, based on the raw meteorological and environmental data, to generate a multi-dimensional meteorological state vector. The change vector generation module is used to perform differential processing on the multidimensional meteorological state vector within adjacent time windows based on the multidimensional meteorological state vector, and generate a meteorological state change vector that represents the change of meteorological state over time. The regional change benchmark generation module is used to compare the meteorological state change vectors corresponding to different meteorological environment sensing nodes within the same time window based on the meteorological state change vector, calculate the degree of change difference, and generate a regional change benchmark. The anomaly evolution determination module is used to determine the consistency of the meteorological state change vectors corresponding to each meteorological environment sensing node based on the regional change benchmark. When the meteorological state change vector corresponding to a certain meteorological environment sensing node deviates from the regional change benchmark, a local meteorological anomaly evolution identifier corresponding to that meteorological environment sensing node is generated. The data acquisition time window adjustment module is used to adjust the data acquisition time window of the corresponding area's meteorological environment raw data according to the local meteorological anomaly evolution indicator, and regenerate the meteorological state change vector based on the adjusted data acquisition time window; The monitoring results output module is used to update the local meteorological anomaly evolution markers based on the regenerated meteorological state change vector, and output the real-time meteorological environment monitoring results of the local meteorological anomaly evolution area.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images