Flood detection and early warning method and system based on space-air-ground power equipment

Abstract

The invention discloses a flood detection early warning method and system based on space-air-ground power equipment, and relates to the technical field of big data processing, and the method comprises the steps: carrying out the classification and recognition of data according to the keyword of flood monitoring data; preprocessing the database; extracting a statistical feature vector of the flood monitoring data and extracting a time sequence feature vector; performing multi-source data feature fusion on the statistical feature vector and the time sequence feature vector; and constructing a flood dangerous case model. According to the flood detection and early warning system based on the space-air-ground power equipment, the flood monitoring data has polymorphism, polygon, data discretization and certain correlation and randomness, the obtained initial data are classified and cleaned through data identification and classification, data rejection and filtering and data dimension reduction, and the data are stored in a database. The slope monitoring big data with complex structure and disordered content is converted into a unified standard form, redundant information and error data are screened, and the accuracy and utilization efficiency of flood monitoring data are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of big data processing, and specifically to a method and system for flood detection and early warning of air-space-ground power equipment. Background Art

[0002] Flood disaster data, these data and information have characteristics such as heterogeneity, polymorphism, data discreteness but with certain correlations, randomness, fuzziness, and huge data volume. Existing data preprocessing technologies cannot fully meet the requirements of functions and performance, and how to effectively process these data and information is a difficult point in current flood early warning.

[0003] Therefore, how to solve the comprehensiveness and functionality of data preprocessing is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the technical problems solved by the present invention are: existing flood monitoring systems often rely on a single data source and cannot effectively integrate historical meteorological data, airborne and spaceborne SAR data. In flood monitoring, the integrity and accuracy of data are crucial. Existing methods often lack sufficient evaluation of data quality, which is prone to lead to misjudgment. Existing technologies may miss important information during the feature extraction and fusion process, resulting in a decrease in the prediction accuracy of the model.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: A method for flood detection and early warning of air-space-ground power equipment, which includes the following steps.

[0007] Classify and identify data according to the keywords of flood monitoring data; preprocess the database; extract the statistical feature vectors and time series feature vectors of flood monitoring data; perform multi-source data feature fusion on the statistical feature vectors and time series feature vectors; use the fused feature vectors as the input of machine learning methods to construct a model of flood danger situations.

[0008] As a preferred scheme of the method for flood detection and early warning of air-space-ground power equipment according to the present invention, wherein: keywords of flood monitoring data are established in the database before classifying and identifying the data.

[0009] The flood monitoring data includes historical meteorological data, airborne and spaceborne SAR data, ground sensor data, and digital surface model geographic information basic data.

[0010] As a preferred scheme of the method for flood detection and early warning of air-space-ground power equipment according to the present invention, wherein: data is marked according to the keywords of the flood monitoring data.

[0011] After classifying and identifying the data, the detected data is obtained, and according to the data characteristics and data analysis requirements, the data is classified according to the categories and structures of the detected data.

[0012] As a preferred solution of the method for detecting and warning floods in air-space-ground power equipment according to the present invention, wherein: the preprocessing includes removing and filtering redundant information and error data in the database.

[0013] As a preferred solution of the method for detecting and warning floods in air-space-ground power equipment according to the present invention, wherein: a preset standard format is set based on the detected data, and the preset standard format is written according to the actual use requirements of the meteorological data at that time, including structured, unstructured and semi-structured standard formats.

[0014] As a preferred solution of the method for detecting and warning floods in air-space-ground power equipment according to the present invention, wherein: the evaluation of the flood monitoring data includes the quality dimensions of data integrity, consistency, timeliness and effectiveness, and the reliability index and threshold of the flood monitoring data are determined according to the conditions, compositions and concurrent relationships between the quality dimensions.

[0015] As a preferred solution of the method for detecting and warning floods in air-space-ground power equipment according to the present invention, wherein: the extraction of the time series feature vector includes defining the category of the statistical feature vector, and the statistical features of each frame of data include: mean value, variance, mode, median, upper edge, upper quartile, lower quartile and lower edge.

[0016] Another object of the present invention is to provide a system for detecting and warning floods in air-space-ground power equipment, which can solve the problems of lack of multi-source data integration, insufficient real-time performance and low prediction accuracy in the existing flood monitoring system by integrating a variety of monitoring data sources, real-time data processing and intelligent prediction models.

[0017] To solve the above technical problems, the present invention provides the following technical solutions: a system for detecting and warning floods in air-space-ground power equipment, including: a data acquisition module, a data preprocessing module, a feature extraction module and a machine learning model module.

[0018] The data acquisition module is responsible for collecting flood monitoring data from multiple sources.

[0019] The data preprocessing module filters and cleans the collected data, and removes redundant information and error data.

[0020] The feature extraction module extracts statistical feature vectors and time series feature vectors from the flood monitoring data.

[0021] The machine learning model module uses the fused feature vectors as inputs to construct a prediction model for flood hazards.

[0022] A computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned method for flood detection and early warning of space-air-ground power equipment are implemented.

[0023] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned method for flood detection and early warning of space-air-ground power equipment are implemented.

[0024] Advantages of the present invention: The flood detection and early warning system for space-air-ground power equipment of the present invention has polymorphic and polygonal flood monitoring data, which is discrete and has a certain degree of correlation and randomness. Through data identification and classification, data elimination and filtering, and data dimensionality reduction, the initial data obtained is classified and cleaned, and the large-scale slope monitoring data with complex structure and messy content is transformed into a unified standard form, and redundant information and error data are screened and filtered out. Through data processing, the time series feature vectors and statistical feature vectors of the slope monitoring data are obtained, and then the fused feature vectors of the slope monitoring data are obtained, effectively improving the accuracy and utilization efficiency of the flood monitoring data. Description of the Drawings

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0026] Figure 1 It is the overall flowchart of the method for flood detection and early warning of space-air-ground power equipment provided by the first embodiment of the present invention.

[0027] Figure 2 It is the overall framework diagram of the flood detection and early warning system for space-air-ground power equipment provided by the second embodiment of the present invention. Detailed Embodiments

[0028] To make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0029] Example 1, referring to Figure 1 , which is an embodiment of the present invention, provides a method for flood detection and early warning of power equipment based on space-air-ground, characterized in that:

[0030] S1: Classify and identify data according to the keywords of flood monitoring data.

[0031] The flood monitoring data includes: historical meteorological data, space-based / spaceborne SAR data, ground sensor data (meteorology, water level, displacement), and digital surface model geographic information basic data. Combining meteorological data, SAR data, geographic information basic data, etc., carry out research on the inundation area prediction algorithm based on machine learning, and propose a method for flood inundation deduction of power line corridors. The flood and secondary disaster assessment algorithm based on single-temporal SAR images can meet the assessment of equipment area waterlogging and secondary disasters under harsh meteorological conditions such as heavy rainfall. The time-series clustering algorithm for waterlogging and secondary disasters based on short-time high-frequency revisit SAR images can realize dynamic monitoring of waterlogging and secondary disasters of power transmission and transformation equipment and prediction of spatio-temporal evolution trends through time-series SAR feature clustering.

[0032] S2: Preprocess the database.

[0033] The preprocessing includes data acquisition, data calibration, geometric correction, and registration.

[0034] S3: Extract the statistical feature vectors and time series feature vectors of flood monitoring data.

[0035] Combining meteorological data, SAR data, geographic information basic data, carry out research on the inundation area prediction algorithm based on machine learning, and propose a method for flood inundation deduction of power line corridors.

[0036] Mark the data according to the keywords of the flood monitoring data.

[0037] After classifying and identifying the data to obtain the detection data, according to the data characteristics and data analysis requirements, classify the detection data according to the category and structure of the data.

[0038] Set a preset standard format based on the detection data. The preset standard format is written according to the actual use requirements of the current meteorological data, including structured, unstructured, and semi-structured standard formats.

[0039] S4: Perform multi-source data feature fusion on the statistical feature vectors and time series feature vectors.

[0040] Evaluate the quality of flood monitoring data. The quality evaluation of flood monitoring data mainly includes data integrity, consistency, timeliness, and effectiveness quality dimensions. According to the conditions, composition, and concurrent relationships between the quality dimensions, determine the reliability index and threshold of flood monitoring data.

[0041] Locate and identify inferior multi-source heterogeneous data in flood monitoring data, extract features and generate patterns for instances of data quality problems of different types, and screen and locate suspected inferior data based on statistical indicators and violation pattern information to ensure the effectiveness and reliability of the detection method.

[0042] Select some known data sets for labeling, learn a random forest from the labeled training set, and then learn and label the unlabeled training set. During the labeling process, the more consistent the prediction results of the samples, the higher the confidence. Finally, take out the N samples with the lowest confidence and remove such samples, and then retrain this random forest until the confidence values in the unlabeled training set are all within the confidence threshold.

[0043] Define the category of statistical feature vectors. The statistical features of each frame of data include: mean, variance, mode, median, upper edge, upper quartile, lower quartile, and lower edge.

[0044] Considering that there are outlier data in each frame of data, extract the upper edge, upper quartile, lower quartile, and lower edge as statistical feature vectors;: Define the category of statistical feature vectors. The statistical features of each frame of data include: mean, variance, mode, median, upper edge, upper quartile, lower quartile, and lower edge.

[0045] Define the interval between the upper quartile and the lower quartile, and then determine the upper edge and lower edge of the data within the frame.

[0046] Extract time series feature vectors; use the clustering method to extract the time series feature vectors of the data within the frame.

[0047] Data feature fusion. Perform data feature fusion on the statistical feature vectors and time series feature vectors. The feature fusion is to merge the obtained statistical feature vectors and time series feature vectors of each frame of data to obtain the fused feature vectors.

[0048] Among them, the amount of information separating the data is measured by the information entropy, expressed as

[0049] E = -∑p(x)logp(x)

[0050] If the information entropy E exceeds the first threshold, it means that the data does not meet the first requirement, then the data points deviating from the main category are removed through the anomaly detection method. If the information entropy E is less than or equal to the first threshold, it means that the data meets the first requirement, then the impurity of the data is measured by the Gini index, expressed as

[0051] G = 1 - ∑p(x) 2

[0052] Among them, p(x) represents the probability of class x.

[0053] If the Gini index is less than or equal to the second threshold, it indicates that the data purity meets the requirements. If the Gini index is greater than the second threshold, it means that the data purity does not meet the requirements. When the data purity does not meet the requirements, detect and remove the remaining noise and outliers in the dataset.

[0054] Through the measurement of information entropy, data points deviating from the main categories in the data can be initially removed. It can significantly reduce the degree of data category mixing, improve the distinguishability between different categories, and thus provide more reliable data support for subsequent classification or prediction tasks.

[0055] First, perform preliminary filtering using information entropy, and then further refine the purity requirements using the Gini index. When the information entropy exceeds the first threshold, it indicates that the overall uncertainty of the data is high, and there may be data deviating from the main categories. After removing these data through anomaly detection, then use the Gini index to measure the purity of the remaining data, thereby gradually improving the data quality.

[0056] In the information entropy filtering stage, data points that significantly deviate from the main categories are removed; when the Gini index further measures the purity, it focuses on subtle noise and anomalies. The double-layer control effectively reduces the misjudgment caused by data anomalies during the training and prediction processes of the model, and improves the stability and robustness of the model.

[0057] After the fusion of statistical features and time series feature vectors, the double screening of information entropy and Gini index can ensure the consistency and integrity of the data after fusion, and prevent potential contradictions between features from different sources from having a negative impact on the model.

[0058] S5: Use the fused feature vector as the input of the machine learning method to construct a model for flood danger situations.

[0059] Example 2, refer to Figure 2 , which is an embodiment of the present invention, provides a system for flood detection and early warning method for air-ground-space power equipment, characterized in that it includes a data acquisition module 100, a data preprocessing module 200, a feature extraction module 300, and a machine learning model module 400.

[0060] The data acquisition module 100 is responsible for collecting flood monitoring data from multiple sources.

[0061] The data preprocessing module 200 filters and cleans the collected data, removing redundant information and incorrect data.

[0062] The feature extraction module 300 extracts statistical feature vectors and time series feature vectors from the flood monitoring data.

[0063] The machine learning model module 400 uses the fused feature vectors as input to construct a prediction model for flood danger situations.

[0064] If the above-mentioned functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, etc., all of which can store program codes.

[0065] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or in combination with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in combination with an instruction execution system, apparatus, or device.

[0066] More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection parts with one or more wirings (electronic devices), portable computer disk cartridges (magnetic devices), random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories (EPROMs or flash memories), fiber optic devices, and portable compact disc read-only memories (CDROMs). Additionally, a computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, then editing, interpreting, or otherwise processing it as necessary, and then storing it in a computer memory.

[0067] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0068] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A method for flood detection and early warning of air, space and ground power equipment, characterized in that Including: Classify and identify data according to the keywords of flood monitoring data; Preprocess the database; Extract the statistical feature vectors and time series feature vectors of flood monitoring data; Perform multi-source data feature fusion on the statistical feature vectors and time series feature vectors; Use the fused feature vectors as the input of machine learning methods to construct a model for flood hazards.

2. The flood detection and early warning method for power equipment based on space-air-ground as claimed in claim 1, wherein: Before classifying and identifying the data, establish keywords for flood monitoring data in the database; The flood monitoring data includes historical meteorological data, airborne and spaceborne SAR data, ground sensor data, and digital surface model geographic information basic data.

3. The method for flood detection and early warning of air, space and ground power equipment according to claim 2, wherein: Mark the data according to the keywords of the flood monitoring data; After classifying and identifying the data, obtain the detection data, and classify the detection data according to the data characteristics and data analysis requirements, aiming at the categories and structures of the detection data.

4. The method for detecting and warning floods in air, space and ground-based power equipment according to claim 3, wherein: The preprocessing includes removing and filtering redundant information and incorrect data in the database.

5. The method for detecting and warning floods in air, space and ground-based power equipment according to claim 4, wherein: Set a preset standard format based on the detection data. The preset standard format is written according to the actual usage requirements of the meteorological data at that time, including structured, unstructured, and semi-structured standard formats.

6. The method for flood detection and early warning of air, space and ground power equipment according to claim 5, characterized in that: The evaluation of the flood monitoring data includes the quality dimensions of data integrity, consistency, timeliness, and effectiveness. Determine the reliability indicators and thresholds of the flood monitoring data according to the conditions, compositions, and concurrent relationships between the quality dimensions.

7. The flood detection and early warning method for airborne, spaceborne and terrestrial power equipment according to claim 6, wherein: The extraction of the time series feature vector includes defining Categories of statistical feature vectors. The statistical features of each frame of data include: mean, variance, mode, median, upper edge, upper quartile, lower quartile, and lower edge.

8. A system adopting the method for flood detection and early warning of power equipment based on space-air-ground as described in any one of claims 1 to 7, characterized in that: Including a data acquisition module (100), a data preprocessing module (200), a feature extraction module (300), and a machine learning model module (400); The data acquisition module (100) is responsible for collecting flood monitoring data from multiple sources; The data preprocessing module (200) filters and cleans the collected data, and removes redundant information and incorrect data; The feature extraction module (300) extracts statistical feature vectors and time series feature vectors from flood monitoring data; The machine learning model module (400) uses the fused feature vectors as the input to construct a prediction model for flood hazards.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for flood detection and early warning of air-ground-space power equipment according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for flood detection and early warning of air-ground-space power equipment according to any one of claims 1 to 7.

Images