Electromagnetic environment data prediction method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to an electromagnetic data prediction method and device, equipment, a storage medium and a program product. The method comprises the following steps: acquiring historical electromagnetic environment data, influence factors and electromagnetic sensor information; classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; classifying the influence factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; constructing a space-time cube according to the dependent variable data set and the sensor information; according to the independent variable data set and the space-time cube, screening out influence factors meeting conditions; based on a distributed lag nonlinear model, analyzing a lag effect of the influence factor to obtain a maximum lag period of the influence factor; and based on the Bayesian maximum entropy, setting a time point and a space lag distance, carrying out space-time interpolation on the historical electromagnetic environment data, and predicting the electromagnetic environment data. By adopting the method, electromagnetic induction data can be predicted, corresponding conditions can be handled in advance, and harm can be prevented.

Description

Technical Field

[0001] The present application relates to the technical field of data analysis, and in particular to an electromagnetic environment data prediction method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] With the continuous development of the power industry, the scale of substation construction has been continuously expanded. In recent years, many cities have carried out reconstruction and expansion of power supply grids. The electromagnetic radiation generated by high-voltage transmission lines will inevitably cause environmental pollution. Therefore, the study of the impact of electromagnetic radiation on the environment has very important engineering application value and academic significance.

[0003] Although technicians in related fields pay more and more attention to the monitoring of electromagnetic environment, and have realized electromagnetic monitoring based on various data collection and data analysis to analyze its impact on human body and communication lines in the environment, the existing technology can only monitor the changes of electromagnetic data in real time through electromagnetic sensors. When adverse changes in electromagnetic data are detected, it is too late to do further processing and the harm has already been caused. Summary of the invention

[0004] Based on this, it is necessary to provide an electromagnetic environment data prediction method, device, computer equipment, computer-readable storage medium and computer program product that can predict electromagnetic environment data in advance in response to the above technical problems.

[0005] In a first aspect, the present application provides an electromagnetic environment data prediction method, comprising: Acquire historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor; Statistically classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; Statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; Constructing a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor; According to the independent variable data set and the space-time cube, filtering out influencing factors that meet preset conditions; Based on the distributed hysteresis nonlinear model, the hysteresis effect of the selected influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the final influencing factors; Based on Bayesian maximum entropy, the set time point and space lag distance are used to perform spatiotemporal interpolation on the historical electromagnetic environment data to predict the electromagnetic environment data; the time distance between the time point and the current time point does not exceed the maximum lag period.

[0006] In one embodiment, the space-time cube is constructed according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor, including: Aggregating the dependent variable data set and the related information to obtain the spatiotemporal characteristics of the spatiotemporal cube; Compressing the dependent variable data set and the related information according to the spatiotemporal characteristics to obtain compressed data; The space-time cube is obtained based on the compressed data.

[0007] In one embodiment, the influencing factors meeting the preset conditions are screened out according to the independent variable data set and the space-time cube, including: Converting the independent variable data in the independent variable data set into categorical variables; According to the dependent variable data corresponding to the categorical variable in the dependent variable data set, obtaining the spatial heterogeneity value of the independent variable data; The independent variable data corresponding to the spatial heterogeneity values ​​exceeding the preset values ​​are regarded as the influencing factors that meet the preset conditions.

[0008] In one embodiment, converting the independent variable data in the independent variable data set into categorical variables includes: When the independent variable is a type, the type corresponding to the independent variable is converted into a categorical variable; when the independent variable data is a numerical value, the independent variable data is classified based on a classification condition to obtain a categorical variable; wherein the classification condition includes that when the independent variable data is grouped, after the independent variable data is grouped, the within-group variance formed by the final grouping is the smallest among the within-group variances formed when the independent variable data is divided into each group, and the between-group variance formed by the final grouping is the largest among the between-group variances formed when the independent variable data is divided into each group.

[0009] In one embodiment, the method of analyzing the hysteresis effect of the selected influencing factors on the electromagnetic environment data based on the distributed hysteresis nonlinear model to obtain the maximum hysteresis period of the selected influencing factors includes: Based on the Akaike information criterion, a target combination is selected from a plurality of combinations consisting of the historical electromagnetic environment data and the screened influencing factors; Based on the target combination, a least square linear regression model is established between the historical electromagnetic environment data and the screened influencing factors; In the process of performing stepwise regression on the least squares regression model, a collinearity test is performed on the historical electromagnetic environment data and the screened influencing factors in the target combination to obtain a variance inflation coefficient value corresponding to the target combination, and the screened influencing factors included in the target combination whose variance inflation coefficient value is less than a preset value are used as the influencing factors of the electromagnetic environment data finally determined; The basis functions corresponding to the historical electromagnetic environment data and the basis functions corresponding to the final influencing factors are determined respectively, and the distributed lag nonlinear models of different lag periods are fitted respectively through the basis functions of the two, and the lag period with the best fitting effect is selected as the maximum lag period of the final influencing factor.

[0010] In one embodiment, based on Bayesian maximum entropy, the time point and the spatial lag distance are set, the historical electromagnetic environment data is subjected to spatiotemporal interpolation to predict the electromagnetic environment data, including: According to the area to be predicted, set the time point and spatial lag distance; Based on the final impact factor, the distribution of the historical electromagnetic environment data is displayed in a preset manner; The time series characteristics and spatial distribution characteristics of the historical electromagnetic environment data are respectively displayed; the time series characteristics are for a single spatial position; the spatial distribution characteristics are for a single time point; Analyze the distribution situation and the time series characteristics and the spatial distribution characteristics, and when the overall average trend characteristic value of the historical electromagnetic environment data is less than a preset trend characteristic value, remove the average trend of the historical electromagnetic environment data; the average trend characteristic value represents the degree of flatness of the overall average trend; Determine the original covariance between the set time point and the spatial lag distance, modify the time point and the spatial lag distance, and calculate the corresponding experimental covariance after each modification, and determine the final time point and spatial lag distance from each modified time point and spatial lag distance according to the comparison result between the corresponding experimental covariance after each modification and the original covariance; Based on the final time point and the spatial lag distance, the historical electromagnetic environment data is subjected to spatiotemporal interpolation to obtain a prediction result of the electromagnetic environment data.

[0011] In a second aspect, the present application also provides an electromagnetic environment data prediction device, comprising: Acquire historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor; Statistically classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; Statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; Constructing a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor; According to the independent variable data set and the space-time cube, filtering out influencing factors that meet preset conditions; Based on the distributed hysteresis nonlinear model, the hysteresis effect of the selected influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the final influencing factors; Based on Bayesian maximum entropy, the set time point and space lag distance are used to perform spatiotemporal interpolation on the historical electromagnetic environment data to predict the electromagnetic environment data; the time distance between the time point and the current time point does not exceed the maximum lag period.

[0012] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented: Acquire historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor; Statistically classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; Statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; Constructing a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor; According to the independent variable data set and the space-time cube, filtering out influencing factors that meet preset conditions; Based on the distributed hysteresis nonlinear model, the hysteresis effect of the selected influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the final influencing factors; Based on Bayesian maximum entropy, the set time point and space lag distance are used to perform spatiotemporal interpolation on the historical electromagnetic environment data to predict the electromagnetic environment data; the time distance between the time point and the current time point does not exceed the maximum lag period.

[0013] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented: Acquire historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor; Statistically classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; Statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; Constructing a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor; According to the independent variable data set and the space-time cube, filtering out influencing factors that meet preset conditions; Based on the distributed hysteresis nonlinear model, the hysteresis effect of the selected influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the final influencing factors; Based on Bayesian maximum entropy, the set time point and space lag distance are used to perform spatiotemporal interpolation on the historical electromagnetic environment data to predict the electromagnetic environment data; the time distance between the time point and the current time point does not exceed the maximum lag period.

[0014] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor: Acquire historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor; Statistically classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; Statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; Constructing a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor; According to the independent variable data set and the space-time cube, filtering out influencing factors that meet preset conditions; Based on the distributed hysteresis nonlinear model, the hysteresis effect of the selected influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the final influencing factors; Based on Bayesian maximum entropy, the set time point and space lag distance are used to perform spatiotemporal interpolation on the historical electromagnetic environment data to predict the electromagnetic environment data; the time distance between the time point and the current time point does not exceed the maximum lag period.

[0015] The above-mentioned electromagnetic environment data prediction method, device, computer equipment, computer-readable storage medium and computer program product first obtain historical electromagnetic environment sensing data, influencing factors of electromagnetic environment data and relevant information of electromagnetic sensors; the relevant information of electromagnetic sensors includes the location information of electromagnetic sensors, the time point of collecting electromagnetic environment data and the accuracy of electromagnetic sensors. The historical electromagnetic environment data is statistically classified according to the time dimension to obtain a dependent variable data set. The influencing factors of electromagnetic environment data are classified according to the time dimension corresponding to the electromagnetic induction data to obtain an independent variable data set. All data are sorted into data sets according to the corresponding time dimensions, making subsequent data processing more convenient. A time-space cube is established based on the dependent variable data set and the relevant information of the electromagnetic sensor, so as to link the time and space of the electromagnetic environment data, so that the analyzed electromagnetic environment data is more comprehensive. According to the independent variable data set and the time-space cube, the influencing factors of the electromagnetic environment data are screened according to preset conditions to obtain the influencing factors that meet the conditions. Based on the distributed lag nonlinear model, the hysteresis effect of the screened influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the screened influencing factors. Finally, based on the Bayesian maximum entropy, the time point and the number of spatial lags are set, and the historical electromagnetic environment data are interpolated in time and space to predict the electromagnetic environment data. In this way, by analyzing the historical electromagnetic environment data, the electromagnetic environment data of the area to be predicted can be predicted in advance, and processing can be done in advance to effectively prevent accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0017] FIG1 is a schematic diagram of a flow chart of an electromagnetic environment data prediction method according to an embodiment; FIG2 is a structural block diagram of an electromagnetic environment data prediction device in one embodiment; FIG. 3 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0019] In an exemplary embodiment, as shown in FIG1 , an electromagnetic environment prediction method is provided. This embodiment uses the method applied to a terminal as an example for illustration. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps 102 to 114. Among them: Step 102, obtaining historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor.

[0020] Among them, electromagnetic environment data includes electric field data, magnetic field data and electromagnetic field frequency. The influencing factors of electromagnetic environment data include substation scale, cable density, substation equipment power, substation electromagnetic shielding method, etc.

[0021] Specifically, the terminal can obtain historical electromagnetic environment data, influencing factors of the electromagnetic environment data, location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor.

[0022] Step 104 , statistically classify the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set.

[0023] Among them, the time dimensions include hours, days, months, quarters, and years, among which the year also includes peak periods and trough periods.

[0024] In some embodiments, the terminal acquires historical electromagnetic environment data as a dependent variable, and organizes the acquired historical electromagnetic environment data into a dependent variable data set according to a time dimension.

[0025] Step 106 , statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set.

[0026] Specifically, the terminal obtains the influencing factors of the electromagnetic environment data as independent variables, and organizes the factors into independent variable data sets according to the corresponding electromagnetic environment data.

[0027] Step 108, constructing a space-time cube according to the dependent variable data set and related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor.

[0028] In some of the embodiments, the terminal integrates the dependent variable data in the dependent variable data set and the related information of the electromagnetic sensor, aggregates the time and space information corresponding to the dependent variable data, and obtains a space-time cube.

[0029] Step 110, screening out influencing factors that meet preset conditions based on the independent variable data set and the space-time cube.

[0030] Among them, the preset condition is that the spatial heterogeneity value corresponding to the independent variable in the independent variable data set exceeds the preset value.

[0031] Specifically, the terminal calculates the spatial heterogeneity value of the independent variable, that is, the influencing factor of the electromagnetic environment data, and determines whether the spatial heterogeneity value exceeds a preset value. The influencing factor exceeding the preset value is the influencing factor that meets the preset conditions.

[0032] Step 112: Analyze the hysteresis effect of the selected influencing factors on the electromagnetic environment data based on the distributed hysteresis nonlinear model to obtain the maximum hysteresis period of the selected influencing factors.

[0033] Among them, the distributed lag nonlinear model is a statistical model used to describe and estimate the exposure-lag-response association. The maximum lag period is the maximum time delay involved in the impact of the independent variable on the dependent variable.

[0034] In one of the embodiments, the terminal substitutes the screened influencing factors and the corresponding electromagnetic environment data into the distributed lag nonlinear model, fits the distributed lag nonlinear models with different lag periods respectively, and calculates two indicators, namely the Akaike information criterion and the Bayesian information criterion, and selects the lag period corresponding to the relatively minimum indicator value as the maximum lag period of the screened influencing factors.

[0035] Step 114, based on the Bayesian maximum entropy, the set time point and space lag distance, perform spatiotemporal interpolation on the historical electromagnetic environment data to predict the electromagnetic environment data; the time distance between the time point and the current time point does not exceed the maximum lag period.

[0036] The spatial lag distance is the distance between the area to be predicted and the electromagnetic sensor that collects historical electromagnetic environment data. ‌Spatiotemporal interpolation‌ is a method of inferring values ​​at unknown spatiotemporal locations using known spatiotemporal data points.

[0037] Specifically, the terminal will use the Bayesian maximum entropy method to set the prediction time of the area to be predicted and the distance between the electromagnetic sensor that collects historical electromagnetic environment data, and perform spatiotemporal interpolation on the historical electromagnetic environment data to obtain the final predicted electromagnetic environment data.

[0038] In the above electromagnetic environment data prediction method, historical electromagnetic environment sensing data, influencing factors of electromagnetic environment data and relevant information of electromagnetic sensors are obtained; the relevant information of electromagnetic sensors includes the location information of electromagnetic sensors, the time point of collecting electromagnetic environment data and the accuracy of electromagnetic sensors. The historical electromagnetic environment data is statistically classified according to the time dimension to obtain a dependent variable data set. The influencing factors of electromagnetic environment data are classified according to the time dimension corresponding to the electromagnetic induction data to obtain an independent variable data set. All data are sorted into data sets according to the corresponding time dimensions, making subsequent data processing more convenient. A time-space cube is established based on the dependent variable data set and the relevant information of the electromagnetic sensor, so as to link the time and space of the electromagnetic environment data, so that the analyzed electromagnetic environment data is more comprehensive. According to the independent variable data set and the time-space cube, the influencing factors of the electromagnetic environment data are screened according to preset conditions to obtain the influencing factors that meet the conditions. Based on the distributed lag nonlinear model, the hysteresis effect of the screened influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the screened influencing factors. Finally, based on the Bayesian maximum entropy, the time point and the number of spatial lags are set, and the historical electromagnetic environment data are time-space interpolated to predict the electromagnetic environment data. In this way, by analyzing the historical electromagnetic environment data, the electromagnetic environment data of the area to be predicted can be predicted in advance, and processing can be done in advance to effectively prevent accidents.

[0039] In an exemplary embodiment, a space-time cube is constructed based on a dependent variable data set and related information, including: aggregating the dependent variable data set and related information to obtain the space-time characteristics of the space-time cube; compressing the dependent variable data set and related information according to the space-time characteristics to obtain compressed data; and obtaining the space-time cube based on the compressed data.

[0040] In actual implementation, the terminal will aggregate the dependent variable data set and the relevant information of the electromagnetic sensor, that is, aggregate the electromagnetic environment data belonging to the electromagnetic sensor with the time and space data to obtain the corresponding relationship between the electromagnetic environment data of each electromagnetic sensor. Among them, the spatiotemporal feature is that the electromagnetic environment data collected by each electromagnetic sensor at each time constitutes the time-electromagnetic time series data, which constitutes the spatiotemporal feature of the spatiotemporal cube as a whole. The space is the position of the corresponding electromagnetic sensor.

[0041] Furthermore, the terminal will compress the electromagnetic environment data obtained in the time series according to the time-space characteristics, and compress the continuous electromagnetic environment data into electromagnetic environment data at several time points according to the change of the electromagnetic environment data in the time series. All compressed data are combined to obtain a time-space cube. Among them, the specific compression method can be to take the average, or to take the actual value at fixed intervals, or other compression methods, which are not limited in the embodiments of the present application.

[0042] In the above embodiment, the electromagnetic environment data is aggregated to combine the spatial and temporal features of the electromagnetic environment data, so that the analysis result is more comprehensive. After the data is aggregated, it is compressed to reduce the difficulty of subsequent data processing.

[0043] In an exemplary embodiment, influencing factors that meet preset conditions are screened out based on the independent variable data set and the space-time cube, including: converting the independent variable data in the independent variable data set into categorical variables; obtaining the spatial heterogeneity value of the independent variable data based on the corresponding dependent variable data of the categorical variables in the dependent variable data set; and taking the independent variable data corresponding to the spatial heterogeneity value exceeding the preset value as the influencing factor that meets the preset conditions.

[0044] In an exemplary embodiment, the independent variables in the independent variable data set are converted into categorical variables, including: when the independent variable is a type, converting the type corresponding to the independent variable into a categorical variable; when the independent variable data is a numerical value, classifying the independent variable data based on a classification condition to obtain a categorical variable; wherein the classification condition includes that when the independent variable data is grouped, after the independent variable data is grouped, the within-group variance formed by the final grouping is the smallest among the within-group variances formed when the independent variable data is divided into each group, and the between-group variance formed by the final grouping is the largest among the between-group variances formed when the independent variable data is divided into each group.

[0045] In actual implementation, the terminal will determine whether the independent variable in the independent variable data set is a type quantity or a numerical quantity, and then convert the independent variable into a categorical variable according to different situations. If the independent variable is a type quantity, it is directly converted into a categorical variable; if the independent variable is a numerical quantity, it is grouped or stratified, and the variance within the group is minimized and the variance between groups is minimized. Among them, the categorical variable can be a classification of the dependent variable or an explanatory variable of the dependent variable.

[0046] Specifically, the grouping method includes grouping according to expert knowledge / historical experience / experimental verification, or according to a classification algorithm, and each group or layer formed after classification of a single independent variable contains at least two dependent variables. Among them, the classification algorithm includes the k-means algorithm and the sorting and equal division algorithm, and can also be other classification algorithms, which are not limited in the embodiments of the present application.

[0047] Furthermore, the within-group variance is the variance calculated from the variable values ​​within each group and the arithmetic mean within the group; the between-group variance is the variance calculated from the arithmetic mean of each group and the overall arithmetic mean; when grouping numerical quantities, it is necessary to ensure that the variance of the independent variable in the group formed by the final grouping based on the values ​​of each variable and the arithmetic mean within the group is the smallest, and the variance calculated based on the arithmetic mean of the group formed by the final grouping and the overall arithmetic mean is the largest.

[0048] In the above embodiment, by converting the independent variables into categorical variables, subsequent data processing is simplified, the interpretability of the final prediction results is improved, and the prediction performance of the model is improved.

[0049] In some embodiments, after the terminal converts the independent variable into a categorical variable, the data of the converted categorical variable and the corresponding dependent variable are imported into the calculation module in the physical knowledge base for calculation, and the spatial heterogeneity q value of the independent variable, the explanatory power of the independent variable on the dependent variable, or the inter-layer mean comparison result is obtained. Among them, the calculation module in the physical knowledge base is a calculation module used to support multi-sensor composite measurement. The explanatory power is expressed as 100×q%, and the inter-layer mean comparison result is the relationship between the mean of the dependent variable in the current layer of the independent variable and the mean of the dependent variable in the adjacent layer of the current layer of the independent variable.

[0050] Furthermore, a t test with a preset significance level is used to test it, and the independent variables corresponding to the spatial heterogeneity values ​​exceeding the preset value are regarded as influencing factors that meet the preset conditions.

[0051] In the above embodiment, the influencing factors of the electromagnetic environment data are further screened to determine the main influencing factors affecting the electromagnetic environment data, so that the final prediction result is more accurate.

[0052] In an exemplary implementation, based on the distributed hysteresis nonlinear model, the hysteresis effect of the selected influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the selected influencing factors, including: Specifically, based on the Akaike information criterion, the terminal selects a target combination from multiple combinations of historical electromagnetic environment data and screened influencing factors, and establishes a least squares linear regression model based on the target combination. In the process of stepwise regression of the least squares regression model, a collinearity test is performed on the historical electromagnetic environment data and the screened influencing factors in the target combination to obtain the variance inflation coefficient value corresponding to the target combination, and the screened influencing factors contained in the target combination whose variance inflation coefficient value is less than the preset value are used as the influencing factors of the electromagnetic environment data finally determined.

[0053] Furthermore, the terminal determines the basis functions corresponding to the historical electromagnetic environment data and the basis functions corresponding to the final influencing factors, selects different lag periods to fit the model, and calculates the two indicators of Akaike information criterion and Bayesian information criterion, and selects the lag period corresponding to the minimum value of the indicator as the maximum lag period of the selected influencing factors. The basis function of the electromagnetic environment data selects the natural spline function, and the basis function of the influencing factor selects the polynomial function.

[0054] In the specific implementation, the terminal will apply the distributed lag nonlinear model to analyze the lag effect of the independent variable on the electromagnetic environment data. The specific formula is shown in (1).

[0055]

[0056] Where t is time, is the independent variable, is the basis function of the independent variable, i.e. the influencing factor, is the linear effect of other confounding factors, for The coefficient of for The coefficient of , α is a custom constant, It is the electromagnetic environment data.

[0057] In another embodiment, the terminal may also use three-dimensional graphics to express the estimation result of the hysteresis effect of the distributed hysteresis nonlinear model.

[0058] In the above embodiment, by calculating the maximum lag period of the final influencing factor, the maximum time period that can be predicted in the final prediction can be determined, thereby controlling the prediction time and making the final prediction result more accurate.

[0059] In an exemplary implementation, based on Bayesian maximum entropy, the time points and spatial lag distances are set, and the historical electromagnetic environment data are temporally and spatially interpolated to predict the electromagnetic environment data, including: setting the time points and spatial lag distances according to the area to be predicted; displaying the distribution of the historical electromagnetic environment data in a preset manner according to the final influencing factors; displaying the time series characteristics and spatial distribution characteristics of the historical electromagnetic environment data respectively; the time series characteristics are for a single spatial position; the spatial distribution characteristics are for a single time point; analyzing the distribution and the time series characteristics, the spatial distribution characteristics, when the overall average trend characteristic value of the historical electromagnetic environment data When it is less than the preset trend characteristic value, the average trend of the historical electromagnetic environment data is removed; the average trend characteristic value represents the flatness of the overall average trend; the original covariance between the set time point and the spatial lag distance is determined, the time point and the spatial lag distance are modified, and the corresponding experimental covariance after each modification is calculated, and the final time point and spatial lag distance are determined from the time point and spatial lag distance of each modification according to the comparison result between the corresponding experimental covariance and the original covariance after each modification; based on the final time point and spatial lag distance, the historical electromagnetic environment data is temporally and spatially interpolated to obtain the prediction result of the electromagnetic environment data.

[0060] In actual implementation, the terminal will set the corresponding prediction time point and the distance from the electromagnetic sensor location that collects historical electromagnetic environment data according to the area to be predicted. The time series characteristics, spatial distribution characteristics and distribution of historical electromagnetic environment data are displayed in a preset manner. The time series characteristics, spatial distribution characteristics and distribution of historical electromagnetic environment data are analyzed. When the overall average trend characteristic value of the historical electromagnetic environment data is less than the preset trend characteristic value, the average trend of the historical electromagnetic environment data is removed, and the residual data after removing the average trend of the historical electromagnetic environment data is used in the final prediction; otherwise, the original historical electromagnetic environment data is used in the final prediction.

[0061] The setting method is descriptive statistics and / or histogram. The trend characteristic value is a characteristic that characterizes the flatness of the overall average trend. The flatter the trend, the smaller the trend characteristic value. The preset trend characteristic value can be set according to the actual detection situation.

[0062] Furthermore, the terminal determines the original covariance between the set time point and the spatial lag distance, then modifies the time point and the spatial lag distance, and calculates the corresponding experimental covariance after each modification, and determines the final time point and spatial lag distance from each modified time point and spatial lag distance based on the comparison result between the corresponding experimental covariance and the original covariance after each modification.

[0063] Specifically, the terminal compares the covariance curves of the time point and the spatial lag distance under different circumstances each time, and selects the time point and the spatial lag distance when the trends of the two variables are most consistent or opposite as the final time point and the spatial lag distance.

[0064] In a specific embodiment, after determining the final time point and spatial lag distance, the Bayesian maximum entropy is used to perform spatiotemporal interpolation on the historical electromagnetic environment data to obtain the prediction results of the electromagnetic environment data. The specific process includes: for spatial distribution estimation, selecting the time to be predicted, the number of nodes to be predicted on the coordinates with the independent variable as the x-axis and the influencing factor as the y-axis, and the spatial range to be calculated and displayed; for time distribution estimation, determining the location and time range of the electromagnetic sensor to be predicted.

[0065] In the above embodiment, by analyzing the overall trend of the historical electromagnetic environment data and determining whether to use the average trend of the historical electromagnetic environment data without using the average trend, data with more significant changes can be obtained. At the same time, by analyzing the variables through covariance and selecting more appropriate variable parameters, the final result is more accurate and practical.

[0066] This application conducts scenario simulation of the electromagnetic situation in the study area by comprehensively considering the time and space dimensions of the electromagnetic environment data and combining it with physics models. It effectively analyzes the electromagnetic spatiotemporal situation through the two dimensions of time and space and a variety of electromagnetic influencing factors, and can simulate the actual scenario of the electromagnetic environment data to improve the accuracy of electromagnetic assessment.

[0067] To explain the electromagnetic environment data prediction method in the present application in detail, a specific embodiment is described below.

[0068] First, the terminal can obtain historical electromagnetic environment data, the influencing factors of electromagnetic environment data, the location information of electromagnetic sensors, the time point of collecting electromagnetic environment data, and the accuracy of electromagnetic sensors. Among them, electromagnetic environment data includes electric field data, magnetic field data, and electromagnetic field frequency. The influencing factors of electromagnetic environment data include substation scale, cable density, substation equipment power, substation electromagnetic shielding method, etc.

[0069] The historical electromagnetic environment data is used as the dependent variable and organized into a dependent variable data set according to the time dimension; the time dimension includes hours, days, months, quarters, and years, and the year also includes peak periods and trough periods. The influencing factors of the electromagnetic environment data are used as independent variables and organized into independent variable data sets according to the corresponding electromagnetic environment data.

[0070] The terminal aggregates the dependent variable data set and the related information of the electromagnetic sensor, that is, aggregates the electromagnetic environment data belonging to the electromagnetic sensor with the time and space data to obtain the corresponding relationship between the electromagnetic environment data of each electromagnetic sensor. Among them, the spatiotemporal feature is that the electromagnetic environment data collected by each electromagnetic sensor at each time constitutes the time-electromagnetic time series data, which constitutes the spatiotemporal feature of the spatiotemporal cube as a whole. The space is the position of the corresponding electromagnetic sensor.

[0071] Furthermore, the terminal will compress the electromagnetic environment data obtained in time series by taking the average value according to the time-space characteristics, and compress the continuous electromagnetic environment data into electromagnetic environment data at several time points according to the change of the electromagnetic environment data in time series. All compressed data are integrated to obtain a time-space cube.

[0072] Determine whether the independent variable in the independent variable data set is a categorical variable or a numerical variable, and then convert the independent variable into a categorical variable according to different situations. If the independent variable is a categorical variable, directly convert it into a categorical variable; if the independent variable is a numerical variable, group or stratify it based on historical experience, and meet the minimum variance within the group and the minimum variance between groups.

[0073] The transformed categorical variable data and the corresponding dependent variable data are imported into the calculation module in the physical knowledge base for calculation to obtain the spatial heterogeneity q value of the independent variable, the explanatory power of the independent variable on the dependent variable, or the comparison result of the mean between layers.

[0074] Furthermore, the significance level was tested with a preset value of t=0.05, and the independent variables corresponding to the spatial heterogeneity values ​​exceeding 0.05 were regarded as qualified influencing factors.

[0075] Based on the Akaike information criterion, a target combination is selected from multiple combinations of historical electromagnetic environment data and screened influencing factors, and a least squares linear regression model is established based on the target combination. In the process of stepwise regression of the least squares regression model, a collinearity test is performed on the historical electromagnetic environment data and the screened influencing factors in the target combination to obtain the variance inflation coefficient value corresponding to the target combination, and the screened influencing factors contained in the target combination with a variance inflation coefficient value less than 10 are used as the influencing factors of the final electromagnetic environment data.

[0076] The basis functions corresponding to the historical electromagnetic environment data and the basis functions corresponding to the final influencing factors are determined respectively, and different lag periods are selected to fit the model. The two indicators of Akaike information criterion and Bayesian information criterion are calculated, and the lag period corresponding to the minimum value of the indicator is selected as the maximum lag period of the selected influencing factors. The distributed lag nonlinear model is applied to analyze the lag effect of the independent variable on the electromagnetic environment data. The specific formula is shown in (1).

[0077] According to the area you want to predict, set the corresponding prediction time point and the distance from the electromagnetic sensor location that collects historical electromagnetic environment data. Use a histogram to display the time series characteristics, spatial distribution characteristics, and distribution of historical electromagnetic environment data. Analyze the time series characteristics, spatial distribution characteristics, and distribution of historical electromagnetic environment data. When the overall average trend characteristic value of the historical electromagnetic environment data is less than the preset trend characteristic value, remove the average trend of the historical electromagnetic environment data, and use the residual data after removing the average trend of the historical electromagnetic environment data in the final prediction; otherwise, use the original historical electromagnetic environment data in the final prediction.

[0078] Determine the original covariance between the set time point and spatial lag distance, then modify the time point and spatial lag distance, and calculate the corresponding experimental covariance after each modification. According to the comparison results between the corresponding experimental covariance and the original covariance after each modification, select the time point and spatial lag distance when the trends of the two variables are most consistent or opposite from each modified time point and spatial lag distance as the final time point and spatial lag distance.

[0079] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0080] Based on the same inventive concept, the embodiment of the present application also provides an electromagnetic environment data prediction device for implementing the electromagnetic environment data prediction method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more electromagnetic environment data prediction device embodiments provided below can refer to the limitations of the electromagnetic environment data prediction method above, and will not be repeated here.

[0081] In an exemplary embodiment, as shown in FIG2 , an electromagnetic environment data prediction device is provided, comprising: an acquisition module 201, a classification module 202, a construction module 203, an analysis module 204 and a prediction module 205, wherein: An acquisition module, used to acquire historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor; A classification module, used for statistically classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; The classification module is further used to statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; A construction module, used to construct a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor; An analysis module is used to analyze the hysteresis effect of the selected influencing factors on the electromagnetic environment data based on a distributed hysteresis nonlinear model, and obtain the maximum hysteresis period of the final influencing factors; The prediction module is used to perform spatiotemporal interpolation on the historical electromagnetic environment data and predict the electromagnetic environment data based on the Bayesian maximum entropy, the set time point and the spatial lag distance; the time distance between the time point and the current time point does not exceed the maximum lag period.

[0082] In some embodiments, the above-mentioned construction module is also used to aggregate the dependent variable data set and related information to obtain the spatiotemporal characteristics of the spatiotemporal cube; compress the dependent variable data set and related information according to the spatiotemporal characteristics to obtain compressed data; and obtain the spatiotemporal cube based on the compressed data.

[0083] In some embodiments, the above-mentioned screening module is also used to convert the independent variable data in the independent variable data set into categorical variables; obtain the spatial heterogeneity value of the independent variable data according to the corresponding dependent variable data of the categorical variables in the dependent variable data set; and use the independent variable data corresponding to the spatial heterogeneity value exceeding the preset value as an influencing factor that meets the preset conditions.

[0084] In some embodiments, the device also includes a conversion module for converting the type corresponding to the independent variable into a categorical variable when the independent variable is a type; and for classifying the independent variable data based on a classification condition to obtain a categorical variable when the independent variable data is a numerical value; wherein the classification condition includes that when the independent variable data is grouped, after the independent variable data is grouped, the intra-group variance formed by the final grouping is the smallest among the intra-group variances formed when the independent variable data is divided into each group, and the inter-group variance formed by the final grouping is the largest among the inter-group variances formed when the independent variable data is divided into each group.

[0085] In some embodiments, the above-mentioned analysis module is also used to select a target combination from multiple combinations consisting of historical electromagnetic environment data and screened influencing factors based on the Akaike information criterion; based on the target combination, a least squares linear regression model between the historical electromagnetic environment data and the screened influencing factors is established; in the process of stepwise regression of the least squares regression model, a collinearity test is performed on the historical electromagnetic environment data and the screened influencing factors in the target combination to obtain the variance inflation coefficient value corresponding to the target combination, and the screened influencing factors included in the target combination whose variance inflation coefficient value is less than a preset value are used as the influencing factors of the electromagnetic environment data that are finally determined; the basis functions corresponding to the historical electromagnetic environment data and the basis functions corresponding to the final influencing factors are respectively determined, and the distributed lag nonlinear models of different lag periods are respectively fitted through the basis functions of the two, and the lag period with the best fitting effect is selected as the maximum lag period of the final influencing factor.

[0086] In some embodiments, the above-mentioned prediction module is also used to set the time point and the spatial lag distance according to the area to be predicted; according to the final influencing factor, the distribution of the historical electromagnetic environment data is displayed in a preset manner; the time series characteristics and the spatial distribution characteristics of the historical electromagnetic environment data are displayed separately; the time series characteristics are for a single spatial position; the spatial distribution characteristics are for a single time point; the distribution situation and the time series characteristics and the spatial distribution characteristics are analyzed, and when the overall average trend characteristic value of the historical electromagnetic environment data is less than the preset trend characteristic value, the average trend of the historical electromagnetic environment data is removed; the average trend characteristic value represents the degree of flatness of the overall average trend; determine the original covariance between the set time point and the spatial lag distance, modify the time point and the spatial lag distance, and calculate the corresponding experimental covariance after each modification, and determine the final time point and spatial lag distance from the time point and the spatial lag distance of each modification according to the comparison result between the corresponding experimental covariance and the original covariance after each modification; based on the final time point and the spatial lag distance, perform spatiotemporal interpolation on the historical electromagnetic environment data to obtain the prediction result of the electromagnetic environment data.

[0087] Each module in the above electromagnetic environment data device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0088] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG3. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, the memory, and the input / output interface are connected via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, a near field communication (NFC) or other technologies. When the computer program is executed by the processor, a method for predicting electromagnetic environment data is implemented.

[0089] Those skilled in the art will understand that the structure shown in FIG. 3 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0090] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented: Acquire historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor; Statistically classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; Statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; Constructing a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor; According to the independent variable data set and the space-time cube, filtering out influencing factors that meet preset conditions; Based on the distributed hysteresis nonlinear model, the hysteresis effect of the selected influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the final influencing factors; Based on Bayesian maximum entropy, the set time point and space lag distance are used to perform spatiotemporal interpolation on the historical electromagnetic environment data to predict the electromagnetic environment data; the time distance between the time point and the current time point does not exceed the maximum lag period.

[0091] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: Acquire historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor; Statistically classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; Statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; Constructing a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor; According to the independent variable data set and the space-time cube, filtering out influencing factors that meet preset conditions; Based on the distributed hysteresis nonlinear model, the hysteresis effect of the selected influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the final influencing factors; Based on Bayesian maximum entropy, the set time point and space lag distance are used to perform spatiotemporal interpolation on the historical electromagnetic environment data to predict the electromagnetic environment data; the time distance between the time point and the current time point does not exceed the maximum lag period.

[0092] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps: Acquire historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor; Statistically classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; Statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; Constructing a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor; According to the independent variable data set and the space-time cube, filtering out influencing factors that meet preset conditions; Based on the distributed hysteresis nonlinear model, the hysteresis effect of the selected influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the final influencing factors; Based on Bayesian maximum entropy, the set time point and space lag distance are used to perform spatiotemporal interpolation on the historical electromagnetic environment data to predict the electromagnetic environment data; the time distance between the time point and the current time point does not exceed the maximum lag period.

[0093] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0094] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0095] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0096] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for predicting electromagnetic environment data, characterized in that: The method comprises: Acquire historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor; Statistically classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; Statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; Constructing a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor; According to the independent variable data set and the space-time cube, filtering out influencing factors that meet preset conditions; Based on the distributed hysteresis nonlinear model, the hysteresis effect of the selected influencing factors on the electromagnetic environment data is analyzed to obtain the maximum hysteresis period of the final influencing factors; Based on Bayesian maximum entropy, the set time point and space lag distance are used to perform spatiotemporal interpolation on the historical electromagnetic environment data to predict the electromagnetic environment data; the time distance between the time point and the current time point does not exceed the maximum lag period.

2. The method according to claim 1, characterized in that The step of constructing a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor, including: Aggregating the dependent variable data set and the related information to obtain the spatiotemporal characteristics of the spatiotemporal cube; Compressing the dependent variable data set and the related information according to the spatiotemporal characteristics to obtain compressed data; The space-time cube is obtained based on the compressed data.

3. The method according to claim 1, characterized in that The step of screening out influencing factors that meet preset conditions based on the independent variable data set and the space-time cube includes: Converting the independent variable data in the independent variable data set into categorical variables; According to the dependent variable data corresponding to the categorical variable in the dependent variable data set, obtaining the spatial heterogeneity value of the independent variable data; The independent variable data corresponding to the spatial heterogeneity values ​​exceeding the preset values ​​are regarded as the influencing factors that meet the preset conditions.

4. The method according to claim 3, characterized in that The converting the independent variable data in the independent variable data set into categorical variables comprises: In the case where the independent variable is a type, the type corresponding to the independent variable is converted into a categorical variable; When the independent variable data is a numerical value, the independent variable data is classified based on a classification condition to obtain a categorical variable; wherein the classification condition includes that when the independent variable data is grouped, after the independent variable data is grouped, the intra-group variance formed by the final grouping is the smallest among the intra-group variances formed when the independent variable data is divided into each group, and the inter-group variance formed by the final grouping is the largest among the inter-group variances formed when the independent variable data is divided into each group.

5. The method according to claim 1, characterized in that The distributed hysteresis nonlinear model is used to analyze the hysteresis effect of the selected influencing factors on the electromagnetic environment data to obtain the maximum hysteresis period of the final influencing factors, including: Based on the Akaike information criterion, a target combination is selected from a plurality of combinations consisting of the historical electromagnetic environment data and the screened influencing factors; Based on the target combination, a least square linear regression model is established between the historical electromagnetic environment data and the screened influencing factors; In the process of performing stepwise regression on the least squares regression model, a collinearity test is performed on the historical electromagnetic environment data and the screened influencing factors in the target combination to obtain a variance inflation coefficient value corresponding to the target combination, and the screened influencing factors included in the target combination whose variance inflation coefficient value is less than a preset value are used as the influencing factors of the electromagnetic environment data finally determined; The basis functions corresponding to the historical electromagnetic environment data and the basis functions corresponding to the final influencing factors are determined respectively, and the distributed lag nonlinear models of different lag periods are fitted respectively through the basis functions of the two, and the lag period with the best fitting effect is selected as the maximum lag period of the final influencing factor.

6. The method according to claim 1, characterized in that The method of performing spatiotemporal interpolation on the historical electromagnetic environment data based on the Bayesian maximum entropy, setting the time point and the spatial lag distance, and predicting the electromagnetic environment data includes: According to the area to be predicted, set the time point and spatial lag distance; Based on the final impact factor, the distribution of the historical electromagnetic environment data is displayed in a preset manner; The time series characteristics and spatial distribution characteristics of the historical electromagnetic environment data are respectively displayed; the time series characteristics are for a single spatial position; the spatial distribution characteristics are for a single time point; Analyze the distribution situation and the time series characteristics and the spatial distribution characteristics, and when the overall average trend characteristic value of the historical electromagnetic environment data is less than a preset trend characteristic value, remove the average trend of the historical electromagnetic environment data; the average trend characteristic value represents the degree of flatness of the overall average trend; Determine the original covariance between the set time point and the spatial lag distance, modify the time point and the spatial lag distance, and calculate the corresponding experimental covariance after each modification, and determine the final time point and spatial lag distance from each modified time point and spatial lag distance according to the comparison result between the corresponding experimental covariance after each modification and the original covariance; Based on the final time point and the spatial lag distance, the historical electromagnetic environment data is subjected to spatiotemporal interpolation to obtain a prediction result of the electromagnetic environment data.

7. An electromagnetic environment data prediction device, characterized in that: The device comprises: An acquisition module, used to acquire historical electromagnetic environment data, influencing factors of the electromagnetic environment data, and related information of the electromagnetic sensor; the electromagnetic sensor related information includes the location information of the electromagnetic sensor, the time point of collecting the electromagnetic environment data, and the accuracy of the electromagnetic sensor; A classification module, used for statistically classifying the historical electromagnetic environment data according to the time dimension to obtain a dependent variable data set; The classification module is further used to statistically classify the influencing factors according to the time dimension corresponding to the historical electromagnetic environment data to obtain an independent variable data set; A construction module, used to construct a space-time cube according to the dependent variable data set and the related information; the space-time cube includes the electromagnetic environment data collected by any electromagnetic sensor; An analysis module is used to analyze the hysteresis effect of the selected influencing factors on the electromagnetic environment data based on a distributed hysteresis nonlinear model, and obtain the maximum hysteresis period of the final influencing factors; The prediction module is used to perform spatiotemporal interpolation on the historical electromagnetic environment data and predict the electromagnetic environment data based on the Bayesian maximum entropy, the set time point and the spatial lag distance; the time distance between the time point and the current time point does not exceed the maximum lag period.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images