A model construction method for water conservancy information analysis

Abstract

The application relates to the technical field of data processing, in particular to a model construction method for water conservancy information analysis, which comprises the following steps: acquiring regional information corresponding to a target monitoring area and a plurality of water conservancy data; determining the water conservancy data selected according to the number of similar monitoring parameters or parameter characteristic values according to a difference data comparison value; segmenting monitoring time periods corresponding to the selected water conservancy data based on a first division point and a second division point; determining whether to perform segmentation optimization according to a characteristic difference between a water level characteristic value corresponding to a third division paragraph and a preset water level characteristic value; and taking the segmented water conservancy data as training data to construct a target model. The application can improve the accuracy of model risk prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a model construction method for water conservancy information analysis. BACKGROUND

[0002] In water conservancy projects, once the water level rises sharply or the water flow increases suddenly, it is often difficult to make effective feedback quickly and timely, and to reserve enough time to deal with potential risks, which undoubtedly greatly increases the safety hazards. Although the prior art has tried to use models for prediction, but in the model training stage, due to the lack of data segmentation mechanism matching the dynamic characteristics of hydrological events, the division and selection method of data is still insufficient, which affects the prediction effect and limits the prediction accuracy. Therefore, how to select and segment the training data to improve the accuracy of risk prediction is a problem to be solved by those skilled in the art.

[0003] Chinese Patent Publication No. CN117371337A discloses a water conservancy model construction method and system based on digital twinning, including: collecting data by a collection point located in a detection area, and using a convolutional neural network model to establish a water conservancy digital twinning model; performing data analysis on a plurality of sets of monitoring data obtained from the monitoring point and obtaining a data quality set, generating a stability coefficient from the data quality set, and screening out target data, using the target data as input, using the trained water conservancy digital twinning model to obtain corresponding prediction data, and generating a corresponding error coefficient according to the deviation of the prediction data and the actual data, if it exceeds the error threshold, selecting a corresponding correction scheme according to the parameter characteristics and algorithm characteristics of the abnormal module, and correcting the water conservancy twinning model. However, the above-mentioned scheme has the following problems: only the data quality set is used to generate the stability coefficient to screen the target data, which cannot fully consider the diversity and complexity of the data under different monitoring scenarios and does not segment the data, resulting in poor accuracy of model risk prediction. SUMMARY

[0004] Therefore, the present application provides a model construction method for water conservancy information analysis to overcome the problem that in the prior art, only the data quality set is used to generate the stability coefficient to screen the target data, which cannot fully consider the diversity and complexity of the data under different monitoring scenarios and does not segment the data, resulting in poor accuracy of model risk prediction.

[0005] To achieve the above-mentioned purpose, the present application provides a model construction method for water conservancy information analysis, comprising:

[0006] obtaining the regional information corresponding to the target monitoring area and a plurality of water conservancy data;

[0007] The water conservancy data is selected according to the similar monitoring parameter quantity or the parameter characteristic value according to the difference data comparison value;

[0008] The first division point is determined according to the characteristic coefficient corresponding to the first time period, and the second division point is determined based on the weather parameter characteristic value of the first reference interval, and the selected monitoring time period corresponding to each water conservancy data is segmented based on the first division point and the second division point;

[0009] The characteristic difference between the water level characteristic value corresponding to the third division paragraph and the preset water level characteristic value is determined to determine whether to perform segmentation optimization;

[0010] In the segmentation optimization, the associated combination is determined according to the parameter association degree or the similar parameter quantity proportion according to the associated reference value, and the division point adjustment or the characteristic paragraph is selected for segmentation according to the combination characteristic value corresponding to each associated combination to segment the monitoring time period corresponding to the water conservancy data;

[0011] The water conservancy data to be segmented is used as training data to construct a target model.

[0012] Further, the confirmation method of the difference data comparison value comprises:

[0013] The similar data combination is determined based on the information association degree, the combination fluctuation degree is determined based on the standard deviation of the combination evaluation value corresponding to each similar data combination, and the difference data comparison value is determined based on the difference between the combination fluctuation degree and the preset combination fluctuation degree.

[0014] Further, if the difference data comparison value is greater than or equal to the preset difference data comparison value, the water conservancy data is selected according to the similar monitoring parameter quantity;

[0015] When the water conservancy data is selected according to the similar monitoring parameter quantity, the water conservancy data with a similar monitoring parameter quantity corresponding to the regional information greater than a preset similar monitoring parameter quantity is selected.

[0016] Further, if the difference data comparison value is less than the preset difference data comparison value, the water conservancy data is selected according to the parameter characteristic value;

[0017] When the water conservancy data is selected according to the parameter characteristic value, data selection analysis is performed for each parameter, and when data selection analysis is performed for a single parameter, each water conservancy data is sorted in order of the coefficient of variation corresponding to the parameter from small to large, and the water conservancy data is selected according to a preset interval number corresponding to the parameter;

[0018] The preset interval number corresponding to the single parameter is determined according to the coefficient of variation comparison value corresponding to the parameter.

[0019] Further, the first division point is determined according to the characteristic coefficient corresponding to the first time period, comprising:

[0020] The monitoring period corresponding to the water conservancy data is divided into three equal parts, two equal points are obtained, and then the first period, the second period and the third period are sequentially divided in the order from early to late time;

[0021] If the characteristic coefficient corresponding to the first period is greater than or equal to the preset characteristic coefficient, the equal point with earlier time is recorded as the first division point;

[0022] If the characteristic coefficient corresponding to the first period is less than the preset characteristic coefficient, the absolute value of the difference between the characteristic coefficient and the preset characteristic coefficient is adjusted to increase the first period, and the last time point of the first period after the increase adjustment is recorded as the first division point.

[0023] Further, the second division point is determined based on the weather parameter characteristic value of the first reference interval, comprising:

[0024] If the weather parameter characteristic value of the first reference interval is greater than or equal to the preset weather parameter characteristic value, the equal point with later time is recorded as the second division point;

[0025] If the weather parameter characteristic value of the first reference interval is less than the preset weather parameter characteristic value, the absolute value of the difference between the weather parameter characteristic value and the preset weather parameter characteristic value is adjusted to decrease the third period, and the earliest time point in the third period after the decrease adjustment is recorded as the second division point;

[0026] The first reference interval is the time period between the first division point and the equal point with later time.

[0027] Further, whether to perform segmentation optimization is determined according to the characteristic difference between the water level characteristic value corresponding to the third division period and the preset water level characteristic value, comprising:

[0028] If the characteristic difference is greater than or equal to the preset characteristic difference, no segmentation optimization is needed;

[0029] If the characteristic difference is less than the preset characteristic difference, segmentation optimization is performed.

[0030] Further, the associated reference value is determined according to the parameter association degree or the similar parameter quantity proportion to determine the associated combination, comprising:

[0031] If the associated reference value is greater than or equal to the preset associated reference value, the associated combination is determined according to the parameter association degree;

[0032] If the associated reference value is less than the preset associated reference value, the associated combination is determined according to the similar parameter quantity proportion.

[0033] Further, if the combination characteristic value corresponding to a single associated combination is greater than or equal to the preset combination characteristic value, the division point adjustment is performed for the associated combination;

[0034] In the adjustment of the division points, the first division paragraph and the third division paragraph are adjusted according to the paragraph comparison value, and the two adjusted division points are taken as the division points.

[0035] Further, if the combination feature value corresponding to a single association combination is less than a preset combination feature value, the feature paragraph is equally selected for the association combination.

[0036] In the feature paragraph equal selection, the feature paragraph length is determined based on the comparison difference between the paragraph comparison value and the preset paragraph comparison value, the feature paragraph is selected based on the paragraph correlation degree, and the three equal division points corresponding to each feature paragraph are taken as the division points.

[0037] Compared with the prior art, the beneficial effects of the present application are that, in the technical scheme of the present application, the difference data comparison value effectively reflects the difference degree between the evaluation value fluctuation level and the historical effective fluctuation level of each similar data combination, and then the water conservancy data is adaptively selected according to the number of similar monitoring parameters or parameter feature values according to the difference data comparison value, so that the selection of water conservancy data is more in line with the actual application scene, avoiding the problem of insufficient data representativeness or disconnection with regional characteristics caused by a single selection standard, and then the pertinence and effectiveness of the selected data are improved, providing a higher quality training data basis for the construction of subsequent water conservancy information analysis models, and then improving the accuracy and reliability of the model analysis results.

[0038] Further, in the present application, the monitoring period corresponding to each selected water conservancy data is segmented based on the first division point and the second division point, which is beneficial to improve the pertinence and accuracy of water conservancy data period division, and ensures that the division result can not only fit the actual characteristics of monitoring parameters and weather parameters, but also flexibly respond to data rules in different scenarios, providing a more reliable period division basis for the construction of subsequent target models.

[0039] Further, in the present application, the overall association closeness between the selected water conservancy data is reflected by the association reference value, and then the association combination is adaptively selected according to the parameter correlation degree or the similar parameter quantity proportion according to the association reference value, so that the division of the association combination can accurately capture the deep association between data through the parameter correlation degree when the overall correlation degree is high, and can construct effective association from the parameter commonality level by means of the similar parameter quantity proportion when the overall correlation degree is low, so that the determination of the association combination is more in line with the actual association state of the data.

[0040] Further, the deviation degree of the overall characteristics of each divided paragraph in the association combination from the reference standard is effectively reflected by the combination characteristic value corresponding to each association combination in the application, and then the combination point adjustment or characteristic paragraph selection is determined according to the combination characteristic value, which is beneficial to improve the accuracy and flexibility of water conservancy data division, and ensure that the division result can be optimized in details through fine tuning when the characteristics are stable, and the deviation can be corrected through reconfiguration of paragraphs when the characteristics deviation is large, thereby providing more reliable basic data support for the construction of subsequent target models. BRIEF DESCRIPTION OF DRAWINGS

[0041] Fig. 1 A module connection diagram of the hydrogen energy vehicle safety warning system supported by the Internet of Vehicles big data of the application;

[0042] Fig. 2 A flowchart for selecting water conservancy data according to the number of similar monitoring parameters or parameter characteristic values according to the difference data comparison value of the application;

[0043] Fig. 3 A flowchart for determining whether to perform segmented optimization according to the characteristic difference between the water level characteristic value corresponding to the third divided paragraph and the preset water level characteristic value of the application;

[0044] Fig. 4 A flowchart for determining to perform combination point adjustment or characteristic paragraph selection according to the combination characteristic value corresponding to the association combination of the application. DETAILED DESCRIPTION

[0045] In order to make the purpose and advantages of the application more clear and apparent, the application will be further described below with examples; it should be understood that the specific examples described herein are only used to explain the application, and are not used to limit the protection scope of the application.

[0046] The preferred embodiments of the application will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the application, and are not used to limit the protection scope of the application.

[0047] It should be noted that in the description of the application, the terms "upper", "lower", "left", "right", "inner", "outer" and the like indicate the direction or positional relationship of the terms based on the direction or positional relationship shown in the drawings, which is only for the convenience of description, and does not indicate or imply that the device or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the application.

[0048] Please refer to Figs. 1 to 4 The application provides a model construction method for water conservancy information analysis, which comprises:

[0049] Obtaining the region information corresponding to the target monitoring region and a plurality of water conservancy data;

[0050] determining the water conservancy data according to the similar monitoring parameter quantity or parameter characteristic value according to the difference data comparison value;

[0051] determining the first division point according to the characteristic coefficient corresponding to the first time period, determining the second division point based on the weather parameter characteristic value of the first reference interval, and segmenting the monitoring time period corresponding to each selected water conservancy data based on the first division point and the second division point;

[0052] determining whether to perform segmentation optimization according to the characteristic difference between the water level characteristic value corresponding to the third division paragraph and the preset water level characteristic value;

[0053] In the segmentation optimization, the associated combination is determined according to the parameter correlation degree or the proportion of the number of similar parameters according to the associated reference value, and the division point adjustment or the characteristic paragraph is selected for segmentation according to the combination characteristic value corresponding to each associated combination to segment the monitoring time period corresponding to the water conservancy data.

[0054] The water conservancy data to be segmented is used as training data to construct a target model.

[0055] The application scenario of the present application is the selection and segmentation of training data for model construction for water conservancy information analysis; the target monitoring area is a river or reservoir catchment area that needs to be predicted at present;

[0056] The regional information is the monitoring parameters corresponding to each time point in the preset time period of the target monitoring area, and the monitoring parameters include but are not limited to gate opening, flow rate average and water level average;

[0057] It can be understood that each monitoring area in the present application is provided with a plurality of monitoring points, the positions and quantities of which can be set by the user, and the flow rate average and the water level average respectively correspond to the average value of the flow rate and the average value of the water level of each monitoring point at a single time. The flow rate and the water level of each monitoring point are measured by a flow rate meter and a water level gauge, respectively. The gate opening is converted into an electrical signal by installing a rotary encoder or a wire type displacement sensor on the motor or transmission mechanism of the gate hoist, and the gate opening is output in real time. This is a common technical means for those skilled in the art, and will not be described in detail;

[0058] A single water conservancy data includes parameters corresponding to each time point monitored in real time by a single monitoring area over time, including monitoring parameters and weather parameters, and the weather parameters include but are not limited to rainfall, air temperature, air pressure and wind speed. Rainfall, air temperature, air pressure and wind speed are obtained by referring to a meteorological station or forecast data, and will not be described in detail;

[0059] For a single water conservancy data or regional information, a starting point is set as the starting time of the water conservancy data or regional information, and an interval point is set every 1 hour, and the starting point and each interval point are recorded as a time point;

[0060] The monitoring period corresponding to a single water conservancy data is [the earliest time point corresponding to the water conservancy data, the latest time point corresponding to the water conservancy data], and it should be noted that the lengths of the monitoring periods corresponding to the water conservancy data are the same, and the length of the monitoring period corresponding to a single water conservancy data is the time length between the earliest time point and the latest time point of the water conservancy data;

[0061] In the present application, a plurality of historical records are correspondingly set, each of which records the combination fluctuation degree, the difference data comparison value, the number of similar monitoring parameters, the variation coefficient comparison value and the interval number in the historical process of the selection and segmentation of the training data for model construction for water conservancy information analysis at least once, and each of the historical records corresponds to a qualified mark, which records whether the process of the selection and segmentation of the training data for model construction for water conservancy information analysis meets the user's demand, and the qualified mark can be recorded manually. It can be understood that the user can determine whether the process of the selection and segmentation of the training data for model construction for water conservancy information analysis meets the demand according to the self-set index, and the self-set index can be but is not limited to the number of errors, which is the cumulative number of errors in the prediction of the water level by the target model, and will not be described here.

[0062] Based on the first division point and the second division point, the monitoring periods corresponding to the selected water conservancy data are segmented in the order of time from early to late, and are sequentially divided into a first division paragraph, a second division paragraph and a third division paragraph.

[0063] Specifically, the confirmation method of the difference data comparison value includes:

[0064] The similar data combinations are determined based on the information correlation degree, the combination fluctuation degree is determined based on the standard deviation of the combination evaluation value corresponding to each similar data combination, and the difference data comparison value is determined based on the difference between the combination fluctuation degree and the preset combination fluctuation degree.

[0065] In which, the similar data combinations are determined based on the information correlation degree, including: clustering analysis is performed on each water conservancy data, clustering analysis is performed on a single water conservancy data, the water conservancy data is recorded as a target water conservancy data, the water conservancy data outside the similar data combination is recorded as a reference water conservancy data, the reference water conservancy data and the target water conservancy data with an information correlation degree greater than a preset information correlation degree are recorded in a similar data combination, and clustering analysis is continued on other water conservancy data not recorded in the similar data set, until each water conservancy data is recorded in a similar data combination, and the clustering analysis is stopped;

[0066] The information correlation degree corresponding to any two water conservancy data is the minimum value in the mean value convergence degree corresponding to each monitoring parameter; the average value of the values of the monitoring parameter corresponding to each time point of a water conservancy data is denoted as a1, and the average value of the values of the monitoring parameter corresponding to each time point of another water conservancy data is denoted as a2, and the mean value convergence degree corresponding to a single monitoring parameter is |a1-a2| / a1 and the larger value of a1 and a2;

[0067] The value of the preset information correlation degree can be determined by the user according to the actual application scenario. The greater the user's demand for improving the difference data comparison value judgment accuracy, the greater the value of the preset information correlation degree. A value of the preset information correlation degree is provided, and the preset information correlation degree is 70%.

[0068] The combined evaluation value corresponding to a single similar data combination is the average value of the maximum water level mean values corresponding to each water conservancy data in the similar data combination;

[0069] The maximum water level mean value corresponding to a single water conservancy data is the maximum value in the water level mean values corresponding to each time point in the water conservancy data;

[0070] The combined fluctuation degree is the standard deviation of the combined evaluation values corresponding to each similar data combination;

[0071] The difference data comparison value is the combined fluctuation degree minus the preset combined fluctuation degree;

[0072] The history record of the water conservancy data selected according to the data characteristic value is detected, and the average value of the combined fluctuation degrees corresponding to each history record that can meet the user's demand is denoted as the preset combined fluctuation degree.

[0073] It can be understood that the information correlation degree effectively reflects the similarity of different water conservancy data in the parameter dimension, and then the similar data combination is determined based on the information correlation degree. The average level of extreme water levels in the similar data combination is effectively reflected by the combined evaluation value corresponding to the similar data combination, and then the combined fluctuation degree is determined based on the standard deviation of the combined evaluation values corresponding to each similar data combination. The dispersion degree of the difference between extreme water levels in a single similar data combination is effectively reflected by the combined fluctuation degree, and then the difference data comparison value is determined based on the difference between the combined fluctuation degree and the preset combined fluctuation degree. The difference between the fluctuation degree of the extreme water level of different similar data combinations and the historical effective fluctuation benchmark is effectively reflected by the difference data comparison value.

[0074] Specifically, if the difference data comparison value is greater than or equal to the preset difference data comparison value, the water conservancy data is selected according to the number of similar monitoring parameters;

[0075] When the water conservancy data is selected according to the number of similar monitoring parameters, the water conservancy data with a number of similar monitoring parameters corresponding to the regional information greater than a preset number of similar monitoring parameters is selected.

[0076] The value of the preset difference data comparison value can be determined by the user according to the actual application scenario. The smaller the value of the preset difference data comparison value is, the greater the user's demand for selecting water conservancy data according to the number of similar monitoring parameters is. A method for determining the value of the preset difference data comparison value is provided. The average value of the difference data comparison values corresponding to the historical records of the user selecting water conservancy data according to the number of similar monitoring parameters is recorded as the preset difference data comparison value.

[0077] The number of similar monitoring parameters corresponding to a single water conservancy data and regional information is the total number of similar monitoring parameters corresponding to the single water conservancy data and the regional information.

[0078] For a single monitoring parameter, if the mean convergence degree of the monitoring parameter corresponding to a single water conservancy data and regional information is greater than a preset mean convergence degree, the monitoring parameter is a similar monitoring parameter corresponding to the single water conservancy data and the regional information.

[0079] It should be noted that the calculation method of the mean convergence degree of a single monitoring parameter corresponding to a single water conservancy data and regional information is the same as the calculation method of the mean convergence degree of a single monitoring parameter corresponding to any two water conservancy data. Details are not described here.

[0080] The value of the preset mean convergence degree can be determined by the user according to the actual application scenario. The greater the user's demand for improving the accuracy of similar monitoring parameter determination is, the greater the value of the preset mean convergence degree is. A method for determining the value of the preset mean convergence degree is provided. The preset mean convergence degree is 78%.

[0081] The value of the preset number of similar monitoring parameters can be determined by the user according to the actual application scenario. The greater the user's demand for improving the accuracy of model training is, the greater the value of the preset number of similar monitoring parameters is. A method for determining the value of the preset number of similar monitoring parameters is provided. The average number of similar monitoring parameters corresponding to the historical records of the user selecting water conservancy data according to the number of similar monitoring parameters is recorded as the preset number of similar monitoring parameters.

[0082] Specifically, if the difference data comparison value is less than the preset difference data comparison value, the water conservancy data is selected according to the parameter characteristic value.

[0083] When selecting water conservancy data according to the parameter characteristic value, data selection analysis is performed for each parameter. When performing data selection analysis for a single parameter, each water conservancy data is sorted in order of the coefficient of variation corresponding to the parameter from small to large. The water conservancy data is selected according to the preset interval number corresponding to the parameter.

[0084] The preset interval number corresponding to a single parameter is determined according to the coefficient of variation comparison value corresponding to the parameter.

[0085] wherein, for a single parameter, the coefficient of variation corresponding to the parameter in a single water conservancy data = the standard deviation of the values of the parameter corresponding to each time point in the water conservancy data / the average value of the values of the parameter corresponding to each time point in the water conservancy data;

[0086] The coefficient of variation span value corresponding to a single parameter is the absolute value of the difference between the maximum value and the minimum value in the coefficients of variation corresponding to the parameter in each water conservancy data;

[0087] The coefficient of variation span value corresponding to a single parameter is denoted as b, and the average value of the coefficients of variation span values corresponding to each parameter is denoted as b0; the coefficient of variation ratio value = b / b0;

[0088] The preset interval number corresponding to a single parameter is the smallest integer greater than or equal to w, w = the coefficient of variation ratio value / the average value of the coefficients of variation ratio values corresponding to each parameter in each historical record that can meet the user's demand x (the average value of the preset interval numbers corresponding to each parameter in each historical record that can meet the user's demand);

[0089] The sequence obtained by sorting each water conservancy data in the order of the coefficient of variation corresponding to the parameter from small to large is denoted as the reference sequence corresponding to the parameter;

[0090] When selecting water conservancy data according to the preset interval number corresponding to the parameter, the first water conservancy data in the reference sequence corresponding to the parameter is selected as the first selected water conservancy data, and the water conservancy data is selected according to the preset interval number corresponding to the parameter, until the number of water conservancy data after the last selected water conservancy data is less than the preset interval number, then the selection of water conservancy data for the parameter is stopped;

[0091] The preset interval number corresponding to a single parameter is the number of water conservancy data between two adjacent water conservancy data selected in the reference sequence corresponding to the parameter.

[0092] It can be understood that when the difference data comparison value is greater than or equal to the preset difference data comparison value, it means that the fluctuation degree of the combination evaluation value of each similar data combination exceeds the historical fluctuation level that meets the demand, and the difference in evaluation value between each combination is more significant. At this time, it is necessary to prioritize the matching of the selected water conservancy data and the core features of the target area, so the water conservancy data is selected according to the number of similar monitoring parameters;

[0093] When the difference data comparison value is less than the preset difference data comparison value, it means that the fluctuation degree of the combination evaluation value of each similar data combination is lower than the historical effective fluctuation level, and the difference in evaluation value between each combination is small. At this time, more attention should be paid to the feature distribution of the data itself to ensure the representativeness and diversity of the selected data, so the water conservancy data is selected according to the parameter feature value.

[0094] Specifically, the first division point is determined according to a characteristic coefficient corresponding to the first time period, including:

[0095] The monitoring time period corresponding to the water conservancy data is divided into three equal parts to obtain two equal division points, and then the first time period, the second time period and the third time period are sequentially divided in the order of time from early to late;

[0096] If the characteristic coefficient corresponding to the first time period is greater than or equal to the preset characteristic coefficient, the equal division point with earlier time is recorded as the first division point;

[0097] If the characteristic coefficient corresponding to the first time period is less than the preset characteristic coefficient, the first time period is adjusted according to the absolute value of the difference between the characteristic coefficient and the preset characteristic coefficient, and the last time point of the first time period after the adjustment is recorded as the first division point.

[0098] The characteristic coefficient corresponding to the first time period is the average value of the sub-characteristic values corresponding to the monitoring parameters in the first time period;

[0099] For a single monitoring parameter, the monitoring parameter is recorded as a target monitoring parameter, and the sub-characteristic value corresponding to the target monitoring parameter is the standard deviation of the reference value of the target monitoring parameter corresponding to each water conservancy data selected;

[0100] For a single water conservancy data, the average value of the values of the target monitoring parameter corresponding to each time point in the first time period in the water conservancy data is recorded as the average value of the target monitoring parameter corresponding to the water conservancy data, and the standard deviation of the values of the target monitoring parameter corresponding to each time point in the first time period in the water conservancy data is recorded as the fluctuation value of the target monitoring parameter corresponding to the water conservancy data. The historical record in which the equal division point with earlier time is recorded as the first division point and can meet the user's demand is recorded as the first reference historical record;

[0101] The reference value of the target monitoring parameter in a single water conservancy data = the average value of the target monitoring parameter corresponding to the water conservancy data / the average value of the average value of the target monitoring parameter corresponding to each water conservancy data in each first reference historical record + the fluctuation value of the target monitoring parameter corresponding to the water conservancy data / the average value of the fluctuation value of the target monitoring parameter corresponding to each water conservancy data in each first reference historical record;

[0102] The value of the preset characteristic coefficient can be determined by the user according to the actual application scene. The smaller the value of the preset characteristic coefficient, the greater the demand of the user to record the equal division point with earlier time as the first division point. A method for determining the value of the preset characteristic coefficient is provided. The historical record in which the equal division point with earlier time is recorded as the first division point and can meet the user's demand is recorded as the preset characteristic coefficient;

[0103] The increasing value of the first time period is a minimum integer greater than or equal to w0, w0 = |characteristic coefficient - preset characteristic coefficient| / preset characteristic coefficient x number of time points in the first time period x ratio coefficient, wherein the ratio coefficient is 1 / 4, when the absolute value of the difference between the characteristic coefficient and the preset characteristic coefficient is used to increase the first time period;

[0104] When the first time period is increased, the number of time points calculated above (i.e., the increasing value of the first time period) is increased from the end of the first time period (i.e., the latest time point of the original first time period) to form the first time period after the increase.

[0105] It can be understood that the characteristic coefficient corresponding to the first time period effectively reflects the overall characteristic performance of the monitoring parameter in the first time period. When the characteristic coefficient corresponding to the first time period is greater than or equal to the preset characteristic coefficient, it means that the characteristic performance of the first time period has reached the expected standard, and the monitoring data contained therein can better reflect the typical characteristics of the time period, without adjusting the time period range, and the earlier equal division point is recorded as the first division point.

[0106] When the characteristic coefficient corresponding to the first time period is less than the preset characteristic coefficient, it means that the characteristic performance of the first time period has not reached the expected standard, and the existing time period range has not fully reflected the monitoring characteristics of the time period, which is insufficient in representation, and the integrity of the characteristics needs to be enhanced by expanding the time period range. The last time point of the first time period after the increase is recorded as the first division point.

[0107] Specifically, the second division point is determined based on the weather parameter characteristic value of the first reference interval, comprising:

[0108] If the weather parameter characteristic value of the first reference interval is greater than or equal to the preset weather parameter characteristic value, the later equal division point is recorded as the second division point;

[0109] If the weather parameter characteristic value of the first reference interval is less than the preset weather parameter characteristic value, the third time period is adjusted according to the absolute value of the difference between the weather parameter characteristic value and the preset weather parameter characteristic value, and the earliest time point in the third time period after the decrease is recorded as the second division point;

[0110] The first reference interval is a time period between the first division point and the later equal division point.

[0111] The weather parameter characteristic value of the first reference interval is the average of the weather characteristic values corresponding to each weather parameter in the first reference interval.

[0112] For a single weather parameter, the weather parameter is recorded as a target weather parameter, and the sub-weather characteristic value corresponding to the target weather parameter is the standard deviation of the weather reference value corresponding to the target weather parameter in the selected water conservancy data.

[0113] For a single water conservancy data, the average value of the values of the target weather parameter corresponding to each time point in the first reference interval in the water conservancy data is recorded as the average value of the target weather parameter corresponding to the water conservancy data, and the standard deviation of the values of the target weather parameter corresponding to each time point in the first reference interval in the water conservancy data is recorded as the characteristic value of the target weather parameter corresponding to the water conservancy data; the equally divided point with a later time and meeting the user's demand is detected as the second division point and recorded as the second reference history record;

[0114] The weather reference value corresponding to the target weather parameter in a single water conservancy data = the average value of the target weather parameter corresponding to the water conservancy data / the average value of the average value of the target weather parameter corresponding to each water conservancy data in each second reference history record + the characteristic value of the target weather parameter corresponding to the water conservancy data / the average value of the characteristic value of the target weather parameter corresponding to each water conservancy data in each second reference history record;

[0115] The value of the preset weather parameter characteristic value can be determined by the user according to the actual application scene. The smaller the value of the preset weather parameter characteristic value, the greater the user's demand for recording the equally divided point with a later time as the second division point. A method for determining the value of the preset weather parameter characteristic value is provided. The average value of the weather parameter characteristic value corresponding to the history record meeting the user's demand is recorded as the preset weather parameter characteristic value;

[0116] The reduction value of the third period is the smallest integer greater than or equal to w1, w1 = |weather parameter characteristic value-preset weather parameter characteristic value| / preset weather parameter characteristic value x number of time points in the first reference interval x ratio coefficient, wherein the ratio coefficient is 1 / 4;

[0117] When the third period is reduced and adjusted, the number of time points calculated above (i.e. the reduction value of the third period) is shortened from the starting end of the third period (i.e. the earliest time point of the original third period) to form a shortened third period.

[0118] It can be understood that the weather parameter characteristic value of the first reference interval effectively reflects the overall characteristic performance of the weather parameter in the first reference interval. When the weather parameter characteristic value of the first reference interval is greater than or equal to the preset weather parameter characteristic value, it means that the weather parameter characteristic of the interval has reached the expected standard, and the weather data contained therein can well support the original setting of the second division point, and there is no need to adjust the range of the third period to record the equally divided point with a later time as the second division point;

[0119] When the weather parameter characteristic value of the first reference interval is less than the preset weather parameter characteristic value, it indicates that the weather parameter characteristic of the interval does not reach the expected standard, and the existing third time period range fails to fully reflect the necessary weather characteristic correlation, so the pertinence of the characteristic needs to be enhanced by shortening the time period, and the earliest time point in the adjusted third time period is recorded as a second division point.

[0120] Specifically, whether to perform segmentation optimization is determined according to a characteristic difference between the water level characteristic value corresponding to the third division paragraph and a preset water level characteristic value, including:

[0121] If the characteristic difference is greater than or equal to a preset characteristic difference, segmentation optimization is not needed.

[0122] If the characteristic difference is less than the preset characteristic difference, segmentation optimization is performed.

[0123] The characteristic difference = the water level characteristic value corresponding to the third division paragraph - the preset water level characteristic value.

[0124] The water level characteristic value corresponding to the third division paragraph is a standard deviation of interval water level values corresponding to each water conservancy data selected, and the interval water level value corresponding to a single water conservancy data is a maximum value in water level mean values corresponding to each time point in the third division paragraph of the water conservancy data.

[0125] The preset water level characteristic value is an average value of water level characteristic values corresponding to each historical record that does not need segmentation optimization and can meet user demand.

[0126] The preset characteristic difference value can be determined by the user according to the actual application scenario. The greater the user's demand for improving training accuracy, the smaller the preset characteristic difference value. A method for determining the preset characteristic difference value is provided. The average value of the characteristic difference corresponding to the historical record that can meet the user's demand is recorded as the preset characteristic difference.

[0127] It can be understood that the characteristic difference between the water level characteristic value corresponding to the third division paragraph and the preset water level characteristic value effectively reflects the difference between the water level characteristic of the third division paragraph and the preset value. When the characteristic difference is less than the preset characteristic difference, it indicates that the gap between the water level characteristic of the current third division paragraph and the expected standard does not reach an acceptable range, and the water level data contained is insufficient to support the training accuracy. The existing segmentation method cannot meet the user's demand for accuracy, so segmentation optimization is needed.

[0128] Specifically, the associated reference value is used to determine the associated combination according to the parameter correlation degree or the proportion of the number of similar parameters, including:

[0129] If the associated reference value is greater than or equal to a preset associated reference value, the associated combination is determined according to the parameter correlation degree.

[0130] If the association reference value is less than the preset association reference value, the association combination is determined according to the similar parameter quantity proportion.

[0131] The selected water conservancy data is recorded as selected water conservancy data, and the association reference value is the average value of the number of associated water conservancy data corresponding to each selected water conservancy data.

[0132] The associated water conservancy data corresponding to a single selected water conservancy data is the total amount of selected water conservancy data with a parameter correlation greater than a preset parameter correlation corresponding to the selected water conservancy data.

[0133] The parameter correlation degree corresponding to any two selected water conservancy data is the minimum value of the correlation coefficients corresponding to each parameter in the two selected water conservancy data.

[0134] For a single parameter, the calculation formula of the correlation coefficient r corresponding to the parameter in the two selected water conservancy data is:

[0135]

[0136] Where n is the number of time points in a single selected water conservancy data; xi and yi are the values of the parameter corresponding to the i-th time point in the two selected water conservancy data, is the average value of the values of the parameter corresponding to each time point in the selected water conservancy data corresponding to xi, is the average value of the values of the parameter corresponding to each time point in the selected water conservancy data corresponding to yi, i = 1, 2, 3, …, n.

[0137] The value of the preset association reference value can be determined by the user according to the actual application scenario. The greater the value of the preset association reference value, the greater the user's demand for determining the association combination according to the similar parameter quantity proportion. A method for determining the value of the preset association reference value is provided. The average value of the association reference value corresponding to the historical record that can meet the user's demand is recorded as the preset association reference value.

[0138] When determining the association combination according to the parameter correlation degree or the similar parameter quantity proportion, the association analysis is performed on each selected water conservancy data. When performing association analysis on a single selected water conservancy data, the selected water conservancy data is recorded as a target selected water conservancy data, and other selected water conservancy data outside the target selected water conservancy data is recorded as reference selected water conservancy data. The reference selected water conservancy data and the target selected water conservancy data that meet the preset condition are recorded as an association combination, and the association analysis is continued on the selected water conservancy data that is not recorded in the association combination until each selected water conservancy data is recorded in the association combination, and then the association analysis is stopped.

[0139] The preset condition is that the parameter correlation degree of the target selected water conservancy data is greater than the preset parameter correlation degree when the correlation combination is determined according to the parameter correlation degree.

[0140] The preset condition is that the similar parameter quantity proportion of the target selected water conservancy data is greater than the preset similar parameter quantity proportion when the correlation combination is determined according to the similar parameter quantity proportion.

[0141] The similar parameter quantity proportion of any two selected water conservancy data is the number of similar parameters corresponding to the two selected water conservancy data / the total number of parameters corresponding to the two selected water conservancy data.

[0142] For a single parameter in any two selected water conservancy data, if the correlation coefficient corresponding to the parameter is greater than the preset correlation coefficient, the parameter is a similar parameter corresponding to the two selected water conservancy data.

[0143] The value of the preset correlation coefficient can be determined by the user according to the actual application scene. The greater the user's demand for improving the accuracy of similar parameter determination, the greater the value of the preset correlation coefficient. A value of the preset correlation coefficient is provided, and the preset correlation coefficient is 0.8.

[0144] The values of the preset parameter correlation degree and the preset similar parameter quantity proportion can be determined by the user according to the actual application scene. The greater the user's demand for improving the similarity of data in the correlation combination, the greater the values of the preset parameter correlation degree and the preset similar parameter quantity proportion. A value of the preset parameter correlation degree and a value of the preset similar parameter quantity proportion are provided, the preset parameter correlation degree is 0.6, and the preset similar parameter quantity proportion is 75%.

[0145] It can be understood that the correlation reference value effectively reflects the overall correlation closeness between the selected water conservancy data. When the correlation reference value is greater than or equal to the preset correlation reference value, it means that on average each selected water conservancy data has more associated water conservancy data, and the overall correlation degree is high. At this time, the correlation combination is determined by the parameter correlation degree, which can more accurately capture the deep correlation between data.

[0146] When the correlation reference value is less than the preset correlation reference value, it means that on average each selected water conservancy data has less associated water conservancy data, and the overall correlation degree is low. At this time, the correlation combination is determined by the similar parameter quantity proportion, which can compensate for the problem of insufficient overall correlation degree through the common characteristics at the parameter level, and is more in line with the actual correlation state of data.

[0147] Specifically, if the combination feature value corresponding to a single correlation combination is greater than or equal to the preset combination feature value, the division point adjustment is performed for the correlation combination.

[0148] In the division point adjustment, the first division paragraph and the third division paragraph are adjusted according to the paragraph comparison value, and the two adjusted separation points are taken as the division points.

[0149] The historical record that can meet the user's demand and adjust the division point for the associated combination is recorded as a reference record;

[0150] The paragraph comparison value corresponding to the first division paragraph = (the monitoring parameter characteristic value - the average value of the monitoring parameter characteristic values corresponding to each associated combination in each reference record) / the average value of the monitoring parameter characteristic values corresponding to each associated combination in each reference record; the monitoring parameter characteristic value corresponding to a single associated combination is the average value of the parameter mean fluctuation degrees corresponding to each monitoring parameter in the first division paragraph of the associated combination;

[0151] The paragraph comparison value corresponding to the second division paragraph = (the weather parameter characteristic value - the average value of the weather parameter characteristic values corresponding to each associated combination in each reference record) / the average value of the weather parameter characteristic values corresponding to each associated combination in each reference record; the weather parameter characteristic value corresponding to a single associated combination is the average value of the parameter mean fluctuation degrees corresponding to each weather parameter in the second division paragraph of the associated combination;

[0152] The parameter mean fluctuation degree corresponding to a single parameter in the associated combination is the standard deviation of the parameter mean values corresponding to each selected water conservancy data in the associated combination; the parameter mean value corresponding to a single selected water conservancy data is the average value of the values of the parameter corresponding to each time point in the single division paragraph of the selected water conservancy data;

[0153] The paragraph comparison value corresponding to the third division paragraph = (the water level characteristic value - the average value of the water level characteristic values corresponding to each associated combination in each reference record) / the average value of the water level characteristic values corresponding to each associated combination in each reference record; the water level characteristic value is the standard deviation of the parameter mean values corresponding to the third division paragraph of each selected water conservancy data in the associated combination when the parameter is the water level mean value;

[0154] The combination feature value is the average value of the paragraph comparison values corresponding to each division paragraph;

[0155] The value of the preset combination feature value can be determined by the user according to the actual application scenario. The smaller the value of the preset combination feature value is, the greater the user's demand for division point adjustment is. A method for determining the value of the preset combination feature value is provided. The average value of the combination feature values corresponding to each associated combination in the historical record that can meet the user's demand is recorded as the preset combination feature value;

[0156] It should be noted that, for the first division paragraph or the third division paragraph, if the paragraph comparison value corresponding to a single division stage is greater than the average value of the paragraph comparison values corresponding to each division stage, the division stage is adjusted in a decreasing manner;

[0157] If the paragraph comparison value corresponding to a single division stage is less than the average of the paragraph comparison values corresponding to each division stage, the single division stage is adjusted by increasing;

[0158] If the paragraph comparison value corresponding to a single division stage is equal to the average of the paragraph comparison values corresponding to each division stage, no adjustment is needed for the single division stage;

[0159] The first division paragraph and the third division paragraph are adjusted, and when a single division paragraph is adjusted, the adjustment amount corresponding to the single division stage is the smallest integer greater than or equal to w2, w2=|the paragraph comparison value corresponding to the division stage-the average of the paragraph comparison values corresponding to each division stage| / |the average of the paragraph comparison values corresponding to each division stage|×|the average of the lengths of each division stage|×a ratio coefficient, wherein the ratio coefficient is 1 / 4.

[0160] The latest time point in the division paragraph after adjustment for the first division paragraph and the earliest time point in the division paragraph after adjustment for the third division paragraph are recorded as division points, and the monitoring period corresponding to the water data is segmented based on the two division points.

[0161] Specifically, if the combination feature value corresponding to a single associated combination is less than the preset combination feature value, the associated combination is selected by equally dividing the feature paragraph;

[0162] In the equally divided selection of the feature paragraph, the feature paragraph length is determined based on the comparison difference between the paragraph comparison value and the preset paragraph comparison value, and the feature paragraph is selected based on the paragraph association degree, and the three equal division points corresponding to each feature paragraph are taken as division points.

[0163] The comparison difference is |the combination feature value-the preset combination feature value|;

[0164] The historical records that meet the user's demand and are detected to be equally divided and selected for the feature paragraph are recorded as effective historical records;

[0165] The feature paragraph length is the smallest integer greater than or equal to w3, w3=the comparison difference corresponding to a single associated combination / average of the comparison differences corresponding to each associated combination in each effective historical record×average of the feature paragraph lengths corresponding to each associated combination in each effective historical record;

[0166] The feature paragraph length is the number of time points in a single time paragraph;

[0167] When selecting a feature paragraph for a single water data, any data with a length of the feature paragraph length in a water data is randomly selected;

[0168] When selecting other water data, the feature paragraph is selected as the split paragraph with the smallest paragraph association degree with the selected feature paragraph.

[0169] The paragraph correlation degree of the single split paragraph and the selected feature paragraph is the average of the parameter correlation degrees of the single split paragraph and each selected feature paragraph. It should be noted that the calculation method of the parameter correlation degrees of the split paragraph and the selected feature paragraph is the same as that of the parameter correlation degrees of the two water conservancy data, and the only difference is that the number of time points contained in the two data is different, which is easily understood by those skilled in the art, and will not be described in detail.

[0170] For a single water conservancy data, the split analysis is performed for each time point in the order from early to late. For a single time point, a data paragraph with the time point as the starting point and the length of the feature paragraph is recorded as a split paragraph, and the split analysis is stopped until the number of time points that have not been split analyzed is less than the length of the feature paragraph.

[0171] It can be understood that the combined feature value intuitively quantifies the deviation degree of the overall feature of each divided paragraph in the associated combination from the reference standard;

[0172] When the combined feature value of a single associated combination is greater than or equal to the preset combined feature value, it means that the deviation of the overall feature of each divided paragraph in the associated combination from the reference standard is in an acceptable range, and the feature performance of the existing divided paragraph has a basis for further optimization. At this time, by adjusting the division point, the division can be more in line with the data details on the premise of retaining the existing paragraph framework, so the division point adjustment for the associated combination is selected.

[0173] When the combined feature value is less than the preset combined feature value, it indicates that the overall feature of each divided paragraph deviates greatly from the reference standard, and the feature performance of the existing paragraph has difficulty supporting effective division point adjustment. Continuing to use the original paragraph framework may lead to accumulated deviation, so it is necessary to select more representative feature paragraphs by feature paragraph equal division to correct the deviation and reconstruct the division basis from the paragraph level to improve the feature matching degree of the associated combination.

[0174] The three paragraphs divided according to the water conservancy data are sequentially recorded as the first paragraph, the second paragraph and the third paragraph in the order from early to late;

[0175] The step of taking the segmented water conservancy data as training data to build the target model includes building a training sample set, feature extraction, data set division, and setting hyperparameters. When building the training sample set, based on the divided three-section data, the training sample set is built according to the "input-output" logic: the input of a single sample is the monitoring parameter sequence (including gate opening, flow rate average, and water level average, etc.) of the first paragraph and the weather parameter sequence (including rainfall, air temperature, air pressure, and wind speed, etc.) of the second paragraph, and the output is the maximum water level average of the third paragraph; when feature extraction, the statistical features of the first paragraph and the second paragraph are extracted, the statistical features corresponding to the first paragraph include but are not limited to the peak value (such as the maximum flow rate) of each parameter, the standard deviation (reflecting the stability of fluctuation) and the cumulative value (such as the total rainfall); the statistical features corresponding to the second paragraph include but are not limited to the change rate (such as the hourly air temperature change) of the weather parameter and the extreme value duration (such as the duration of heavy rain ≥ 50 mm / h), and the high-dimensional features are compressed into low-dimensional principal components by principal component analysis or factor analysis, to reduce redundancy and improve training efficiency; when data set division, the data set is divided into training set (70%), validation set (10%), and test set (20%) according to the ratio of 7:1:2, the user can set the hyperparameters according to the actual demand, the hyperparameters include but are not limited to vector dimension, learning rate, and iteration number, all of the above are common technical means for those skilled in the art, and will not be described in detail;

[0176] It should be noted that the data length of the first paragraph and the second paragraph input by the user is not fixedly limited, the user can flexibly set according to the actual analysis demand, and the user can obtain the weather monitoring parameters of the second paragraph through weather forecast.

[0177] So far, the technical solutions of the present application have been described in combination with the preferred embodiments shown in the drawings, but those skilled in the art can easily understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to related technical features without departing from the principles of the present application, and the technical solutions after the changes or replacements will fall within the protection scope of the present application.

Claims

1. A model construction method for water conservancy information analysis, characterized in that, include: Acquire regional information and some water conservancy data corresponding to the target monitoring area; Based on the comparison values ​​of the difference data, water conservancy data are selected according to the number of similar monitoring parameters or the characteristic values ​​of the parameters; The first division point is determined based on the characteristic coefficient corresponding to the first time period, and the second division point is determined based on the weather parameter characteristic value of the first reference interval. Based on the first division point and the second division point, the monitoring time period corresponding to each selected water conservancy data is segmented. Whether to perform segmentation optimization is determined based on the feature difference between the water level feature value corresponding to the third segment and the preset water level feature value. In the segmented optimization, the correlation combination is determined based on the correlation reference value, the correlation degree of parameters or the proportion of similar parameters, and the division point adjustment or feature segment selection is carried out according to the combination feature value corresponding to each correlation combination to segment the monitoring period corresponding to the water conservancy data. Segmented water conservancy data will be used as training data to construct the target model; The methods for confirming the difference data comparison values ​​include: Similar data combinations are determined based on information correlation, and the combination volatility is determined based on the standard deviation of the combination evaluation value corresponding to each similar data combination. The difference between the combination volatility and the preset combination volatility is used to determine the difference data comparison value. The first dividing point is determined based on the characteristic coefficients corresponding to the first time period, including: Divide the monitoring period corresponding to the water conservancy data into three equal parts to obtain two equal division points, and then divide it into the first period, the second period, and the third period in order from morning to evening; If the characteristic coefficient corresponding to the first time period is greater than or equal to the preset characteristic coefficient, then the earlier division point is recorded as the first division point. If the characteristic coefficient corresponding to the first time period is less than the preset characteristic coefficient, then the first time period is increased according to the absolute value of the difference between the characteristic coefficient and the preset characteristic coefficient, and the last time point of the first time period after the increase adjustment is recorded as the first division point. The second division point is determined based on the weather parameter characteristic values ​​of the first reference interval, including: If the weather parameter characteristic value of the first reference interval is greater than or equal to the preset weather parameter characteristic value, then the later time division point is recorded as the second division point. If the weather parameter characteristic value of the first reference interval is less than the preset weather parameter characteristic value, then the third time period is reduced based on the absolute value of the difference between the weather parameter characteristic value and the preset weather parameter characteristic value, and the earliest time point in the third time period after the reduction adjustment is recorded as the second division point. The first reference interval is the time segment between the first division point and the later division point.

2. The model construction method for water conservancy information analysis according to claim 1, characterized in that, If the difference data comparison value is greater than or equal to the preset difference data comparison value, then water conservancy data are selected according to the number of similar monitoring parameters; When selecting water conservancy data based on the number of similar monitoring parameters, select water conservancy data whose number of similar monitoring parameters corresponding to the regional information is greater than the preset number of similar monitoring parameters.

3. The model construction method for water conservancy information analysis according to claim 2, characterized in that, If the difference data comparison value is less than the preset difference data comparison value, then water conservancy data is selected based on the parameter characteristic value; When selecting water conservancy data based on parameter characteristic values, data selection and analysis are performed for each parameter. When selecting and analyzing data for a single parameter, the water conservancy data are sorted in ascending order of the coefficient of variation corresponding to that parameter, and water conservancy data are selected according to the preset interval number corresponding to that parameter. The number of preset intervals corresponding to a single parameter is determined based on the coefficient of variation comparison value corresponding to that parameter.

4. The model construction method for water conservancy information analysis according to claim 1, characterized in that, Whether to perform segmentation optimization is determined based on the feature difference between the water level feature value corresponding to the third segment and the preset water level feature value, including: If the feature difference is greater than or equal to the preset feature difference, then no segmentation optimization is required; If the feature difference is less than the preset feature difference, then segmented optimization is performed.

5. The model construction method for water conservancy information analysis according to claim 4, characterized in that, The association combination is determined based on the correlation reference value, the correlation degree of parameters, or the proportion of similar parameters, including: If the associated reference value is greater than or equal to the preset associated reference value, the associated combination is determined based on the parameter association degree; If the associated reference value is less than the preset associated reference value, the associated combination is determined based on the proportion of similar parameters.

6. The model construction method for water conservancy information analysis according to claim 5, characterized in that, If the combined feature value corresponding to a single associated combination is greater than or equal to the preset combined feature value, then the division point is adjusted for that associated combination. During the division point adjustment, adjustments are made to the first and third paragraph divisions based on the paragraph comparison values, and the two adjusted dividing points are used as the division points.

7. The model construction method for water conservancy information analysis according to claim 5, characterized in that, If the feature value of a single associated combination is less than the preset feature value, then the feature segments of that associated combination are selected in equal parts. In the selection of feature paragraphs, the length of the feature paragraph is determined based on the comparison difference between the paragraph comparison value and the preset paragraph comparison value, and the feature paragraphs are selected based on the paragraph relevance. The trisection points corresponding to each feature paragraph are used as the division points.

Images