Intelligent management and early warning system for hydraulic engineering monitoring data

Abstract

The invention relates to the technical field of water level monitoring, in particular to an intelligent management and early warning system for hydraulic engineering monitoring data. The system comprises a data preprocessing module used for obtaining water level data in different time periods; the distribution regularity calculation module is used for obtaining the distribution regularity of the time period according to the skewness coefficient and the kurtosis coefficient of the water level data of the time period; the stability degree calculation module is used for obtaining the fluctuation stability of the time period based on the variable coefficient of the water level data of the same time period of each year and the quantity of the abnormal water level data; integrating the distribution regularity and the fluctuation stability to obtain the stability degree; the water level normal range obtaining module is used for obtaining the water level normal range of one time period when the stability degree is larger than a stability threshold value and obtaining the water level normal range of one time period when the stability degree is smaller than or equal to the stability threshold value; and the early warning module is used for carrying out early warning on the water level according to the normal water level range of each time period. The accuracy of water level early warning can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of water level monitoring, and in particular to an intelligent management and early warning system for water conservancy project monitoring data. Background Art

[0002] Water conservancy projects are projects built to prevent and control water disasters and develop and utilize water resources, including flood control, waterlogging control, irrigation, water supply, hydropower generation, shipping, water resource protection, soil and water conservation, as well as water-related projects in aquatic products, tourism, and improving the ecological environment. Water conservancy project monitoring data includes water level data, water quality data, flow data, etc. In order to effectively reduce the possibility of disasters or mitigate the harm of disasters, it is necessary to conduct intelligent management and early warning of water conservancy project monitoring data.

[0003] The intelligent management and early warning system for water conservancy project monitoring data collects water level, water quality and other data in real time through sensors and other monitoring equipment, and then transmits the collected data to the data storage and processing platform. Thresholds are set based on the various historical monitoring data collected. When the monitoring data reaches the preset threshold, the system will automatically trigger a warning signal and then notify relevant personnel via text messages, emails, etc.

[0004] Among them, water level data is an important basis for flood warning and reasonable water resource scheduling. By monitoring the changes in water levels in real time, the safety and sustainability of water conservancy projects can be improved. In traditional monitoring, after collecting water level data, the fixed threshold set by historical water level monitoring data is often used to warn the currently collected monitoring data. This fixed threshold setting often ignores the seasonal characteristics of water level data. There may be a situation where the water level data in the drought period is abnormal compared with the same period in history but does not exceed the threshold. Therefore, according to the seasonal characteristics of water level data, a box plot is used to set the threshold for each time period. When setting the threshold, the box plot is easily affected by the distribution characteristics and fluctuation characteristics of the data itself, resulting in inaccurate threshold setting, and then leading to inaccurate water level warning. Summary of the invention

[0005] In order to solve the above technical problems, the purpose of the present invention is to provide an intelligent management and early warning system for water conservancy project monitoring data. The technical solutions adopted are as follows: An embodiment of the present invention provides an intelligent management and early warning system for water conservancy project monitoring data, the system comprising: The data preprocessing module is used to obtain the water level monitoring data of the preset years in history and perform preprocessing to obtain water level data; segment the water level data of each year to obtain water level data of different periods; The distribution regularity calculation module is used to process the water level data of the same period of each year using the box plot to obtain the skewness coefficient and kurtosis coefficient of the period; and obtain the distribution regularity of a period according to the skewness coefficient and kurtosis coefficient of the period; The stability calculation module is used to obtain the fluctuation stability of the period based on the coefficient of variation of the water level data in the same period of each year and the number of abnormal water level data; and to obtain the stability of the period by using the distribution regularity and fluctuation stability of the same period of each year; The module for obtaining the normal range of water level is used to calculate the normal range of water level in the same period of each year by using the upper quantile, lower quantile and interquartile range of water level data in the same period of each year if the stability is greater than the stability threshold; if the stability is less than or equal to the stability threshold, the module for obtaining the normal range of water level in the same period of each year is used to calculate the normal range of water level in the same period; The early warning module is used to warn the water level according to the normal range of the water level in each time period.

[0006] Preferably, the water level data of the same period in each year are processed using a box plot to obtain the skewness coefficient and kurtosis coefficient of the period, including: The skewness coefficient and kurtosis coefficient of the period were obtained by eliminating abnormal water level data in the water level data of the same period of each year using the box plot.

[0007] Preferably, obtaining the distribution regularity of a period according to the skewness coefficient and the kurtosis coefficient of the period includes: The absolute value of the skewness coefficient corresponding to the same period of each year is added to the hyperparameter and inverted to obtain the inverse of the skewness of the period; the absolute value of the difference between the kurtosis coefficient corresponding to the same period of each year and the preset value is obtained and added to the hyperparameter to obtain the addition result, and the addition result is inverted to obtain the inverse of the kurtosis of the period; the water level data of the same period of each year is used to construct a data distribution curve, and the Spearman correlation coefficient between the data distribution curve and the standard normal distribution curve is obtained, and the Spearman correlation coefficient is added to the first preset value and averaged to obtain the normalized Spearman correlation coefficient of the period; the inverse of the skewness, the inverse of the kurtosis and the normalized Spearman correlation coefficient of the period are added and averaged to obtain the distribution regularity of the period.

[0008] Preferably, the fluctuation stability of the period is obtained based on the coefficient of variation of the water level data in the same period of each year and the number of abnormal water level data, including: The inverse of the coefficient of variation of the water level data in the same period of each year is obtained and normalized to obtain the inverse of the coefficient of variation; the number of water level data in the same period of each year is taken as the numerator, and the number of abnormal water level data in the same period of each year and the sum of the hyperparameters are taken as the denominator, the two are compared and normalized to obtain the abnormal proportion; the inverse of the coefficient of variation and the abnormal proportion are added and averaged to obtain the fluctuation stability of the period.

[0009] Preferably, the distribution regularity and fluctuation stability of the same period of each year are used to obtain the stability of the period, including: The weighted sum of the distribution regularity and fluctuation stability of the same period of each year is used to obtain the stability of the period.

[0010] Preferably, if the degree of stability is greater than the stability threshold, the upper quantile, lower quantile and interquartile range of the water level data of the same period of each year are used to calculate the normal range of the water level in the period, including: The abnormal water level data in the water level data of the same period of each year are eliminated by using the box plot, and the upper quantile, lower quantile and interquartile range of the period are obtained by using the remaining water level data in the period; the upper quantile of the period is added to the interquartile range of the preset multiple to obtain the upper limit value; the lower quantile of the period is subtracted from the interquartile range of the preset multiple to obtain the lower limit value; the range between the lower limit value and the upper limit value is the normal range of the water level in the period.

[0011] Preferably, if the degree of stability is less than or equal to the stability threshold, the upper quantile, lower quantile, interquartile range and skewness coefficient of the water level data of the same period of each year are used to calculate the normal range of the water level in the period, including: The upper quantile, lower quantile, interquartile range and skewness coefficient of the water level data in the same period of each year are used to calculate the upper and lower limits of the normal range of water level in that period; the range between the lower and upper limits is the normal range of water level in that period.

[0012] Preferably, the calculation formulas for the upper limit and the lower limit are: , , in, represents the upper limit of the normal range of water level in the ith period, Indicates the lower limit of the normal range of water level in the i-th period; It represents the upper quantile of the water level data in the i-th period of each year after eliminating the abnormal water level data; It represents the lower quantile of the water level data in the i-th period of each year after eliminating the abnormal water level data; represents the interquartile range of the water level data in the i-th period of each year after eliminating the abnormal water level data; e represents the natural constant; It represents the skewness coefficient of the water level data in the i-th period of each year after the abnormal water level data are eliminated through the box plot.

[0013] Preferably, the water level is warned according to the normal range of the water level in each period, including: The normal range of water level corresponding to the current moment is determined according to the time period to which the current moment belongs. If the water level at the current moment is within the normal range of water level corresponding to the current moment, no warning is issued. If the water level at the current moment is not within the normal range of water level corresponding to the current moment, a warning is issued.

[0014] The embodiments of the present invention have at least the following beneficial effects: the present invention obtains water level monitoring data of preset years in history and performs preprocessing to obtain water level data, which can improve the quality of the data and make subsequent analysis more accurate. Furthermore, based on seasonal characteristics, the water level data of each year is divided to facilitate subsequent analysis of the water level data based on seasonality. Then, the distribution regularity of the time period is obtained based on the skewness coefficient and kurtosis coefficient of the same time period in each year, and the fluctuation stability of the time period is obtained based on the coefficient of variation of the water level data of the same time period in each year and the number of abnormal water level data. The fluctuation stability and The stability of the time period is obtained by the regularity of the distribution, and it is judged whether the threshold setting of the water level data for this period using the box plot is accurate. If the stability is greater than the stability threshold, the box plot is directly used to obtain the threshold method. The normal range of the water level for this period is calculated based on the upper quantile, lower quantile and interquartile range of the water level data for the same period of each year. If the stability is less than or equal to the stability threshold, the upper quantile, lower quantile, interquartile range and skewness coefficient of the water level data for the same period of each year are used to calculate the normal range of the water level for this period. A more accurate normal range of water level can be obtained, thereby improving the accuracy of water level warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A system block diagram of an intelligent management and early warning system for water conservancy project monitoring data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of an intelligent management and early warning system for water conservancy project monitoring data proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The following is a detailed description of a specific solution of an intelligent management and early warning system for water conservancy project monitoring data provided by the present invention in conjunction with the accompanying drawings.

[0020] Example: The main application scenarios of the present invention are: in water conservancy projects, when monitoring and warning of water levels, the seasonal characteristics of water level data are ignored, so the set fixed threshold value often cannot accurately determine whether an early warning is needed. Therefore, according to the seasonal characteristics of water level data, a box plot is used to set the threshold for each time period. However, when setting the threshold using a box plot, it is easily affected by the distribution characteristics and fluctuation characteristics of the data itself, resulting in inaccurate threshold setting. Therefore, it is necessary to analyze the characteristics of historical water level data, and then improve the setting of water level thresholds using box plots, so as to obtain more accurate water level thresholds and improve the accuracy of water level early warnings.

[0021] See also Figure 1 , which shows a system block diagram of an intelligent management and early warning system for water conservancy project monitoring data provided by an embodiment of the present invention, the system includes the following modules: The data preprocessing module is used to obtain the water level monitoring data of the preset years in history and perform preprocessing to obtain the water level data; the water level data of each year is segmented to obtain the water level data of different time periods.

[0022] Collect water level monitoring data of preset years in history. Preferably, the preset years in the present invention refer to three years before the current time. The implementer can adjust the preset years according to actual conditions. Thus, water level monitoring data of preset years in history can be obtained. After obtaining the water level monitoring data, it is necessary to pre-process it, specifically, to fill in missing values ​​and reduce noise to obtain water level data.

[0023] When obtaining an accurate water level threshold, seasonal characteristics need to be considered. Changes in water level data are closely related to climate, especially precipitation, evaporation, etc. These climate factors often show seasonal fluctuations. For example, there may be more precipitation in spring and summer, and less precipitation in autumn and winter, which will affect the water level data. Seasonal division of water level data according to months can better reflect changes in water level data. Therefore, the present invention divides the water level data of each year into 12 time periods based on one month as a division period, and obtains water level data in different time periods for subsequent analysis.

[0024] The distribution regularity calculation module is used to process the water level data of the same period of each year using the box plot to obtain the skewness coefficient and kurtosis coefficient of the period; and obtain the distribution regularity of a period based on the skewness coefficient and kurtosis coefficient of the period.

[0025] When using box plots to set data thresholds, the data will be affected by the distribution characteristics and fluctuation characteristics of the water level data, resulting in inaccurate threshold setting results. If the water level data distribution presents a relatively standard normal distribution, then using a box plot to set the threshold for the water level data in that period will be more accurate. Therefore, it is necessary to calculate the distribution regularity of the water level data in each divided period, and use the box plot to set the water level data threshold based on the distribution regularity of the water level data in that period.

[0026] Furthermore, the water level data in the same period of each year are processed using a box plot to obtain the skewness coefficient and the kurtosis coefficient; the distribution regularity of a period is obtained based on the skewness coefficient and the kurtosis coefficient of the period.

[0027] Specifically, the water level data in the same time period of each year are processed using a box plot to obtain the skewness coefficient and the kurtosis coefficient. It should be noted that when obtaining the skewness coefficient and the kurtosis coefficient, the box plot is used to eliminate the abnormal water level data. The water level data in the same time period of each year in this application is the water level data in the same time period in three years. For example, January is the data in January of each year in three years.

[0028] The distribution regularity of a time period is obtained according to the skewness coefficient and the kurtosis coefficient, including: adding the absolute value of the skewness coefficient corresponding to the same time period of each year to the hyperparameter and inverting it to obtain the inverse of the skewness of the time period; obtaining the absolute value of the difference between the kurtosis coefficient corresponding to the same time period of each year and the preset value and adding it to the hyperparameter to obtain the addition result, inverting the addition result to obtain the inverse of the kurtosis of the time period; using the water level data of the same time period of each year to construct a data distribution curve, obtaining the Spearman correlation coefficient between the data distribution curve and the standard normal distribution curve, adding the Spearman correlation coefficient to the first preset value and averaging them to obtain the normalized Spearman correlation coefficient of the time period; adding the inverse of the skewness, the inverse of the kurtosis and the normalized Spearman correlation coefficient of the time period and averaging them to obtain the distribution regularity of the time period.

[0029] For the i-th period of each year, the calculation model of the distribution regularity of water level data in this period is: , in, It represents the distribution regularity of the i-th period of each year, or the distribution regularity of the water level data in the i-th divided period. Since the judgment of normal distribution is greatly affected by abnormal values, is the skewness coefficient of the water level data in the ith period of each year after the abnormal water level data are eliminated through the box plot. The closer this value is to 0, that is, the larger the first term of the above formula is, the closer the water level data in the ith period of each year is to the normal distribution, that is, the stronger the distribution regularity of the water level data in the ith divided period is. Represents the inverse of skewness; a is a hyperparameter to prevent the denominator from being 0; It represents the kurtosis coefficient of the water level data in the i-th period of each year after the outliers are removed by the box plot. The preset value is 3. The reason for setting the preset value to 3 is that the kurtosis coefficient of the standard normal distribution is 3. Therefore, the closer the kurtosis coefficient is to 3, that is, the larger the second term of the above formula is, the closer the water level data is to the normal distribution, that is, the stronger the distribution regularity of the water level data in the i-th divided period is. The skewness coefficient measures the symmetry of the data distribution, and the kurtosis coefficient measures the peak state of the data distribution. Both are existing technologies and are not elaborated in this scheme. is the inverse of skewness; represents the Spearman correlation coefficient between the data distribution curve constructed using the data in the i-th period of each year and the standard normal distribution curve, where The value of is between -1 and 1. The purpose is to normalize and obtain the normalized Spearman correlation coefficient. The larger the value, the closer the water level data distribution in the i-th period of each year is to the normal distribution, that is, the stronger the distribution regularity of the water level data in the i-th divided period. In this way, the distribution regularity of each period can be obtained.

[0030] The stability calculation module is used to obtain the fluctuation stability of the period based on the coefficient of variation of the water level data in the same period of each year and the number of abnormal water level data; and to obtain the stability of the period by using the distribution regularity and fluctuation stability of the same period of each year.

[0031] Because the use of box plots for threshold setting will be affected by the volatility of water level data, if the water level data fluctuates greatly, the box plot may treat normal high-volatility data values ​​as abnormal water level data, thereby affecting the threshold setting. Therefore, it is necessary to calculate the distribution regularity and fluctuation stability of the water level data in each divided time period, and comprehensively obtain the stability of the water level data in each time period, and then conduct subsequent analysis based on the stability.

[0032] Furthermore, based on the coefficient of variation of the water level data in the same period of each year and the number of abnormal water level data, the fluctuation stability of the period is obtained. Specifically, the inverse of the coefficient of variation of the water level data in the same period of each year is obtained and normalized to obtain the inverse of the coefficient of variation; the number of water level data in the same period of each year is taken as the numerator, and the number of abnormal water level data in the same period of each year and the sum of the hyperparameters are taken as the denominator, the two are compared and normalized to obtain the abnormal proportion; the inverse of the coefficient of variation and the abnormal proportion are added and averaged to obtain the fluctuation stability of the period.

[0033] The specific calculation model of fluctuation stability is: , in, is the fluctuation stability of water level data in the i-th period of each year; is the total number of all water level data in the i-th period (the same period) of each year, is the number of all abnormal water level data in the i-th period of each year, which is the abnormal water level data obtained after screening based on the box plot. It represents the ratio of the total number of water level data in the i-th period of each year to the number of abnormal water level data in the period. The larger the value, the smaller the number of abnormal data in the i-th period of each year. is the coefficient of variation of the water level data in the i-th period of each year. The smaller the value, that is, the larger the second term in the above formula, indicates that the volatility of the water level data in the i-th period of each year is smaller, that is, the more stable the fluctuation of the water level data in this period is; norm( ) is the normalization function. and They represent the abnormal proportion and the inverse of the coefficient of variation respectively.

[0034] Thus, the distribution regularity and fluctuation stability of the same period of each year can be obtained, and then the stability of the period can be obtained by combining them. Specifically, the distribution regularity and fluctuation stability of the same period of each year are weighted and summed to obtain the stability of the period. The specific calculation formula is: , in, and are weight coefficients respectively. Since the distribution regularity of water level data in a period has a greater impact on the box plot threshold setting than the fluctuation stability, this scheme sets , , implementers can make adjustments based on specific circumstances; Indicates the stability of the water level data in the i-th period, represents the distribution regularity of water level data in the i-th period, Indicates the fluctuation stability of the water level data in the i-th period. From this, we can get the stability of each of the 12 periods.

[0035] The normal water level range acquisition module is used to calculate the normal water level range of the period using the upper quantile, lower quantile and interquartile range of the water level data of the same period of each year if the stability degree is greater than the stability threshold; if the stability degree is less than or equal to the stability threshold, the upper quantile, lower quantile, interquartile range and skewness coefficient of the water level data of the same period of each year are used to calculate the normal water level range of the period.

[0036] If the water level data within a period of time is more stable, it means that the water level data within the period of time is more accurate in screening outliers after the box plot threshold is set. At this time, the original threshold setting method can be used. If the water level data within a period of time is less stable, it means that the water level data within the period of time is less accurate in screening outliers after the original box plot threshold is set. At this time, the box plot threshold setting needs to be adaptively adjusted according to the data characteristics within the period.

[0037] Set a stability threshold. Preferably, the value of the stability threshold in the embodiment of the present invention is 0.7. The implementer can adjust it according to the actual situation. If the stability level of a time period is greater than the stability threshold, the threshold setting method will not be changed when the box plot is used to set the threshold of the water level data in the time period. It should be noted that due to the influence of abnormal values ​​that may exist in the historical water level data on the threshold setting, if the upper and lower thresholds are adjusted directly, the warning results may be inaccurate. Therefore, it is necessary to use the box plot to eliminate the abnormal values ​​before calculation.

[0038] If the degree of stability is greater than the stability threshold, the upper quantile, lower quantile and interquartile range of the water level data in the same period of each year are used to calculate the normal range of the water level in that period. Specifically, the abnormal water level data in the water level data in the same period of each year are eliminated using a box plot, and the upper quantile, lower quantile and interquartile range of that period are obtained using the remaining water level data in that period; the upper quantile of that period is added to the interquartile range of a preset multiple to obtain the upper limit value; the lower quantile of that period is subtracted from the interquartile range of a preset multiple to obtain the lower limit value; the range between the lower limit value and the upper limit value is the normal range of the water level in that period.

[0039] Among them, the specific calculation formula for the upper limit and lower limit is: and , is the upper quantile of the water level data in the i-th period of each year after removing the abnormal water level data, is the lower quantile of the water level data in the i-th period of each year after removing the abnormal water level data, It is the interquartile range of the water level data in the i-th period of each year after eliminating the abnormal water level data.

[0040] If the degree of stability is less than or equal to the stability threshold, the upper quantile, lower quantile, interquartile range and skewness coefficient of the water level data of the same period of each year are used to calculate the normal range of the water level in the period. Specifically, the upper quantile, lower quantile, interquartile range and skewness coefficient of the water level data of the same period of each year are used to calculate the upper limit and lower limit of the normal range of the water level in the period. The range between the lower limit and the upper limit is the normal range of the water level in the period. The calculation formula of the upper limit is: , The calculation formula of the lower limit is: , in, represents the upper limit of the normal range of water level in the ith period, Indicates the lower limit of the normal range of water level in the i-th period; It represents the upper quantile of the water level data in the i-th period of each year after eliminating the abnormal water level data; It represents the lower quantile of the water level data in the i-th period of each year after eliminating the abnormal water level data; represents the interquartile range of the water level data in the i-th period of each year after eliminating the abnormal water level data; e represents the natural constant; It represents the skewness coefficient of the water level data in the i-th period of each year after the abnormal water level data are eliminated through the box plot.

[0041] On x>0, it is an increasing function with a range of [0,1]; in the calculation of the upper limit, This indicates that the water level data value in the i-th period is too high. For example, if the water level rises due to heavy rainfall in the rainy season, the normal data value may be considered to be greater than the threshold when judging the current data threshold, and the normal data is classified as abnormal. Therefore, the upper limit of the normal range of the water level needs to be increased; in the calculation of the lower limit, the skewness coefficient When , it means that there are more data points that are less than the mean value of the water level data, indicating that the water level data value in the i-th period is low. For example, when the water level drops in the dry season, when judging the current data threshold, it may be considered that the normal data value is less than the threshold, and the normal water level data is classified as abnormal. At this time, the lower limit of the normal range of the water level needs to be lowered.

[0042] This achieves adaptive threshold adjustment of the water level data for each divided time period, and obtains the normal water level range of the water level data for each time period.

[0043] The early warning module is used to warn the water level according to the normal range of the water level in each time period.

[0044] After obtaining the normal range of water level for each time period, the specific operation to determine whether the water level at the current moment is normal is as follows: determine the normal range of water level corresponding to the current moment according to the time period to which the current moment belongs; if the water level at the current moment is within the normal range of water level corresponding to the current moment, no warning is issued; if the water level at the current moment is not within the normal range of water level corresponding to the current moment, a warning is issued. In addition, the normal range of water level corresponding to each month of each year can be updated in real time based on previous historical data to ensure the accuracy of the normal range of water level and the accuracy of warning.

[0045] To summarize, this scheme divides the historical water level data into seasonal categories, obtains the stability of the water level data in each time period according to the distribution characteristics and fluctuation characteristics of the historical water level data in each divided time period, and then adaptively sets the water level threshold according to the box plot according to the stability of each time period of the historical data, obtains the threshold setting for each time period, and then issues an early warning for the water level data in the current time period by judging whether the current water level data exceeds the set threshold for the time period. This method of obtaining the threshold setting for water level data in each time period based on the seasonal characteristics of historical water level data largely avoids the situation where a single threshold setting is inaccurate for early warning of water level data, and adjusts the threshold setting of the box plot according to the characteristics of the water level data in each time period, so that the box plot threshold setting is more accurate, thereby making the early warning of water conservancy project monitoring data more accurate and effective.

[0046] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0047] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0048] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent management and early warning system for water conservancy project monitoring data, characterized in that: The system includes: The data preprocessing module is used to obtain the water level monitoring data of the preset years in history and perform preprocessing to obtain water level data; segment the water level data of each year to obtain water level data of different periods; The distribution regularity calculation module is used to process the water level data of the same period of each year using the box plot to obtain the skewness coefficient and kurtosis coefficient of the period; and obtain the distribution regularity of a period according to the skewness coefficient and kurtosis coefficient of the period; The stability calculation module is used to obtain the fluctuation stability of the period based on the coefficient of variation of the water level data in the same period of each year and the number of abnormal water level data; and to obtain the stability of the period by using the distribution regularity and fluctuation stability of the same period of each year; The module for obtaining the normal range of water level is used to calculate the normal range of water level in the same period of each year by using the upper quantile, lower quantile and interquartile range of water level data in the same period of each year if the stability is greater than the stability threshold; if the stability is less than or equal to the stability threshold, the module for obtaining the normal range of water level in the same period of each year is used to calculate the normal range of water level in the same period; The early warning module is used to warn the water level according to the normal range of the water level in each time period.

2. The intelligent management and early warning system for water conservancy project monitoring data according to claim 1 is characterized in that: The box plot is used to process the water level data of the same period of each year to obtain the skewness coefficient and kurtosis coefficient of the period, including: The skewness coefficient and kurtosis coefficient of the period were obtained by eliminating abnormal water level data in the water level data of the same period of each year using the box plot.

3. The intelligent management and early warning system for water conservancy project monitoring data according to claim 1 is characterized in that: The step of obtaining the distribution regularity of a time period according to the skewness coefficient and the kurtosis coefficient of the time period includes: The absolute value of the skewness coefficient corresponding to the same period of each year is added to the hyperparameter and inverted to obtain the inverse of the skewness of the period; the absolute value of the difference between the kurtosis coefficient corresponding to the same period of each year and the preset value is obtained and added to the hyperparameter to obtain the addition result, and the addition result is inverted to obtain the inverse of the kurtosis of the period; the water level data of the same period of each year is used to construct a data distribution curve, and the Spearman correlation coefficient between the data distribution curve and the standard normal distribution curve is obtained, and the Spearman correlation coefficient is added to the first preset value and averaged to obtain the normalized Spearman correlation coefficient of the period; the inverse of the skewness, the inverse of the kurtosis and the normalized Spearman correlation coefficient of the period are added and averaged to obtain the distribution regularity of the period.

4. The intelligent management and early warning system for water conservancy project monitoring data according to claim 1 is characterized in that: The fluctuation stability of the period is obtained based on the coefficient of variation of the water level data in the same period of each year and the number of abnormal water level data, including: The inverse of the coefficient of variation of the water level data in the same period of each year is obtained and normalized to obtain the inverse of the coefficient of variation; the number of water level data in the same period of each year is taken as the numerator, and the number of abnormal water level data in the same period of each year and the sum of the hyperparameters are taken as the denominator, the two are compared and normalized to obtain the abnormal proportion; the inverse of the coefficient of variation and the abnormal proportion are added and averaged to obtain the fluctuation stability of the period.

5. The intelligent management and early warning system for water conservancy project monitoring data according to claim 1 is characterized in that: The method of using the distribution regularity and fluctuation stability of the same period of each year to obtain the stability of the period includes: The weighted sum of the distribution regularity and fluctuation stability of the same period of each year is used to obtain the stability of the period.

6. The intelligent management and early warning system for water conservancy project monitoring data according to claim 1 is characterized in that: If the stability level is greater than the stability threshold, the upper quantile, lower quantile and interquartile range of the water level data for the same period of each year are used to calculate the normal range of the water level for that period, including: The abnormal water level data in the water level data of the same period of each year are eliminated by using the box plot, and the upper quantile, lower quantile and interquartile range of the period are obtained by using the remaining water level data in the period; the upper quantile of the period is added to the interquartile range of the preset multiple to obtain the upper limit value; the lower quantile of the period is subtracted from the interquartile range of the preset multiple to obtain the lower limit value; the range between the lower limit value and the upper limit value is the normal range of the water level in the period.

7. The intelligent management and early warning system for water conservancy project monitoring data according to claim 1 is characterized in that: If the stability level is less than or equal to the stability threshold, the upper quantile, lower quantile, interquartile range and skewness coefficient of the water level data of the same period of each year are used to calculate the normal range of the water level in that period, including: The upper quantile, lower quantile, interquartile range and skewness coefficient of the water level data in the same period of each year are used to calculate the upper and lower limits of the normal range of water level in that period; the range between the lower and upper limits is the normal range of water level in that period.

8. The intelligent management and early warning system for water conservancy project monitoring data according to claim 7 is characterized in that: The calculation formulas for the upper limit and the lower limit are: ， ， in, represents the upper limit of the normal range of water level in the ith period, Indicates the lower limit of the normal range of water level in the i-th period; It represents the upper quantile of the water level data in the i-th period of each year after eliminating the abnormal water level data; It represents the lower quantile of the water level data in the i-th period of each year after eliminating the abnormal water level data; represents the interquartile range of the water level data in the i-th period of each year after eliminating the abnormal water level data; e represents the natural constant; It represents the skewness coefficient of the water level data in the i-th period of each year after the abnormal water level data are eliminated through the box plot.

9. The intelligent management and early warning system for water conservancy project monitoring data according to claim 1 is characterized in that: The water level early warning according to the normal range of water level in each period includes: The normal range of water level corresponding to the current moment is determined according to the time period to which the current moment belongs. If the water level at the current moment is within the normal range of water level corresponding to the current moment, no warning is issued. If the water level at the current moment is not within the normal range of water level corresponding to the current moment, a warning is issued.

Images