A flood disaster early warning method and system based on hydrogeological data

Abstract

The present invention relates to the field of flood disaster warning, and specifically to a flood disaster warning method and system based on hydrogeological data. The method first obtains the monitoring data segments of different types of monitoring parameters and runoff parameters in the measured basin during each flood disaster process, divides the monitoring data segments of the runoff parameters during the flood disaster process based on the warning threshold of the runoff parameters, obtains the data sub-segments of the monitoring parameters at each disaster level during the flood disaster process, obtains the response sensitivity of the monitoring parameters during the flood disaster process according to the data distribution of the data sub-segments at each disaster level, obtains the disaster impact weight of the monitoring parameters according to the correlation between the monitoring parameters during each flood disaster process and the difference in response sensitivity, and then warns the flood disaster in the measured basin. The present invention improves the accuracy of flood disaster warning by comprehensively analyzing the various hydrogeological data collected.

Description

Technical Field

[0001] The present invention relates to the field of flood disaster early warning, and in particular to a flood disaster early warning method and system based on hydrogeological data. Background Art

[0002] Hydrogeological data can provide basic data support for flood disaster warning. By collecting and analyzing hydrogeological data in specific areas, the possibility of flood disasters can be accurately predicted, and early warnings can be issued to reduce the losses caused by flood disasters.

[0003] In the related technology, various types of hydrogeological data collected are usually analyzed in an indiscriminate and comprehensive manner to achieve early warning of flood disasters. However, since there are many types of hydrogeological data related to flood disasters, there are large differences in the degree of mutual connection between different types of hydrogeological data and flood disasters, which makes it impossible for existing methods to accurately predict flood disasters, resulting in low accuracy of flood disaster early warnings. Summary of the invention

[0004] In order to solve the technical problem that the existing methods cannot accurately predict flood disasters, resulting in low accuracy of flood disaster warning, the purpose of the present invention is to provide a flood disaster warning method and system based on hydrogeological data. The technical solution adopted is as follows:

[0005] The present invention proposes a flood disaster early warning method based on hydrogeological data, the method comprising:

[0006] Acquire monitoring data segments of different types of monitoring parameters and runoff parameters in the measured basin during each flood disaster process, and acquire early warning thresholds of different disaster levels of the runoff parameters, wherein the monitoring data segments include monitoring data at different times;

[0007] Taking any flood disaster process as the target disaster process and any monitoring parameter as the target monitoring parameter, the monitoring data segments of the runoff parameter in the target disaster process are divided based on different early warning thresholds of the runoff parameter, and multiple data sub-segments of the target monitoring parameter in each disaster level in the target disaster process are obtained; according to the distribution of monitoring data of the target monitoring parameter in the data sub-segments at different disaster levels in the target disaster process, and the value of each disaster level, the response sensitivity of the target monitoring parameter in the target disaster process is obtained;

[0008] According to the correlation between the target monitoring parameter and other monitoring parameters except the target monitoring parameter, the correlation between the target monitoring parameter and the runoff parameter, and the difference in the response sensitivity between the target monitoring parameter and the other monitoring parameters in the target disaster process, the influence level of the target monitoring parameter in the target disaster process is obtained; according to the difference in the influence level between the target monitoring parameter in the target disaster process and other flood disaster processes except the target disaster process, and the influence level of the target monitoring parameter in each flood disaster process, the disaster impact weight of the target monitoring parameter is obtained;

[0009] According to the real-time monitoring data of each monitoring parameter and the disaster impact weight, an early warning is issued for the flood disaster in the monitored river basin.

[0010] Furthermore, the multiple data subsegments of the target monitoring parameters at each disaster level in the process of obtaining the target disaster include:

[0011] Based on different early warning thresholds of runoff parameters, the monitoring data segments of runoff parameters in the target disaster process are divided to obtain multiple data sub-segments of runoff parameters in each disaster level in the target disaster process;

[0012] During the target disaster process, the time periods corresponding to the data sub-segments of the runoff parameters at each disaster level are used as the disaster time periods of the target disaster process at each disaster level, and the monitoring data segments corresponding to the disaster time periods of the target monitoring parameters at each disaster level are used as the multiple data sub-segments of the target monitoring parameters at each disaster level during the target disaster process.

[0013] Furthermore, obtaining the response sensitivity of the target monitoring parameter in the target disaster process includes:

[0014] Analyze the discrete degree of the monitoring data of each data sub-segment of the target monitoring parameter in the target disaster process at each disaster level, and obtain the data fluctuation level of the target monitoring parameter in each disaster level during the target disaster process;

[0015] After negative correlation mapping and normalization of the values ​​of each disaster level, the reference weight of each disaster level is obtained;

[0016] By using the reference weight of each disaster level, the data fluctuation level of the target monitoring parameter at each disaster level in the target disaster process is weightedly summed to obtain the response sensitivity of the target monitoring parameter in the target disaster process.

[0017] Furthermore, the data fluctuation level of the target monitoring parameter at each disaster level in the process of obtaining the target disaster includes:

[0018] The range of the monitoring data of each data sub-segment of the target monitoring parameter at each disaster level during the target disaster process is taken as the fluctuation degree of each data sub-segment of the target monitoring parameter at each disaster level during the target disaster process;

[0019] The average value of the fluctuation degree of all data subsegments of the target monitoring parameter at each disaster level during the target disaster process is taken as the data fluctuation level of the target monitoring parameter at each disaster level during the target disaster process.

[0020] Furthermore, obtaining the impact level of the target monitoring parameter on the target disaster process includes:

[0021] The absolute value of the Pearson correlation coefficient between the target monitoring parameter and each other monitoring parameter during the target disaster process is used as the first correlation coefficient between the target monitoring parameter and each other monitoring parameter during the target disaster process;

[0022] Obtaining the impact priority of the target monitoring parameter in the target disaster process according to the difference in the response sensitivity between the target monitoring parameter and each other monitoring parameter in the target disaster process, and the first correlation coefficient;

[0023] The absolute value of the Pearson correlation coefficient between the target monitoring parameter and the runoff parameter during the target disaster process is used as the second correlation coefficient between the target monitoring parameter and the runoff parameter during the target disaster process;

[0024] The impact priority and the second correlation coefficient are integrated to obtain the impact level of the target monitoring parameter on the target disaster process.

[0025] Furthermore, obtaining the impact priority of the target monitoring parameter on the target disaster process includes:

[0026] Performing negative correlation mapping on the absolute value of the difference between the target monitoring parameter and each other monitoring parameter in the response sensitivity of the target disaster process, and obtaining the sensitivity similarity between the target monitoring parameter and each other monitoring parameter in the target disaster process;

[0027] Combining the sensitive similarity and the first correlation coefficient between the target monitoring parameter in the target disaster process and each other monitoring parameter to obtain a connection parameter between the target monitoring parameter in the target disaster process and each other monitoring parameter;

[0028] The average value of the connection parameter between the target monitoring parameter and all other monitoring parameters in the target disaster process is used as the impact priority of the target monitoring parameter in the target disaster process.

[0029] Furthermore, the disaster impact weights of the target monitoring parameters are obtained including:

[0030] The absolute value of the difference between the impact level of the target monitoring parameter between the target disaster process and each other flood disaster process is taken as the impact level difference value between the target monitoring parameter and each other flood disaster process;

[0031] Normalizing the accumulated value of the impact level difference value of the target monitoring parameter between the target disaster process and all other flood disaster processes to obtain the impact confidence of the target monitoring parameter in the target disaster process;

[0032] By using the impact confidence of the target monitoring parameter in each flood disaster process, the impact level of the target monitoring parameter in each flood disaster process is weighted and summed and normalized to obtain the disaster impact weight of the target monitoring parameter.

[0033] Furthermore, the early warning of flood disasters in the measured basin includes:

[0034] The difference between the real-time monitoring data of each monitoring parameter and the minimum value of the warning threshold of each monitoring parameter is used as the real-time initial judgment value of each monitoring parameter;

[0035] Using the disaster impact weight of each monitoring parameter, weighted summation is performed on the initial judgment value to obtain a real-time comprehensive judgment value of the watershed to be measured;

[0036] If the real-time comprehensive judgment value of the watershed to be tested is not less than 0, use The function maps the comprehensive judgment value to obtain the real-time disaster possibility of the watershed to be tested; if the real-time comprehensive judgment value of the watershed to be tested is less than the value 0, the real-time disaster possibility of the watershed to be tested is set to the value 0;

[0037] Based on the real-time disaster possibility of the tested basin, early warning of flood disasters in the tested basin is issued.

[0038] Furthermore, the early warning of flood disasters in the tested basin based on the real-time disaster possibility of the tested basin includes:

[0039] If the disaster probability of the tested basin at each moment in the preset time period is greater than the preset probability threshold, a flood disaster warning will be issued.

[0040] The present invention also proposes a flood disaster warning system based on hydrogeological data, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor implementing any one of the steps of a flood disaster warning method based on hydrogeological data when executing the computer program.

[0041] The present invention has the following beneficial effects:

[0042] The present invention takes into account that the existing methods cannot accurately predict flood disasters, resulting in low accuracy in flood disaster warnings. Therefore, the present invention first obtains the monitoring data segments of different types of monitoring parameters and runoff parameters in the measured basin during each flood disaster process, as well as the runoff parameters and warning thresholds of different disaster levels for each monitoring parameter. Since the runoff parameter is directly related to the occurrence of flood disasters, and flood disasters are generally divided into multiple levels, and there are large differences in the data fluctuation performance of the target monitoring parameters at different levels, the monitoring data segments of the runoff parameter in the target disaster process are first divided to obtain multiple data sub-segments of the target monitoring parameter at each disaster level in the target disaster process, and then the response sensitivity obtained is used to obtain the warning thresholds. It reflects the sensitivity of the target monitoring parameters to the data changes in the target disaster process, and provides a data basis for the subsequent accurate calculation of the impact weights of each monitoring parameter on flood disasters. Taking into account the different connections between different monitoring parameters and the target disaster process, the obtained impact level reflects the impact of the target monitoring parameters on the target disaster process. Taking into account that there are certain differences in the impact levels of the target monitoring parameters on different flood disaster processes, the obtained disaster impact weights are used to further reflect the importance of the target monitoring parameters in reflecting flood disasters. Then, the disaster impact weights of each monitoring parameter, as well as the real-time monitoring data and the warning threshold of the initial disaster level of the monitoring parameters are combined to accurately warn of flood disasters in the measured basin. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] 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.

[0044] Figure 1 A flow chart of a flood disaster early warning method based on hydrogeological data provided by an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of multiple data subsegments of a target monitoring parameter at an initial disaster level during a target disaster process provided by an embodiment of the present invention;

[0046] Figure 3 A flow chart of a method for obtaining the response sensitivity of a target monitoring parameter in a target disaster process provided by an embodiment of the present invention;

[0047] Figure 4A flow chart of a method for obtaining the impact level of a target monitoring parameter on a target disaster process provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] 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 a flood disaster early warning method and system based on hydrogeological data proposed by the present invention, its specific implementation method, structure, features and effects, in conjunction 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.

[0049] 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.

[0050] The specific scheme of a flood disaster early warning method and system based on hydrogeological data provided by the present invention is described in detail below with reference to the accompanying drawings.

[0051] See also Figure 1 , which shows a flow chart of a flood disaster early warning method based on hydrogeological data provided by an embodiment of the present invention, the method comprising:

[0052] Step S1: Obtain monitoring data segments of different types of monitoring parameters and runoff parameters in the measured basin during each flood disaster process, and obtain warning thresholds of different disaster levels of runoff parameters. The monitoring data segments include monitoring data at different times.

[0053] Hydrogeological data can provide basic data support for flood disaster warning. By collecting and analyzing hydrogeological data in specific areas, the possibility of flood disasters can be accurately predicted, and early warnings can be issued to reduce the losses caused by flood disasters. The occurrence of flood disasters is affected by many factors, such as runoff, precipitation, groundwater level and soil moisture, etc. Various types of data are directly or indirectly related to the occurrence of flood disasters. For example, changes in runoff data can directly reflect the situation of flood disasters, and its connection with flood disasters is relatively strong. Compared with runoff data, other types of data are less related to flood disasters, but there is also a certain connection.

[0054] At the same time, considering that there are certain differences in the data performance of the runoff parameter or a specific monitoring parameter in different flood disaster processes, it is necessary to analyze the data performance of the runoff parameter or a specific monitoring parameter in multiple flood disaster processes in the future to avoid the impact of randomness on flood disaster warning. Therefore, the embodiment of the present invention first installs various types of sensors in the watershed to be tested, and uses the sensors to collect the monitoring data segments of the runoff parameter and various types of monitoring parameters in each flood disaster process. The monitoring data segments include monitoring data at different times, wherein the time intervals for data collection of various sensors are the same. In one embodiment of the present invention, the time interval for data collection is set to 30 minutes, that is, monitoring data is collected every 30 minutes. The time interval for data collection can also be set by the implementer according to the specific implementation scenario, which is not limited here.

[0055] In actual situations, flood disasters are usually divided into four different disaster levels, namely the first disaster level, the second disaster level, the third disaster level and the fourth disaster level, among which the first disaster level has the highest severity and is the highest disaster level, and the fourth disaster level has the lowest severity and is the initial disaster level. Different disaster levels correspond to different warning thresholds for runoff parameters and each monitoring parameter. Since the runoff parameter has the strongest correlation with flood disasters, the disaster levels are generally divided based on different warning thresholds for the runoff parameters. For example, for the runoff parameter, there are three warning thresholds of different sizes. When the monitoring data of the runoff parameter does not exceed the minimum value of the warning threshold, it indicates that the flood is at the fourth disaster level. When it exceeds the minimum value of the warning threshold and does not exceed the second minimum value of the warning threshold, it indicates that the flood is at the third disaster level, and so on. Therefore, the embodiment of the present invention also needs to obtain the warning thresholds for different disaster levels of the runoff parameter, wherein the warning threshold of the runoff parameter is a known value specified in the local watershed.

[0056] Step S2: taking any flood disaster process as the target disaster process and any monitoring parameter as the target monitoring parameter, dividing the monitoring data segments of the runoff parameter in the target disaster process based on different warning thresholds of the runoff parameter, and obtaining multiple data sub-segments of the target monitoring parameter at each disaster level in the target disaster process; according to the distribution of monitoring data of the target monitoring parameter in the data sub-segments at different disaster levels in the target disaster process and the value of each disaster level, obtaining the response sensitivity of the target monitoring parameter in the target disaster process.

[0057] Since the data performance of runoff parameters and various monitoring parameters are different in different flood disaster processes, it is necessary to analyze any flood disaster process. First, any flood disaster process is taken as the target disaster process. At the same time, different types of monitoring parameters have different data performances in a specific flood disaster process, and the correlation between different types of monitoring parameters and flood disasters is also different. Therefore, it is necessary to analyze any monitoring parameter and take any monitoring parameter as the target monitoring parameter.

[0058] Since the runoff parameters are directly related to the occurrence of flood disasters, and flood disasters are divided into multiple disaster levels, there are large differences in the data fluctuation performance of the target monitoring parameters in different disaster levels. Therefore, the embodiment of the present invention first divides the monitoring data segments of the runoff parameters in the target disaster process based on different warning thresholds of the runoff parameters, and differentiates the monitoring data of the runoff parameters in the target disaster process into different disaster levels, and obtains multiple data sub-segments of the target monitoring parameters in each disaster level during the target disaster process. Subsequently, the data sub-segments of the target monitoring parameters in each disaster level during the target disaster process can be analyzed, and the sensitivity of the target monitoring parameters to the occurrence of the target disaster can be accurately calculated and analyzed.

[0059] Preferably, in one embodiment of the present invention, the method for acquiring multiple data subsegments of target monitoring parameters at each disaster level during a target disaster process specifically includes:

[0060] Based on different early warning thresholds of runoff parameters, the monitoring data segments of runoff parameters in the target disaster process are divided to obtain multiple data sub-segments of runoff parameters in the target disaster process at each disaster level; in the target disaster process, the time periods corresponding to each data sub-segment of the runoff parameters at each disaster level are used as the disaster time periods of the target disaster process at each disaster level, and the monitoring data segments corresponding to each disaster time period of the target monitoring parameters at each disaster level are used as the multiple data sub-segments of the target monitoring parameters in the target disaster process at each disaster level. Figure 2 , which shows a schematic diagram of multiple data sub-segments of a target monitoring parameter at an initial disaster level during a target disaster process provided by an embodiment of the present invention, wherein A represents a monitoring data segment of a runoff parameter during a target disaster process, B represents a monitoring data segment of a target monitoring parameter during a target disaster process, S1, S2 and S3 respectively represent three warning thresholds of the runoff parameter, curve segments a1~a2, a3~a4 and a5~a6 represent data sub-segments of the runoff parameter during a target disaster process at an initial disaster level, time periods t1~t2, t3~t4 and t5~t6 represent disaster time periods of a target disaster process at an initial disaster level, and curve segments b1~b2, b3~b4 and b5~b6 represent data sub-segments of the target monitoring parameter during a target disaster process at an initial disaster level.

[0061] For the target monitoring parameters in the target disaster process, the fluctuation of the monitoring data in the data subsegments at different disaster levels is different, and the severity of the flood disaster represented by different disaster levels is different. Therefore, the distribution of the monitoring data of the target monitoring parameters in the data subsegments at the same disaster level during the target disaster process can be analyzed, and combined with the numerical values ​​of each disaster level, the obtained response sensitivity can be used to reflect the sensitivity of the target monitoring parameters to the data changes during the target disaster process. The greater the response sensitivity, the more sensitive the target monitoring parameters are to the occurrence of flood disasters during the target disaster process, which provides a data basis for the subsequent accurate calculation of the impact weights of each monitoring parameter on flood disasters.

[0062] Preferably, in one embodiment of the present invention, a method for acquiring the response sensitivity of a target monitoring parameter in a target disaster process specifically includes:

[0063] See also Figure 3 , which shows a flow chart of a method for obtaining the response sensitivity of a target monitoring parameter in a target disaster process provided by an embodiment of the present invention.

[0064] Step S201: Analyze the discrete degree of the monitoring data of each data sub-segment of the target monitoring parameter in the target disaster process at each disaster level to obtain the data fluctuation level of the target monitoring parameter in the target disaster process at each disaster level.

[0065] The more obvious the data fluctuation characteristics of the target monitoring parameters in the data sub-segment of each disaster level during the target disaster process, the more sensitive the target monitoring parameters are to the data response of the flood disaster at that disaster level. Therefore, the discrete degree of the monitoring data of each data sub-segment of the target monitoring parameters in the target disaster process at each disaster level can be analyzed to obtain the data fluctuation level of the target monitoring parameters in the target disaster process at each disaster level, providing a data basis for the subsequent calculation of the response sensitivity of the target monitoring parameters in the target disaster process.

[0066] Preferably, in one embodiment of the present invention, the method for obtaining the data fluctuation level of the target monitoring parameter at each disaster level during the target disaster process specifically includes:

[0067] The range of the monitoring data of each data subsegment of the target monitoring parameter at each disaster level during the target disaster process is taken as the fluctuation degree of each data subsegment of the target monitoring parameter at each disaster level during the target disaster process. The greater the fluctuation degree, the more obvious the data fluctuation characteristics of the data subsegment. Furthermore, the average value of the fluctuation degrees of all data subsegments of the target monitoring parameter at each disaster level during the target disaster process can be taken as the data fluctuation level of the target monitoring parameter at each disaster level during the target disaster process.

[0068] As an example, in one embodiment of the present invention, the expression of the data fluctuation level of the target monitoring parameter at each disaster level during the target disaster process may be specifically, for example, as follows:

[0069]

[0070] in, Indicates the target monitoring parameters in the target disaster process. The level of data fluctuation for each disaster level; and They represent the target monitoring parameters in the target disaster process in the first The disaster level The maximum and minimum values ​​of the monitoring data of each data sub-segment; Indicates the target monitoring parameters in the target disaster process. The number of data subsegments for each disaster level; Indicates the target monitoring parameters in the target disaster process. The disaster level The degree of fluctuation of a data sub-segment.

[0071] Step S202: negatively correlate the values ​​of each disaster level and perform normalization processing to obtain a reference weight for each disaster level.

[0072] The embodiment of the present invention divides the disaster level into four levels, and uses values ​​1, 2, 3 and 4 to represent the values ​​of different disaster levels respectively. The smaller the value of the disaster level, the more serious the disaster. Therefore, in order to make the target monitoring parameters more sensitive to more serious flood disasters and improve the sensitivity of severe flood disaster prediction, the values ​​of each disaster level are negatively correlated and normalized to obtain the reference weight of each disaster level, which is convenient for subsequent improvement of the accuracy of response sensitivity calculation and analysis.

[0073] In one embodiment of the present invention, the reciprocal of the numerical value of each disaster level can be used as the severity index value of each disaster level to achieve negative correlation mapping of the numerical value of each disaster level. In other embodiments of the present invention, negative correlation mapping can also be achieved based on other basic mathematical operations, which are not limited here. Then, the severity index value of each disaster level is used as the numerator, the cumulative value of the severity index values ​​of all disaster levels is used as the denominator, and the ratio is used as the reference weight of each disaster level, thereby achieving normalization processing and making the sum of the reference weights of all disaster levels equal to the value 1.

[0074] Step S203: Using the reference weight of each disaster level, weighted summation is performed on the data fluctuation level of the target monitoring parameter at each disaster level during the target disaster process to obtain the response sensitivity of the target monitoring parameter during the target disaster process.

[0075] Then, the reference weight of each disaster level can be used to perform weighted summation on the data fluctuation level of the target monitoring parameters at each disaster level during the target disaster process, and the response sensitivity of the target monitoring parameters during the target disaster process can be obtained.

[0076] As an example, in one embodiment of the present invention, the expression of the response sensitivity of the target monitoring parameter in the target disaster process may be specifically, for example, as follows:

[0077]

[0078] in, It represents the response sensitivity of the target monitoring parameters in the target disaster process; Indicates Reference weights for each disaster level; Indicates the target monitoring parameters in the target disaster process. The level of data fluctuation for each disaster level; Indicates the number of disaster levels. In the embodiment of the present invention, there are four disaster levels.

[0079] The response sensitivity of each monitoring parameter in the target disaster process can be obtained by the same method as above.

[0080] Step S3: According to the correlation between the target monitoring parameter and other monitoring parameters except the target monitoring parameter, the correlation between the target monitoring parameter and the runoff parameter, and the difference in response sensitivity between the target monitoring parameter and other monitoring parameters in the target disaster process, the influence level of the target monitoring parameter in the target disaster process is obtained; according to the difference in the influence level between the target monitoring parameter in the target disaster process and other flood disaster processes except the target disaster process, and the influence level of the target monitoring parameter in each flood disaster process, the disaster impact weight of the target monitoring parameter is obtained.

[0081] Since the degree of association between different monitoring parameters and the occurrence of flood disasters is different during the target disaster process, and there is a certain correlation between different monitoring parameters, the stronger the correlation between the target monitoring parameters and other monitoring parameters, the more critical the target monitoring parameters are in the target disaster process, and the smaller the difference in response sensitivity between the target monitoring parameters and other monitoring parameters in the target disaster process, the more similar the sensitivity of the target monitoring parameters and other monitoring parameters to the response of flood disasters in the target disaster process is, which further indicates that during the target disaster process, the stronger the correlation between the target monitoring parameters and the occurrence of flood disasters, the greater their impact on the occurrence of flood disasters. At the same time, considering that the scale of flood disasters depends on the runoff parameters, that is, the runoff parameters are directly related to the occurrence of flood disasters, the stronger the correlation between the target monitoring parameters and the runoff parameters, the greater the impact of the target monitoring parameters on the occurrence of flood disasters during the target disaster process.

[0082] Therefore, the correlation between the target monitoring parameters and other monitoring parameters except the target monitoring parameters, the correlation between the target monitoring parameters and the runoff parameters, and the difference in response sensitivity between the target monitoring parameters and other monitoring parameters in the target disaster process can be analyzed. The obtained influence level reflects the degree of influence of the target monitoring parameters on the target disaster process. The greater the influence level, the greater the influence of the target monitoring parameters on the occurrence of flood disasters compared with other monitoring parameters in the target disaster process. Subsequently, a comprehensive analysis can be conducted on the influence levels of the target monitoring parameters in each flood disaster process, so as to calculate and analyze the influence weights of the target monitoring parameters on the occurrence of flood disasters, and improve the accuracy of subsequent flood disaster warnings.

[0083] Preferably, in one embodiment of the present invention, the method for obtaining the impact level of the target monitoring parameter on the target disaster process specifically includes:

[0084] See also Figure 4 , which shows a flow chart of a method for obtaining the impact level of a target monitoring parameter on a target disaster process provided by an embodiment of the present invention.

[0085] Step S301: taking the absolute value of the Pearson correlation coefficient between the target monitoring parameter in the target disaster process and each other monitoring parameter as the first correlation coefficient between the target monitoring parameter in the target disaster process and each other monitoring parameter.

[0086] From the above analysis, it can be seen that the stronger the correlation between the target monitoring parameter and other monitoring parameters, the more critical the target monitoring parameter is in the target disaster process, which further indicates that in the target disaster process, compared with other monitoring parameters, the target monitoring parameter is more strongly correlated with the occurrence of flood disasters. Therefore, the absolute value of the Pearson correlation coefficient between the target monitoring parameter in the target disaster process and each other monitoring parameter can be used as the first correlation coefficient between the target monitoring parameter in the target disaster process and each other monitoring parameter. Subsequently, the influence level of the target monitoring parameter in the target disaster process can be accurately calculated based on the first correlation coefficient.

[0087] Step S302: Obtain the impact priority of the target monitoring parameter in the target disaster process according to the difference in response sensitivity between the target monitoring parameter and each other monitoring parameter in the target disaster process, and the first correlation coefficient.

[0088] The smaller the difference in response sensitivity between the target monitoring parameter and other monitoring parameters in the target disaster process, the more similar the sensitivity of the target monitoring parameter and other monitoring parameters to flood disaster response in the target disaster process is, which further indicates that in the target disaster process, compared with other monitoring parameters, the target monitoring parameter is more strongly correlated with the occurrence of flood disasters. Therefore, the impact priority of the target monitoring parameter in the target disaster process can be obtained based on the difference in response sensitivity between the target monitoring parameter and each other monitoring parameter in the target disaster process, as well as the first correlation coefficient. The greater the impact priority, the more priority the target monitoring parameter has in affecting flood disasters than other monitoring parameters.

[0089] Preferably, in one embodiment of the present invention, the method for obtaining the impact priority of the target monitoring parameter on the target disaster process specifically includes:

[0090] The absolute value of the difference in response sensitivity between the target monitoring parameter and each other monitoring parameter in the target disaster process is negatively correlated to obtain the sensitivity similarity between the target monitoring parameter and each other monitoring parameter in the target disaster process. The sensitivity similarity and the first correlation coefficient between the target monitoring parameter and each other monitoring parameter in the target disaster process are integrated to obtain the connectivity parameter between the target monitoring parameter and each other monitoring parameter in the target disaster process. The average value of the connectivity parameter between the target monitoring parameter in the target disaster process and all other monitoring parameters is taken as the impact priority of the target monitoring parameter in the target disaster process.

[0091] In an embodiment of the present invention, the sum or product of the sensitive similarity and the first correlation coefficient between the target monitoring parameter in the target disaster process and each other monitoring parameter can be used as a connectivity parameter between the target monitoring parameter in the target disaster process and each other monitoring parameter to achieve integration of the two, which is not limited here.

[0092] As an example, in one embodiment of the present invention, the expression of the impact priority of the target monitoring parameter on the target disaster process can be specifically, for example, as follows:

[0093]

[0094] in, Indicates the impact priority of the target monitoring parameter on the target disaster process; It represents the response sensitivity of the target monitoring parameters in the target disaster process; Indicates the parameters other than the target monitoring parameters. The sensitivity of other monitoring parameters to the target disaster process; Indicates the target monitoring parameters and the The sensitivity similarity between other monitoring parameters; Indicates the target monitoring parameters and the Pearson correlation coefficients between other monitoring parameters; Indicates the target monitoring parameters and the The first correlation coefficient between other monitoring parameters; Indicates the number of other monitoring parameters besides the target monitoring parameters; Indicates the target monitoring parameters and the Linkage parameters between other monitoring parameters; Indicates the preset adjustment parameter, which is used to prevent the denominator from being 0. The value range is In one embodiment of the present invention, Set to 0.01, The specific value of can also be set by the implementer according to the specific implementation scenario and is not limited here.

[0095] In other embodiments of the present invention, negative correlation mapping may also be implemented based on other basic mathematical operations, which is not limited here.

[0096] Step S303: taking the absolute value of the Pearson correlation coefficient between the target monitoring parameter and the runoff parameter during the target disaster process as the second correlation coefficient between the target monitoring parameter and the runoff parameter during the target disaster process.

[0097] Since the runoff parameters are directly related to the occurrence of flood disasters and the correlation between the two is the strongest, the absolute value of the Pearson correlation coefficient between the target monitoring parameters and the runoff parameters during the target disaster process can be used as the second correlation coefficient between the target monitoring parameters and the runoff parameters during the target disaster process. The larger the second correlation coefficient, the stronger the correlation between the target monitoring parameters and the occurrence of flood disasters.

[0098] Step S304: Integrate the impact priority and the second correlation coefficient to obtain the impact level of the target monitoring parameter on the target disaster process.

[0099] Then, the impact priority and the second correlation coefficient can be integrated to obtain the impact level of the target monitoring parameter on the target disaster process.

[0100] In an embodiment of the present invention, the sum or product of the impact priority and the second correlation coefficient may be used as the impact level of the target monitoring parameter on the target disaster process to achieve a combination of the two, which is not limited here.

[0101] As an example, in one embodiment of the present invention, the expression of the influence level of the target monitoring parameter on the target disaster process can be specifically, for example, as follows:

[0102]

[0103] in, Indicates the impact level of the target monitoring parameter on the target disaster process; Indicates the impact priority of the target monitoring parameter on the target disaster process; represents the Pearson correlation coefficient between the target monitoring parameter and the runoff parameter during the target disaster process; Represents the second correlation coefficient between the target monitoring parameter and the runoff parameter during the target disaster process.

[0104] The impact level of the target monitoring parameters in each flood disaster process can be obtained by the same method as above.

[0105] Since there are certain differences in the impact levels of target monitoring parameters on the occurrence of different flood disaster processes, the greater the difference in the impact level of target monitoring parameters between the target disaster process and each other flood disaster process, the more prominent the impact of the target monitoring parameters on the target disaster process. Therefore, the disaster impact weight of the target monitoring parameter can be obtained based on the difference in the impact level of the target monitoring parameter between the target disaster process and other flood disaster processes except the target disaster process, and the comprehensive impact level of the target monitoring parameter in each flood disaster process. The larger the disaster impact weight, the greater the reference value and importance of the target monitoring parameter in predicting the occurrence of flood disasters. Subsequently, the flood disaster can be accurately predicted based on the disaster impact weight of each monitoring parameter.

[0106] Preferably, in one embodiment of the present invention, the method for obtaining the disaster impact weight of the target monitoring parameter specifically includes:

[0107] The absolute value of the difference in the impact level of the target monitoring parameter between the target disaster process and each other flood disaster process is taken as the difference value of the impact level of the target monitoring parameter between the target disaster process and each other flood disaster process.

[0108] The accumulated value of the difference in the impact levels of the target monitoring parameters between the target disaster process and all other flood disaster processes is normalized to obtain the confidence level of the impact of the target monitoring parameters on the target disaster process.

[0109] In one embodiment of the present invention, the accumulated value of the difference in impact levels of the target monitoring parameter between the target disaster process and all other flood disaster processes can be used as the accumulated difference value of the target monitoring parameter in the target disaster process, and the accumulated difference value of the target monitoring parameter in each flood disaster process can be calculated based on this. The accumulated difference value of the target monitoring parameter in the target disaster process is used as the numerator, the accumulated value of the accumulated difference values ​​of the target monitoring parameter in all flood disaster processes is used as the denominator, and the ratio is used as the impact confidence of the target monitoring parameter in the target disaster process to achieve normalization processing. The impact confidence of the target monitoring parameter in each flood disaster process is calculated based on this, so that the sum of the impact confidences of the target monitoring parameter in all flood disaster processes is equal to the value 1.

[0110] Then, the confidence level of the target monitoring parameter in each flood disaster process can be used to perform weighted summation and normalization on the impact level of the target monitoring parameter in each flood disaster process to obtain the disaster impact weight of the target monitoring parameter.

[0111] In one embodiment of the present invention, the confidence level of the impact of the target monitoring parameter in each flood disaster process can be used to perform a weighted summation of the impact level of the target monitoring parameter in each flood disaster process as the comprehensive impact level of the target monitoring parameter, and the comprehensive impact level of each monitoring parameter can be calculated based on this. The comprehensive impact level of the target monitoring parameter is used as the numerator, the cumulative value of the comprehensive impact levels of all monitoring parameters is used as the denominator, and the ratio is used as the disaster impact weight of the target monitoring parameter to achieve normalization processing, so that the sum of the disaster impact weights of all monitoring parameters is equal to the value 1.

[0112] As an example, in one embodiment of the present invention, the expression of the disaster impact weight of the target monitoring parameter may be specifically, for example, as follows:

[0113]

[0114]

[0115] in, Represents the disaster impact weight of the target monitoring parameter; Indicates the comprehensive impact level of the target monitoring parameters; Indicates The comprehensive impact level of the various monitoring parameters; Indicates the number of monitored parameters; Indicates the target monitoring parameter in Confidence level of the impact of the secondary flood disaster process; Indicates the target monitoring parameter in Impact level of sub-flood disaster process; Represents the number of flood hazard processes.

[0116] The disaster impact weight of each monitoring parameter can be obtained by the same method as above.

[0117] Step S4: Based on the real-time monitoring data and disaster impact weight of each monitoring parameter, an early warning is issued for the flood disaster in the monitored basin.

[0118] The greater the disaster impact weight of a certain monitoring parameter, the more the monitoring parameter is a key factor in causing flood disasters, and further the greater the reference value and importance of the monitoring parameter in predicting the occurrence of flood disasters. Therefore, accurate early warning of flood disasters in the measured basin can be carried out based on the real-time monitoring data and disaster impact weight of each monitoring parameter.

[0119] Preferably, in one embodiment of the present invention, the method for early warning of flood disasters in a test basin specifically includes:

[0120] First, the warning thresholds of different disaster levels for each monitoring parameter are obtained, and the difference between the real-time monitoring data of each monitoring parameter and the minimum value of the warning threshold of each monitoring parameter is used as the real-time initial judgment value of each monitoring parameter. When the real-time initial judgment value of each monitoring parameter is less than 0, it means that the monitoring data of the monitoring parameter has not reached the level of flood disaster. When it is not less than 0, it means that the monitoring data of the monitoring parameter exceeds the warning threshold of the initial disaster level, which may cause a larger degree of flood disaster.

[0121] Then, the disaster impact weight of each monitoring parameter can be used to perform weighted summation on the initial judgment value to obtain the real-time comprehensive judgment value of the basin to be tested. If the real-time comprehensive judgment value of the basin to be tested is not less than 0, it means that a flood disaster may occur. The larger the comprehensive judgment value, the greater the possibility of a flood disaster. The function maps the comprehensive judgment value to obtain the real-time disaster possibility of the measured basin; if the real-time comprehensive judgment value of the measured basin is less than 0, it means that the possibility of flood disaster is extremely small. At this time, the real-time disaster possibility of the measured basin can be set to 0, and it is considered that flood disaster is impossible.

[0122] As an example, in one embodiment of the present invention, the expression of the real-time disaster possibility of the watershed to be measured may be specifically, for example, as follows:

[0123]

[0124]

[0125] in, Indicates the real-time disaster possibility of the basin to be tested; Indicates the real-time comprehensive judgment value of the watershed to be tested; Indicates Disaster impact weights of various monitoring parameters; Indicates Real-time monitoring data of various monitoring parameters; Indicates The minimum value of the early warning threshold of the monitoring parameter; Indicates the number of monitored parameters; Indicates Real-time initial judgment value of various monitoring parameters; Represents the activation function.

[0126] Then, based on the real-time disaster possibility of the tested basin, early warning of flood disasters in the tested basin can be issued.

[0127] Preferably, in one embodiment of the present invention, when the disaster probability of the measured basin at each moment within a preset time period is greater than a preset probability threshold, a flood disaster warning is issued, wherein the preset time period is set to 5 hours, and the preset probability threshold is set to 0.7. The specific values ​​of the preset time period and the preset probability threshold can also be set by the implementer according to the specific implementation scenario, which is not limited here.

[0128] By collecting various types of hydrogeological data of the test basin and conducting comprehensive analysis, a flood disaster prediction model is established, which improves the accuracy of flood disaster prediction in the test basin.

[0129] An embodiment of the present invention provides a flood disaster warning system based on hydrogeological data, the system comprising a memory, a processor and a computer program, wherein the memory is used to store the corresponding computer program, the processor is used to run the corresponding computer program, and when the computer program runs in the processor, the method described in steps S1 to S4 can be implemented.

[0130] 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 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.

[0131] 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.

Claims

1. A flood disaster early warning method based on hydrogeological data, It is characterized in that The method comprises: Acquire monitoring data segments of different types of monitoring parameters and runoff parameters in the measured basin during each flood disaster process, and acquire early warning thresholds of different disaster levels of the runoff parameters, wherein the monitoring data segments include monitoring data at different times, and the monitoring parameters do not include the runoff parameters; Taking any flood disaster process as the target disaster process and any monitoring parameter as the target monitoring parameter, the monitoring data segments of the runoff parameter in the target disaster process are divided based on different early warning thresholds of the runoff parameter, and multiple data sub-segments of the target monitoring parameter in each disaster level in the target disaster process are obtained; according to the distribution of monitoring data of the target monitoring parameter in the data sub-segments at different disaster levels in the target disaster process, and the value of each disaster level, the response sensitivity of the target monitoring parameter in the target disaster process is obtained; According to the correlation between the target monitoring parameter and other monitoring parameters except the target monitoring parameter, the correlation between the target monitoring parameter and the runoff parameter, and the difference in the response sensitivity between the target monitoring parameter and the other monitoring parameters in the target disaster process, the influence level of the target monitoring parameter in the target disaster process is obtained; according to the difference in the influence level between the target monitoring parameter in the target disaster process and other flood disaster processes except the target disaster process, and the influence level of the target monitoring parameter in each flood disaster process, the disaster impact weight of the target monitoring parameter is obtained; According to the real-time monitoring data of each monitoring parameter and the disaster impact weight, early warning of flood disasters in the monitored basin is issued; The step of obtaining the response sensitivity of the target monitoring parameter in the target disaster process includes: Analyze the discrete degree of the monitoring data of each data sub-segment of the target monitoring parameter in the target disaster process at each disaster level, and obtain the data fluctuation level of the target monitoring parameter in each disaster level during the target disaster process; After negative correlation mapping and normalization of the values ​​of each disaster level, the reference weight of each disaster level is obtained; Using the reference weight of each disaster level, weighted summation is performed on the data fluctuation level of the target monitoring parameter at each disaster level during the target disaster process to obtain the response sensitivity of the target monitoring parameter during the target disaster process; The data fluctuation level of the target monitoring parameter at each disaster level in the process of obtaining the target disaster includes: The range of the monitoring data of each data sub-segment of the target monitoring parameter at each disaster level during the target disaster process is taken as the fluctuation degree of each data sub-segment of the target monitoring parameter at each disaster level during the target disaster process; The average value of the fluctuation degree of all data subsegments of the target monitoring parameter at each disaster level during the target disaster process is taken as the data fluctuation level of the target monitoring parameter at each disaster level during the target disaster process; The influence level of the target monitoring parameter on the target disaster process is obtained as follows: The absolute value of the Pearson correlation coefficient between the target monitoring parameter and each other monitoring parameter during the target disaster process is used as the first correlation coefficient between the target monitoring parameter and each other monitoring parameter during the target disaster process; Obtaining the impact priority of the target monitoring parameter in the target disaster process according to the difference in the response sensitivity between the target monitoring parameter and each other monitoring parameter in the target disaster process, and the first correlation coefficient; The absolute value of the Pearson correlation coefficient between the target monitoring parameter and the runoff parameter during the target disaster process is used as the second correlation coefficient between the target monitoring parameter and the runoff parameter during the target disaster process; The impact priority and the second correlation coefficient are integrated to obtain the impact level of the target monitoring parameter on the target disaster process; The priority of obtaining the impact of the target monitoring parameters on the target disaster process includes: Performing negative correlation mapping on the absolute value of the difference between the target monitoring parameter and each other monitoring parameter in the response sensitivity of the target disaster process, and obtaining the sensitivity similarity between the target monitoring parameter and each other monitoring parameter in the target disaster process; Combining the sensitive similarity and the first correlation coefficient between the target monitoring parameter in the target disaster process and each other monitoring parameter to obtain a connection parameter between the target monitoring parameter in the target disaster process and each other monitoring parameter; The average value of the connection parameter between the target monitoring parameter and all other monitoring parameters in the target disaster process is used as the impact priority of the target monitoring parameter in the target disaster process; The disaster impact weights of the target monitoring parameters are obtained as follows: The absolute value of the difference between the impact level of the target monitoring parameter between the target disaster process and each other flood disaster process is taken as the impact level difference value between the target monitoring parameter and each other flood disaster process; Normalizing the accumulated value of the impact level difference value of the target monitoring parameter between the target disaster process and all other flood disaster processes to obtain the impact confidence of the target monitoring parameter in the target disaster process; Using the impact confidence of the target monitoring parameter in each flood disaster process, weighted summing and normalizing the impact level of the target monitoring parameter in each flood disaster process is performed to obtain the disaster impact weight of the target monitoring parameter; The early warning of flood disasters in the tested basin includes: The difference between the real-time monitoring data of each monitoring parameter and the minimum value of the warning threshold of each monitoring parameter is used as the real-time initial judgment value of each monitoring parameter; Using the disaster impact weight of each monitoring parameter, weighted summation is performed on the initial judgment value to obtain a real-time comprehensive judgment value of the watershed to be measured; If the real-time comprehensive judgment value of the watershed to be tested is not less than 0, use The function maps the comprehensive judgment value to obtain the real-time disaster possibility of the watershed to be tested; if the real-time comprehensive judgment value of the watershed to be tested is less than the value 0, the real-time disaster possibility of the watershed to be tested is set to the value 0; Based on the real-time disaster possibility of the tested basin, early warning of flood disasters in the tested basin is issued.

2. A flood disaster early warning method based on hydrogeological data according to claim 1, It is characterized in that The multiple data subsegments of the target monitoring parameters at each disaster level in the process of obtaining the target disaster include: Based on different early warning thresholds of runoff parameters, the monitoring data segments of runoff parameters in the target disaster process are divided to obtain multiple data sub-segments of runoff parameters in each disaster level in the target disaster process; During the target disaster process, the time periods corresponding to the data sub-segments of the runoff parameters at each disaster level are used as the disaster time periods of the target disaster process at each disaster level, and the monitoring data segments corresponding to the disaster time periods of the target monitoring parameters at each disaster level are used as the multiple data sub-segments of the target monitoring parameters at each disaster level during the target disaster process.

3. A flood disaster early warning method based on hydrogeological data according to claim 1, It is characterized in that The early warning of flood disasters in the tested basin based on the real-time disaster possibility of the tested basin includes: If the disaster probability of the tested basin at each moment in the preset time period is greater than the preset probability threshold, a flood disaster warning will be issued.

4. A flood disaster early warning system based on hydrogeological data, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 3 are implemented.

Images