A flood prevention early warning method and system based on artificial intelligence

Abstract

The present invention provides a flood prevention and early warning method and system based on artificial intelligence, which belongs to the field of production-type artificial intelligence technology, and specifically includes: obtaining associated mountain areas with different historical flash flood times, determining the flash flood flow of the associated mountain areas with different historical flash flood times, and determining the terrain change risk of the associated mountain areas in combination with the occurrence times of different historical flash flood times, using the terrain change risk to determine the risk change area, using the terrain change risk and position of the risk change area and the flash flood risk as input, using a prediction model based on an artificial intelligence algorithm to obtain a corrected flash flood risk, and using the corrected flash flood risk to determine whether early warning processing is needed, thereby ensuring the accuracy and timeliness of flood prevention and early warning processing.

Description

Technical Field

[0001] The present invention belongs to the technical field of production-oriented artificial intelligence, and in particular relates to a flood prevention and early warning method and system based on artificial intelligence. Background Art

[0002] In recent years, unexpected weather events such as short-term heavy rainfall and snowfall have occurred from time to time. Therefore, in order to carry out flood prevention warnings in a timely manner, the invention patent application CN202311130447.6 "A flash flood disaster forecasting and warning method and system" realizes timely warning of flash flood disasters by comprehensively considering soil data and expected precipitation data, thereby reducing the losses caused by flash flood disasters. However, at the same time, there are the following technical problems:

[0003] For mountains where flash floods have occurred, their terrain conditions often change to a certain extent. At the same time, the changes in terrain conditions tend to develop in the direction of a higher risk of flash floods. Therefore, if the changes in the terrain conditions of the mountains are not taken into account when conducting flood prevention and warning processing, the timeliness and accuracy of the warning processing cannot be guaranteed.

[0004] In response to the above technical problems, the present invention provides a flood prevention and early warning method and system based on artificial intelligence. Summary of the invention

[0005] To achieve the purpose of the present invention, the present invention adopts the following technical solutions:

[0006] According to one aspect of the present invention, a flood prevention and early warning method based on artificial intelligence is provided.

[0007] A flood prevention and early warning method based on artificial intelligence, specifically comprising:

[0008] S1 obtains rainfall data and predicted rainfall data of a mountain within a preset time period, and determines the flash flood risk of the mountain by using the similarity between the rainfall data, predicted rainfall data and data of different historical flash flood times, and proceeds to the next step when the flash flood risk is within a preset interval;

[0009] S2: Determine the deviation of the flash flood flow between different historical flash flood times, and when it is determined that the mountain has a risk of terrain change in combination with the deviation of the rainfall data between different historical flash flood times, proceed to the next step;

[0010] S3: obtaining the associated mountain areas with different historical flash flood times, determining the flash flood flow of the associated mountain areas with different historical flash flood times, and determining the terrain change risk of the associated mountain areas in combination with the occurrence time of different historical flash flood times, and determining the risk change area using the terrain change risk;

[0011] S4 takes the terrain change risk and location of the risk change area and the flash flood risk as input, adopts a prediction model based on an artificial intelligence algorithm to obtain a modified flash flood risk, and uses the modified flash flood risk to determine whether early warning processing is needed.

[0012] The beneficial effects of the present invention are:

[0013] 1. In the present invention, whether there is a risk of terrain change in the mountain is determined by the deviation of the flash flood flow between different historical flash flood times and the deviation of rainfall data, thereby achieving an accurate assessment of the risk of terrain change in the mountain from the deviation of rainfall data under similar flash flood flows, and avoiding the emergence of technical problems such as inaccurate prediction results of the overall flash flood prediction risk caused by the single use of rainfall data, thereby improving the reliability of the management and control of the flash flood risk of the mountain.

[0014] 2. In the present invention, the terrain change risk and location of the risk change area and the flash flood risk are taken as input, and a prediction model based on an artificial intelligence algorithm is used to obtain a modified flash flood risk. This not only takes into account the difference in the size of the flash flood risk caused by the difference in rainfall data, but also comprehensively considers the impact of terrain change on the flash flood risk by further combining the location of the risk change area and the terrain change risk.

[0015] A further technical solution is that the preset time period is determined according to the area of ​​the mountain and the vegetation coverage rate, wherein the larger the area of ​​the mountain and the higher the vegetation coverage rate, the longer the preset time period.

[0016] A further technical solution is that the predicted rainfall data is determined based on weather forecast data from a meteorological station.

[0017] A further technical solution is that the flash flood risk has a value range of 0 to 1, wherein the greater the flash flood risk of the mountain, the greater the risk of flash floods occurring in the mountain.

[0018] A further technical solution is that, when the flash flood risk of the mountain is not within the preset range, it is necessary to further determine whether the flash flood risk of the mountain is greater than the preset risk threshold. If so, it is determined that early warning processing is required; if not, it is determined that early warning processing is not required.

[0019] A further technical solution is that when the risk of terrain change of the mountain does not meet the requirements, it is determined that no early warning processing is required.

[0020] A further technical solution is that the artificial intelligence algorithm includes BP neural network, CNN convolutional neural network, fully connected neural network, and generative adversarial network.

[0021] A further technical solution is to use the modified flash flood risk to determine whether early warning processing is needed, specifically including:

[0022] When the modified flash flood risk is not within the preset risk range, it is determined that the mountain needs to be warned.

[0023] In a second aspect, the present invention provides a computer system comprising: a memory and a processor that are communicatively connected, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-mentioned artificial intelligence-based flood prevention and warning method when running the computer program.

[0024] Other features and advantages will be described in the following description. The objects and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description and drawings.

[0025] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

[0027] Figure 1 It is a flow chart of a flood prevention and early warning method based on artificial intelligence;

[0028] Figure 2 It is a flow chart of the method for determining the flash flood risk of a mountain;

[0029] Figure 3 It is a flow chart of the method for determining the risk change area;

[0030] Figure 4 It is a framework diagram of a computer system. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0032] For mountains where flash floods have occurred, their terrain conditions often change to a certain extent. At the same time, the changes in terrain conditions tend to develop in the direction of a higher risk of flash floods. Therefore, it is necessary to comprehensively consider the terrain changes of flash floods and carry out flood prevention and warning treatment for the mountains.

[0033] In order to solve the above technical problems, the following technical solutions are adopted:

[0034] Determine the flash flood risk of the mountain by using the rainfall data, the predicted rainfall data and the data similarity of different historical flash flood times;

[0035] When the flash flood risk is within the preset range, the risk of terrain change is determined based on the changes in flash flood flow of historical flash floods and the deviation of rainfall data;

[0036] When the terrain change risk does not meet the requirements, the risk change area is determined based on the flash flood flow of the associated mountain area at different historical flash flood times;

[0037] Taking the terrain change risk and location of the risk change area and the flash flood risk as input, a prediction model based on an artificial intelligence algorithm is used to obtain a modified flash flood risk, and the modified flash flood risk is used to determine whether early warning processing is needed.

[0038] Flash flood risk: Utilize the similarity between the rainfall data, predicted rainfall data and the data of different historical flash floods to determine the rainfall periods similar to the different historical flash floods, determine the similarity coefficients with the different historical flash floods based on the proportion of the number of similar rainfall periods, and determine the flash flood risk based on the average value of the similarity coefficients with the different historical flash floods. Specifically, for example, if the similarity coefficients with the different historical flash floods are 0.7, 0.5 and 0.6 respectively, then the flash flood risk is 0.6.

[0039] When the flash flood risk is above 0.8, a flood warning can be directly issued. When the flash flood risk is below 0.4, it is determined that there is no need to issue a flood warning. When it is between 0.4 and 0.8, the risk of terrain change is assessed.

[0040] Determine whether the mountain is at risk of terrain change: determine it based on the deviation of flash flood flow between different historical flash flood times, and the deviation of rainfall data between different historical flash flood times, and determine it based on the change in the ratio of rainfall data to flash flood flow under different historical flash flood times within one year. Specifically, if the ratios are 0.7, 0.5, and 0.3, respectively, and gradually become smaller, it means that less rainfall data results in a larger flash flood flow, then it is determined that the mountain is at risk of terrain change.

[0041] Associated mountain areas: mountain areas where floods have occurred.

[0042] The risk of terrain change in associated mountainous areas: The risk of terrain change is determined based on the flash flood flow of the historical flash floods in the associated mountainous areas and the time when different historical flash floods occurred. Specifically, the corresponding preset change risks under different flash flood flows are determined based on the flash flood flow. Specifically, when the preset change risks of different historical flash floods are 0.6, 0.4, and 0.3, and the interval between the corresponding historical flash flood occurrence times and the current time is 1 year, 6 months, and 3 months, the preset weight coefficients corresponding to the interval time are 0.3, 0.4, and 0.7, then the terrain change risk is 0.18+0.16+0.21=0.55, and when the terrain change risk is greater than 0.5, it is a risk change area.

[0043] Artificial intelligence algorithms are constructed through neural networks such as BP neural network, LSTM neural network, and CNN convolutional neural network.

[0044] Furthermore, the preset time period is determined according to the area of ​​the mountain and the vegetation coverage rate, wherein the larger the area of ​​the mountain and the higher the vegetation coverage rate, the longer the preset time period.

[0045] It should be noted that the predicted rainfall data is determined based on the weather forecast data from the meteorological station.

[0046] In one embodiment, Figure 2 As shown, the method for determining the flash flood risk of the mountain is:

[0047] Obtaining rainfall data of the mountain in different division unit time periods, and determining the proportion of similar rainfall time periods in different historical flash flood times based on the deviation of rainfall data in different division unit time periods;

[0048] Obtaining the predicted rainfall data of the mountain in different divided unit time periods within a preset time period in the future, and determining the proportion of similar predicted rainfall time periods in different historical flash flood times based on the deviation of the predicted rainfall data in different divided unit time periods within the preset time period in the future;

[0049] Determine the similarity of rainfall data between the mountain and different historical flash floods according to the proportion of the number of similar rainfall periods and the proportion of the number of predicted rainfall periods, and determine the number of similar flash floods in the historical flash floods according to the similarity of the rainfall data;

[0050] The flash flood risk is determined by taking the average value of the similarity of rainfall data of similar flash flood times.

[0051] Furthermore, the flash flood risk has a value range of 0 to 1, wherein the greater the flash flood risk of the mountain, the greater the risk of flash floods occurring in the mountain.

[0052] It is understandable that when the flash flood risk of the mountain is not within the preset range, it is necessary to further determine whether the flash flood risk of the mountain is greater than the preset risk threshold. If so, it is determined that early warning processing is required; if not, it is determined that early warning processing is not required.

[0053] In another possible embodiment, the method for determining the flash flood risk of the mountain is:

[0054] Obtaining rainfall data of the mountain in different division unit time periods, and determining similarity with previous rainfall data of different historical flash flood times based on an average value of deviations of the rainfall data in different division unit time periods;

[0055] Obtaining the predicted rainfall data of the mountain in different divided unit time periods within a preset time period in the future, and determining the similarity with the predicted rainfall data of different historical flash flood times based on the average value of the deviation amount of the predicted rainfall data in different divided unit time periods within the preset time period in the future;

[0056] Determine the similarity of rainfall data between the mountain and different historical flash floods according to the average of the similarity of previous rainfall data and the similarity of predicted rainfall data, and determine the similar flash floods in the historical flash floods according to the similarity of rainfall data;

[0057] The flash flood risk is determined by taking the average value of the similarity of rainfall data of similar flash flood times.

[0058] In another possible embodiment, the method for determining the flash flood risk of the mountain is:

[0059] Obtain rainfall data of the mountain in different divided unit time periods, and determine the ratio of similar rainfall periods in different historical flash flood times based on the deviation of rainfall data in different divided unit time periods, and judge whether the ratio of similar rainfall periods is greater than the preset ratio of historical flash flood times. If so, the historical flash flood times with the ratio of similar rainfall periods greater than the preset ratio are regarded as potential similar flash flood times. If not, it is determined that the mountain does not need to be warned.

[0060] Obtain the predicted rainfall data of the mountain in different divided unit time periods within the future preset time period, and determine the proportion of the number of similar predicted rainfall time periods in different potential similar flash flood times based on the deviation of the predicted rainfall data in different divided unit time periods within the future preset time period, and judge whether there is a potential similar flash flood number whose proportion of the number of similar preset rainfall time periods is greater than the preset proportion, if so, proceed to the next step, if not, determine that the mountain does not need to be warned;

[0061] The number of potential similar flash floods in which the number of similar preset rainfall periods accounts for a greater proportion than the preset proportion is taken as the number of suspected similar flash floods, and it is determined whether the number of suspected similar flash floods is greater than the preset number of flash floods. If so, it is determined that the mountain needs to be early-warning processed, and if not, proceed to the next step;

[0062] Based on the average value of the deviation of rainfall data in different divided unit time periods and the average value of the deviation of predicted rainfall data in different divided unit time periods within a preset time period in the future, the similarity of previous rainfall data and predicted rainfall data with different suspected similar flash flood times is determined;

[0063] Determine the similarity of rainfall data between the mountain and different suspected similar flash floods according to the average of the similarity of the previous rainfall data and the similarity of the predicted rainfall data, and determine whether the mountain has similar flash floods according to the similarity of the rainfall data. If yes, proceed to the next step; if not, determine that the mountain needs to be processed for early warning;

[0064] The weight values ​​of different similar flash flood times are determined based on the similarity of the rainfall data of the similar flash flood times, and the flash flood risk is determined based on the sum of the weight values ​​of the different similar flash flood times.

[0065] Specifically, it should be noted that the risk of terrain change in the mountain is determined to include:

[0066] Determine the flow similarity periods between different historical flash floods by using the deviation of flash flood flows in different time periods, determine the flash flood flow similarity between different historical flash floods according to the proportion of the flow similarity periods, and divide the historical flash floods into different flow similarity groups according to the flash flood flow similarity;

[0067] Based on the average value of the deviation of rainfall data in different time periods, the deviation of rainfall data between different historical flash flood times is determined, and the number of abnormal flow flash floods in different flow similar groups is determined by using the deviation of rainfall data of historical flash flood times in different divided time periods, and the flow abnormality values ​​of different flow abnormality groups are determined by the proportion of the number of abnormal flow flash flood times;

[0068] The flow anomaly values ​​of different flow anomaly groups are normalized to obtain processing results, the terrain change risk of the mountain is determined according to the sum of the processing results of different flow anomaly groups, and a preset change risk threshold is used to determine whether the terrain change risk of the mountain meets the requirements.

[0069] Further, the preset change risk threshold is used to determine whether the terrain change risk of the mountain meets the requirements, specifically including:

[0070] When the terrain change risk of the mountain is less than the preset change risk threshold, it is determined that the terrain change risk of the mountain meets the requirement.

[0071] It is understandable that when the terrain change risk of the mountain does not meet the requirement, it is determined that no early warning processing is required.

[0072] Furthermore, the abnormal flow flash flood number is the historical flash flood number whose rainfall data deviation from the historical flash flood number in other divided time periods does not meet the requirements.

[0073] In another possible embodiment, determining that the mountain has a risk of terrain change specifically includes:

[0074] Determine the flow similarity periods between different historical flash floods by using the deviation of flash flood flows in different time periods, determine the flash flood flow similarity between different historical flash floods according to the proportion of the flow similarity periods, and divide the historical flash floods into different flow similarity groups according to the flash flood flow similarity;

[0075] Determine the deviation of rainfall data between different historical flash flood times based on the average value of the deviation of rainfall data in different time periods, and judge whether there is a flow similarity group whose rainfall data deviation between historical flash flood times does not meet the requirements. If so, proceed to the next step; if not, determine that there is no risk of terrain change in the mountain;

[0076] The flow similarity group in which the deviation of rainfall data between the historical flash flood times does not meet the requirements is taken as the rainfall data deviation group, and the rainfall data deviation of the historical flash flood times between different divided time periods is determined by using the occurrence time of different historical flash flood times, and it is judged whether there is a rainfall data deviation group in which the deviation of rainfall data of the historical flash flood times between the divided time periods does not meet the requirements, if so, proceed to the next step, if not, it is determined that the mountain does not have the risk of terrain change;

[0077] The rainfall data deviation group in which the deviation amount of the historical flash flood times between the divided time periods does not meet the requirements is taken as the deviation data group, and it is determined whether the number of the deviation data groups is greater than the preset number of groups. If not, the next step is entered. If yes, it is determined that the mountain has a terrain change risk;

[0078] The number of abnormal flow flash floods in different flow similar groups is determined by using the deviation of rainfall data of the number of historical flash floods between divided time periods, and the flow abnormality values ​​of different deviation data groups are determined by the number ratio of the abnormal flow flash floods, and it is judged whether the number of deviation data groups whose flow abnormality values ​​do not meet the requirements meets the requirements. If so, proceed to the next step, if not, it is determined that the mountain has a risk of terrain change;

[0079] The flow anomaly values ​​of different deviation data groups are normalized to obtain processing results, the terrain change risk of the mountain is determined according to the sum of the processing results of different deviation data groups, and a preset change risk threshold is used to determine whether the terrain change risk of the mountain meets the requirements.

[0080] Furthermore, the associated mountain area is determined based on the path of the flash flood flow of the historical flash flood times.

[0081] Specifically, it should be noted that Figure 3 As shown, the method for determining the risk change area is:

[0082] The impact values ​​of the flash flood flows at different historical flash flood times are determined by using the flash flood flow intervals at which the flash flood flows at different historical flash flood times are located;

[0083] Divide the historical flash flood times into different unit time periods according to the occurrence times of different historical flash flood times, obtain the flash flood flow impact values ​​of the historical flash flood times in different unit time periods, and determine the change risk impact values ​​of different unit time periods in combination with the interval lengths between different historical flash flood times;

[0084] The weight values ​​of different unit time periods are determined based on the time intervals in which they are located, and the terrain change risks of the associated mountain areas are determined using the weight values ​​and the change risk impact values, and the risk change areas in the associated mountain areas are determined based on the terrain change risks.

[0085] Further, determining the risk change area in the associated mountain area according to the terrain change risk specifically includes:

[0086] When the terrain change risk of the associated mountain area does not meet the requirement, the associated mountain area is determined to be a risk change area.

[0087] In another possible embodiment, the method for determining the risk change area is:

[0088] S31: obtaining the number of historical flash floods in the associated mountain area, and determining whether the number of historical flash floods in the associated mountain area is greater than a preset number threshold. If so, proceeding to step S35; if not, proceeding to the next step;

[0089] S32 divides the number of historical flash floods into different unit time periods according to the occurrence time of different historical flash floods, and determines whether the associated mountain area has experienced a flash flood within the preset unit time. If not, proceed to the next step. If not, proceed to step S35.

[0090] S33 determines the impact values ​​of the flash flood flows of different historical flash floods by using the flash flood flow intervals of the flash flood flows of different historical flash floods, and judges whether the average values ​​of the impact values ​​of the flash flood flows of different historical flash floods are less than a preset threshold value. If so, it is determined that the associated mountain area does not belong to the risk change area. If not, proceed to the next step.

[0091] S34 determines the number of flash floods that seriously affect the number of flash floods based on the flash flood flow impact values ​​of different historical flash floods, and judges whether the time between the occurrence time of the flash floods that seriously affect the number of flash floods and the current time is greater than a preset time. If so, it is determined that the associated mountain area does not belong to the risk change area. If not, proceed to the next step;

[0092] S35 divides the historical flash floods into different unit time periods according to the occurrence times of different historical flash floods, obtains the flash flood flow impact values ​​of the historical flash floods in different unit time periods, and determines the change risk impact values ​​of different unit time periods in combination with the intervals between different historical flash floods;

[0093] S36 determines the weight values ​​of different unit time periods based on the time intervals in which they are located, and uses the weight values ​​and the change risk impact values ​​to determine the terrain change risks of the associated mountain areas, and determines the risk change areas in the associated mountain areas through the terrain change risks.

[0094] Furthermore, the artificial intelligence algorithm includes BP neural network, CNN convolutional neural network, fully connected neural network, and generative adversarial network.

[0095] It should be noted that the use of the modified flash flood risk to determine whether early warning processing is needed specifically includes:

[0096] When the modified flash flood risk is not within the preset risk range, it is determined that the mountain needs to be warned.

[0097] Embodiment 2 In the second aspect, as Figure 4 As shown, the present invention provides a computer system, comprising: a memory and a processor that are communicatively connected, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-mentioned artificial intelligence-based flood prevention and warning method when running the computer program.

[0098] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0099] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the 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.

[0100] The above description is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, one or more embodiments of this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included in the scope of the claims of this specification.

Claims

1. A flood prevention and early warning method based on artificial intelligence, characterized in that: Specifically include: Obtain rainfall data and predicted rainfall data of the mountain within a preset time period, and determine the flash flood risk of the mountain by using the similarity between the rainfall data, predicted rainfall data and data of different historical flash flood times, and when the flash flood risk is within a preset interval, proceed to the next step; Determine the deviation of the flash flood flow between different historical flash flood times, and when it is determined that the mountain has a risk of terrain change based on the deviation of the rainfall data between different historical flash flood times, proceed to the next step; Obtaining the associated mountain areas with different historical flash flood times, determining the flash flood flow of the associated mountain areas with different historical flash flood times, and determining the terrain change risk of the associated mountain areas in combination with the occurrence times of different historical flash flood times, and determining the risk change area using the terrain change risk; Taking the terrain change risk and location of the risk change area and the flash flood risk as input, a prediction model based on an artificial intelligence algorithm is used to obtain a modified flash flood risk, and the modified flash flood risk is used to determine whether early warning processing is needed; Determine that the mountain has a risk of terrain change, including: Determine the flow similarity periods between different historical flash floods by using the deviation of flash flood flows in different time periods, determine the flash flood flow similarity between different historical flash floods according to the proportion of the flow similarity periods, and divide the historical flash floods into different flow similarity groups according to the flash flood flow similarity; Based on the average value of the deviation of rainfall data in different time periods, the deviation of rainfall data between different historical flash flood times is determined, and the number of abnormal flow flash floods in different flow similar groups is determined by using the deviation of rainfall data of historical flash flood times in different divided time periods, and the flow abnormality values ​​of different flow abnormality groups are determined by the proportion of the number of abnormal flow flash flood times; Normalizing the flow anomaly values ​​of different flow anomaly groups to obtain processing results, determining the terrain change risk of the mountain according to the sum of the processing results of different flow anomaly groups, and determining whether the terrain change risk of the mountain meets the requirements using a preset change risk threshold; The preset change risk threshold is used to determine whether the terrain change risk of the mountain meets the requirements, specifically including: When the terrain change risk of the mountain is less than the preset change risk threshold, it is determined that the terrain change risk of the mountain meets the requirement.

2. The artificial intelligence-based flood prevention and early warning method according to claim 1, characterized in that: The preset time period is determined according to the area of ​​the mountain and the vegetation coverage rate, wherein the larger the area of ​​the mountain and the higher the vegetation coverage rate, the longer the preset time period.

3. The flood prevention and early warning method based on artificial intelligence according to claim 1, characterized in that: The predicted rainfall data is determined based on the weather forecast data from the meteorological station.

4. The artificial intelligence-based flood prevention and early warning method according to claim 1, characterized in that: The method for determining the flash flood risk of the mountain is: Obtaining rainfall data of the mountain in different division unit time periods, and determining the proportion of similar rainfall time periods in different historical flash flood times based on the deviation of rainfall data in different division unit time periods; Obtaining the predicted rainfall data of the mountain in different divided unit time periods within a preset time period in the future, and determining the proportion of similar predicted rainfall time periods in different historical flash flood times based on the deviation of the predicted rainfall data in different divided unit time periods within the preset time period in the future; Determine the similarity of rainfall data between the mountain and different historical flash floods according to the proportion of the number of similar rainfall periods and the proportion of the number of predicted rainfall periods, and determine the number of similar flash floods in the historical flash floods according to the similarity of the rainfall data; The flash flood risk is determined by taking the average value of the similarity of rainfall data of similar flash flood times.

5. The artificial intelligence-based flood prevention and early warning method according to claim 1, characterized in that: When the flash flood risk of the mountain is not within the preset range, it is necessary to further determine whether the flash flood risk of the mountain is greater than the preset risk threshold. If so, it is determined that early warning processing is required; if not, it is determined that no early warning processing is required.

6. The artificial intelligence-based flood prevention and early warning method according to claim 1, characterized in that: The abnormal flow flash flood times are historical flash flood times whose rainfall data deviation from the historical flash flood times in other divided time periods does not meet the requirements.

7. The artificial intelligence-based flood prevention and early warning method according to claim 1, characterized in that: The associated mountain area is determined according to the path of the flash flood flow of the historical flash flood times.

8. A computer system comprising: A memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, characterized in that when the processor runs the computer program, an artificial intelligence-based flood prevention and early warning method as described in any one of claims 1-7 is executed.

Images