Water conservancy early warning information targeting pushing method and system based on mobile phone signaling big data

Abstract

This invention belongs to the field of water conservancy early warning and big data processing technology, specifically involving a method and system for targeted push of water conservancy early warning information based on mobile phone signaling big data. The method includes: generating the geographic spatial range of the water conservancy early warning area and its surrounding adjacent areas; obtaining the base stations in the water conservancy early warning area and its surrounding adjacent areas; acquiring historical mobile phone signaling datasets related to the base stations within the most recent time period, identifying two categories of mobile phone users in the water conservancy early warning area: permanent residents and short-term transient residents; acquiring real-time mobile phone signaling datasets related to the base stations in the water conservancy early warning area and its surrounding adjacent areas, identifying mobile phone users within the water conservancy early warning area and those about to enter the area; automatically pushing early warning information to the three categories of mobile phone users according to preset early warning rules; and real-time monitoring of the population size and spatial distribution within the water conservancy early warning area. This invention further improves the accuracy and timeliness of water conservancy early warning information push and has wider applicability.

Description

Technical Field

[0001] This invention belongs to the field of water conservancy early warning and big data processing technology, specifically involving a method and system for targeted push of water conservancy early warning information based on mobile phone signaling big data. Background Technology

[0002] Water conservancy emergencies such as floods, mudslides, dam failures, water pollution, and flood discharge are characterized by their suddenness, unpredictability, and irregularity. Therefore, issuing early warnings to the public is crucial for ensuring public safety. Traditional methods of issuing warnings include SMS messages, television, radio, and the internet. However, these methods cannot accurately determine the scope and affected individuals, often only considering local residents and neglecting tourists, visitors, and temporary transit passengers. This results in blind spots in warning dissemination, and also wastes SMS resources by sending warnings to many unrelated individuals, potentially causing unnecessary public panic.

[0003] Mobile signaling data is the time and location information captured and recorded by communication base stations when mobile phone users make calls, send text messages, or move their locations. It has advantages such as high spatiotemporal accuracy, wide coverage, stability, reliability, and timely updates. Mobile signaling data is highly correlated with the spatiotemporal distribution of human activities and has wide applications in various fields of natural resources and social development. Mobile signaling data is used in the water conservancy sector; according to searches...

[0004] Chinese patent document with publication number CN111711920B discloses a method for constructing an urban rainstorm and flood early warning scheme based on mobile phone signaling data. The method identifies the user's residence, workplace, and weekend travel destination through mobile phone signaling data, and then, in combination with a rainstorm spatial distribution map, identifies mobile phone users within the rainstorm and flood danger zone and sends rainstorm warning information to the mobile phone users.

[0005] Chinese patent document CN117576876B discloses a method for early warning of urban flood disasters targeting potential at-risk individuals. This method uses flood risk data and mobile phone signaling data to identify at-risk groups and issue early warning information to them. However, the above research only applies mobile phone signaling data to flood disaster scenarios and cannot be used in emergency water conservancy scenarios such as mudslides, dam breaks, water pollution, and flood discharge, thus limiting its application scope.

[0006] In view of this, the inventors hope to provide a method and system for targeted push of water conservancy early warning information based on mobile phone signaling big data. Summary of the Invention

[0007] The purpose of this invention is to overcome the aforementioned problems in traditional technologies and provide a targeted push method and system for water conservancy early warning information based on mobile phone signaling big data. This method and system are applicable to most water conservancy early warning scenarios, can accurately locate people within the water conservancy early warning area and people who are about to go to the area, and can accurately push early warning information according to different groups of people, significantly improving the accuracy and effectiveness of water conservancy early warning information push.

[0008] To achieve the above-mentioned technical objectives and effects, the present invention is implemented through the following technical solution:

[0009] This invention provides a method for targeted delivery of water conservancy early warning information based on mobile phone signaling big data, comprising the following steps:

[0010] S1. Based on the early warning requirements of the water conservancy scenario, generate the geographical spatial range of the water conservancy early warning area and the surrounding adjacent area;

[0011] S2. Based on the geographic spatial range of the water conservancy early warning area and the surrounding adjacent areas, and mobile communication base station data, obtain the base station number set data of the water conservancy early warning area and the base station number set data of the adjacent areas of the water conservancy early warning area, respectively.

[0012] S3. Based on the base station number set data of the water conservancy early warning area, obtain the historical mobile phone signaling raw data related to the base station in the most recent period, preprocess the historical mobile phone signaling raw data, and generate the historical mobile phone signaling dataset of the water conservancy early warning area.

[0013] S4. Based on the historical mobile signaling dataset of the water conservancy warning area, use the population analysis model to identify two types of mobile phone users related to the water conservancy warning area: permanent residents and short-term mobile users. The permanent residents mobile phone user group includes mobile phone users whose residence or workplace is located in the water conservancy warning area and whose commuting route passes through the water conservancy warning area.

[0014] S5. Based on the base station number set data of the water conservancy early warning area and the base station number set data of the adjacent area of ​​the water conservancy early warning area, respectively obtain the real-time mobile phone signaling raw data related to the base station, preprocess the real-time mobile phone signaling raw data, and generate the real-time mobile phone signaling dataset of the water conservancy early warning area and the real-time mobile phone signaling dataset of the adjacent area of ​​the water conservancy early warning area.

[0015] S6. Based on the real-time mobile signaling dataset of the water conservancy warning area, use a population analysis model to identify the temporary transit mobile phone user group within the water conservancy warning area; based on the real-time mobile signaling dataset of the neighboring areas of the water conservancy warning area, use a population travel trajectory analysis and prediction model to identify the temporary transit mobile phone user group about to enter the water conservancy warning area.

[0016] S7. Based on preset warning rules, automatically push warning information to three categories of mobile phone users—residents, short-term migrants, and temporary transit users—via SMS or voice call, and track the completion of warning information push.

[0017] S8. Monitor and analyze the population size and spatial distribution within the water conservancy early warning area in real time, and display the analysis results and early warning information push completion status in the form of visual charts.

[0018] Furthermore, step S1 is further divided into the following steps:

[0019] S101. Generate the geospatial extent of the water conservancy early warning area using one of the following four methods: manual drawing, buffer analysis, selecting administrative regions, or adding vector files.

[0020] Manual drawing is based on manually drawing rectangular or polygonal areas on a map to generate the geographic spatial range of the water conservancy early warning area;

[0021] Buffer analysis involves drawing point, line, and polygon features on a map and automatically generating buffer zones based on buffer radii to create the geographic spatial extent of water conservancy early warning areas.

[0022] Selecting an administrative region involves choosing the name of the province, city, district / county, street / township, community / village, and using the corresponding administrative boundaries as the geographical spatial range of the water conservancy early warning area.

[0023] Adding vector files involves uploading vector files in formats such as shp, json, and kml, or text files containing continuous coordinate values, to generate the geospatial extent of the water conservancy early warning area.

[0024] S102. Perform GIS buffer analysis on the geographic spatial range of the water conservancy early warning area, set the buffer radius, and generate the geographic spatial range of the adjacent area outside the water conservancy early warning area.

[0025] Furthermore, step S2 is further divided into the following steps:

[0026] S201. Using the signal coverage radius of the mobile communication base station as the buffer radius, perform GIS buffer analysis on the vector points of the mobile communication base station to generate the signal coverage area vector data of the mobile communication base station.

[0027] S202. Perform GIS spatial overlay analysis on the geographic spatial range of the water conservancy early warning area and the mobile communication base station data to obtain the mobile communication base stations related to the water conservancy early warning area and generate a set of base station number data for the water conservancy early warning area.

[0028] S203. Perform GIS spatial overlay analysis on the geographic spatial range and mobile communication base station data of the adjacent area outside the water conservancy early warning area to obtain the mobile communication base stations in the adjacent area of ​​the water conservancy early warning area and generate a set of base station number data for the adjacent area of ​​the water conservancy early warning area.

[0029] Furthermore, step S3 is further divided into the following steps:

[0030] S301. Using the set of base station numbers in the water conservancy early warning area as the query condition, obtain the historical mobile phone signaling data related to the base station within the most recent time period; wherein the mobile phone signaling data contains the mobile user identification code, signaling timestamp, and the encoding information of the current base station;

[0031] S302. Using a distributed computing framework, historical mobile phone signaling data is sorted according to mobile user identification code and signaling timestamp. Preprocessing operations are performed, including removing non-personal SIM cards, deduplication of multiple SIM cards per person, removal of duplicate time points, and handling of abnormal data, to generate a historical mobile phone signaling dataset P for the water conservancy early warning area. Specific information is as follows:

[0032] P = {p1, p2, ..., p} i ,p n |1≤i≤n}, where p i Let n be the mobile signaling dataset for the i-th mobile phone user, where n is the number of mobile phone users.

[0033] p i ={q1,q2,…,q j ,q m |1≤j≤m}, where m is the number of mobile signaling data, q j This refers to the j-th mobile signaling data in the mobile signaling dataset;

[0034] q j = {d, t, s}, where d is the mobile subscriber identification code, t is the timestamp, and s is the base station code.

[0035] Furthermore, step S4 is further divided into the following steps:

[0036] S401. Based on the historical mobile signaling dataset of the water conservancy early warning area, obtain the entire activity chain of each mobile phone user, calculate the cumulative daytime stay value of the same mobile phone user on weekdays, add mobile phone users whose cumulative daytime stay value on weekdays reaches the threshold for daytime stay value to the working population seed set; add mobile phone users whose frequency of appearance in the working population seed set reaches the threshold for working stay days to the set of mobile phone users whose workplace is located in the water conservancy early warning area.

[0037] S402. Based on the historical mobile phone signaling dataset of the water conservancy early warning area, obtain the entire activity chain of each mobile phone user, calculate the cumulative nighttime stay value of the same mobile phone user, add mobile phone users whose cumulative nighttime stay value reaches the threshold of nighttime stay value to the seed set of residents; add mobile phone users whose number of occurrences in the seed set of residents reaches the threshold of number of days of residence to the set of mobile phone users whose residence is located in the water conservancy early warning area.

[0038] S403. Based on the historical mobile phone signaling dataset of the water conservancy warning area, remove the mobile phone signaling data of mobile phone users whose residence or workplace is located in the water conservancy warning area, and generate a temporary mobile phone signaling dataset; based on the temporary mobile phone signaling dataset, obtain the entire activity chain of each mobile phone user, calculate the number of days the same mobile phone user appears during the weekday commuting period, and add the mobile phone users whose number of days appearing during the weekday commuting period reaches the weekday commuting day threshold to the set of mobile phone users commuting through the water conservancy warning area;

[0039] S404. Combine the sets of mobile phone users whose residences and workplaces are located in the water conservancy warning area and those who commute through the water conservancy warning area, and combine them with the basic information of mobile phone users to generate a dataset of mobile phone numbers of permanent residents in the water conservancy warning area; the basic information of mobile phone users includes mobile user identification code, mobile phone number, age, and gender;

[0040] S405. For mobile phone users who are elderly, they are marked as key groups of concern in the resident population mobile phone number dataset by using labels.

[0041] S406. Based on the historical mobile signaling dataset of the water conservancy warning area, obtain the entire activity chain of each mobile phone user, calculate the daily cumulative stay time value of the same mobile phone user, and add mobile phone users whose daily cumulative stay time value reaches the daily stay time threshold of the floating population but is less than the daily stay time threshold of the permanent population to the floating population seed set; add mobile phone users whose frequency of appearance in the floating population seed set is less than the floating population stay days threshold and who are still in the water conservancy warning area on the most recent day to the short-term floating population mobile phone user set of the water conservancy warning area, and generate a short-term floating population mobile phone number dataset of the water conservancy warning area by combining the basic information of the mobile phone users.

[0042] Furthermore, step S5 is further divided into the following steps:

[0043] S501: Using the base station number set of the water conservancy early warning area and the base station number set of the adjacent area of ​​the water conservancy early warning area as query conditions, the real-time mobile phone signaling data of the water conservancy early warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy early warning area are automatically obtained at fixed time intervals.

[0044] S502. Using a distributed computing framework, the real-time mobile phone signaling data of the water conservancy early warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy early warning area are sorted according to the mobile user identification code and the signaling timestamp. After performing operations such as removing non-personal SIM cards, deduplication of multiple SIM cards for one person, removal of duplicate time point data, and abnormal data processing, the real-time mobile phone signaling dataset of the water conservancy early warning area and the real-time mobile phone signaling dataset of the adjacent area of ​​the water conservancy early warning area are generated.

[0045] Furthermore, step S6 is further divided into the following steps:

[0046] S601. Based on the real-time mobile signaling dataset of the water conservancy early warning area, combined with the basic information of mobile phone users, obtain the seed set of mobile phone numbers of temporary transit population in the water conservancy early warning area, remove the mobile phone numbers of the permanent population and short-term floating population mobile phone users in S4, and then remove the mobile phone numbers that have not appeared for the first time, and generate the mobile phone number dataset of temporary transit population in the water conservancy early warning area.

[0047] S602. Based on the real-time mobile phone signaling dataset of the adjacent area of ​​the water conservancy warning area, and using the population travel trajectory analysis and prediction model, obtain the mobile phone user group that is about to enter the water conservancy warning area, exclude mobile phone users that are not appearing for the first time, and generate a temporary transit population mobile phone number dataset that is about to enter the water conservancy warning area.

[0048] Furthermore, in step S602, the population travel trajectory analysis and prediction model is based on the historical travel trajectory data and road network vector data extracted from mobile phone signaling data. It uses deep learning technology to analyze the population travel direction, speed, and trajectory-road correlation to predict the future travel trajectory of the population. It obtains the probability that mobile phone users in the vicinity of the water conservancy warning area will enter the water conservancy warning area, and classifies mobile phone users whose probability exceeds the threshold as mobile phone users who are about to enter the water conservancy warning area.

[0049] Furthermore, step S7 is further divided into the following steps:

[0050] S701. Based on the specific circumstances of water conservancy emergencies, early warning information shall be set in advance for the permanent residents, short-term floating population and temporary transit population in the water conservancy early warning area, and parameters such as start time, push interval and validity period shall be set.

[0051] S702. Based on preset warning rules, automatically push warning information to the mobile phone number datasets of permanent residents, short-term migrants, and temporary transit residents in the water conservancy warning area by sending warning text messages or making voice calls; for mobile phone users marked as "key attention groups" in the mobile phone number dataset of permanent residents, push warning information by making voice calls as needed.

[0052] S703. Mark the mobile phone users whose warning information was successfully pushed and whose warning information was not pushed, and generate a dataset of mobile phone numbers whose warning information was successfully pushed and a dataset of mobile phone numbers whose warning information was not pushed.

[0053] S704. Resend the warning information to the mobile phone numbers where the warning information failed to be sent, and proceed to step S703 until all relevant mobile phone users in the water conservancy warning area have successfully received the warning information.

[0054] Furthermore, step S8 is further divided into the following steps:

[0055] S801: Based on the mobile phone number datasets of permanent residents, short-term migrants, and temporary transit residents in the water conservancy early warning area, count the number of permanent residents, short-term migrants, and temporary transit residents related to the water conservancy early warning area, as well as the number of successful and failed early warning information pushes;

[0056] S802: Spatially locate the base station on the map, and generate corresponding population spatial distribution heat maps by using the number of residents, working people, commuters, short-term migrants, and temporary transit passengers at the base station as indicators.

[0057] This invention also provides a targeted push system for water conservancy early warning information based on mobile phone signaling big data, comprising:

[0058] The water conservancy early warning area generation module generates the geographic spatial range of the water conservancy early warning area and its surrounding neighboring areas by manually drawing, buffer analysis, selecting administrative regions, and adding vector files. It also filters and obtains the base station number set data of the water conservancy early warning area and the base station number set data of the water conservancy early warning area and its neighboring areas.

[0059] The mobile signaling data acquisition and processing module uses the base station number set data of the water conservancy early warning area and the specified time as conditions to obtain the original mobile signaling data related to the base stations in the water conservancy early warning area, and performs data storage and data processing operations through a distributed computing framework to generate a mobile signaling dataset of the water conservancy early warning area.

[0060] The resident population analysis module identifies the residential population, working population, and commuting population related to the water conservancy warning area based on historical mobile phone signaling datasets within a certain time period. Combined with basic mobile phone user information, it generates a dataset of mobile phone numbers of the resident population in the water conservancy warning area.

[0061] The short-term migrant population analysis module identifies the short-term migrant population within the water conservancy warning area based on historical mobile phone signaling datasets over a certain period of time. Combined with basic mobile phone user information, it generates a dataset of mobile phone numbers of the short-term migrant population within the water conservancy warning area.

[0062] The temporary transit population analysis module, based on the real-time mobile phone signaling dataset of the water conservancy warning area and the real-time mobile phone signaling dataset of the adjacent area of ​​the water conservancy warning area, calculates and analyzes the temporary transit mobile phone user groups located in the water conservancy warning area and about to enter the water conservancy emergency area, respectively. Combined with the basic information of mobile phone users, it generates a dataset of mobile phone numbers of the temporary transit population located in the water conservancy warning area and about to enter the water conservancy warning area.

[0063] The early warning information push module automatically pushes early warning information to the relevant permanent residents, short-term floating populations, and temporary transit populations in the water conservancy early warning area via SMS or voice call, based on the preset early warning information and set parameters, and tracks the completion of the early warning information push.

[0064] The real-time population monitoring and analysis module monitors and analyzes the population size and spatial distribution within the water conservancy early warning area in real time, and displays the analysis results and the completion status of early warning SMS push in the form of visual charts.

[0065] The beneficial effects of this invention are:

[0066] 1. This invention, based on mobile signaling big data and GIS technology, accurately identifies the residential population, working population, commuter population, and temporary transit population related to water conservancy emergency areas, and automatically categorizes and pushes early warning information. Specifically, it accurately identifies and categorizes the population within the water conservancy early warning area, enabling targeted delivery of different early warning information to different mobile phone user groups, thus improving the accuracy and timeliness of water conservancy early warnings. This invention further enhances water conservancy safety and, to some extent, reduces the cost, manpower, and material costs of pushing early warning information via SMS.

[0067] 2. This invention can be used in various water conservancy scenarios such as floods, mudslides, dam breaks, water pollution, and flood discharge, and has wide applicability.

[0068] 3. This invention uses a population travel trajectory analysis and prediction model to identify mobile phone users who are about to enter a water conservancy warning area and pushes warning information in advance to avoid or reduce personnel safety problems caused by entering the water conservancy warning area.

[0069] 4. This invention identifies elderly people within water conservancy early warning areas, enabling focused attention on this key population group and reducing the risks they face.

[0070] Of course, any product implementing this invention does not necessarily need to achieve all of the above advantages at the same time. Attached Figure Description

[0071] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0072] Figure 1 This is a flowchart of the method for targeted push of water conservancy early warning information based on mobile phone signaling big data, which is a method of the present invention.

[0073] Figure 2 This is a schematic diagram of the framework of the water conservancy early warning information targeted push system based on mobile phone signaling big data of the present invention. Detailed Implementation

[0074] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0075] Example 1

[0076] like Figure 1 As shown, this embodiment provides a method for targeted push of water conservancy early warning information based on mobile phone signaling big data, including the following steps:

[0077] Step S1: Based on the early warning requirements of the water conservancy scenario, generate the geographic spatial range of the water conservancy early warning area and its surrounding adjacent areas. The specific steps are as follows:

[0078] S101. Generate the geospatial extent of the water conservancy early warning area using one of the following four methods: manual drawing, buffer analysis, selecting administrative regions, or adding vector files. Specifically:

[0079] The manual drawing is based on manually drawing rectangular or polygonal areas on a map to generate the geographic spatial range of the water conservancy early warning area;

[0080] The buffer analysis involves drawing point, line, and polygon features on a map, automatically generating a buffer zone based on the buffer radius, and generating the geographic spatial range of the water conservancy early warning area.

[0081] The selected administrative region refers to the selection of the names of provinces, cities, districts / counties, subdistricts / towns, communities / villages, and the corresponding administrative boundaries are used as the geographical spatial range of the water conservancy early warning area;

[0082] The addition of vector files involves uploading vector files in formats such as shp, json, and kml, or text files containing continuous coordinate values, to generate the geographic spatial range of the water conservancy early warning area.

[0083] S102. Perform GIS buffer analysis on the geographic spatial range of the water conservancy early warning area, set the buffer radius, and generate the geographic spatial range of the adjacent area of ​​the water conservancy early warning area.

[0084] Step S2: Based on the geographic spatial range of the water conservancy early warning area and its surrounding adjacent areas, and mobile communication base station data, obtain the base station number set data of the water conservancy early warning area and the base station number set data of the adjacent areas of the water conservancy early warning area. The specific steps are as follows:

[0085] S201. Using the signal coverage radius of the mobile communication base station as the buffer radius, perform GIS buffer analysis on the vector points of the mobile communication base station to generate the signal coverage area vector data of the mobile communication base station.

[0086] S202. Perform GIS spatial overlay analysis on the geographic spatial range and mobile communication base station data of the water conservancy early warning area to obtain the mobile communication base stations related to the water conservancy early warning area and generate a set of base station number data for the water conservancy early warning area.

[0087] S203. Perform GIS spatial overlay analysis on the geographic spatial range and mobile communication base station data of the adjacent area of ​​the water conservancy early warning area to obtain the mobile communication base stations in the adjacent area of ​​the water conservancy early warning area and generate a set of base station number data of the adjacent area of ​​the water conservancy early warning area.

[0088] Step S3: Based on the base station number set data of the water conservancy early warning area, obtain the original historical mobile phone signaling data related to the base stations within a certain period, preprocess the original historical mobile phone signaling data, and generate the historical mobile phone signaling dataset of the water conservancy early warning area. The specific steps are as follows:

[0089] S301. Using the set of base station numbers in the water conservancy early warning area as the query condition, obtain historical mobile phone signaling data within the last 10 working days; the mobile phone signaling data contains information such as mobile user identification code, signaling timestamp, and the code of the current base station;

[0090] S302. Using Hadoop+Spark to build a distributed computing framework, the historical mobile phone signaling data is sorted according to the mobile user identification code and signaling timestamp. Preprocessing operations such as removing non-personal SIM cards, deduplication of multiple SIM cards per person, removal of duplicate time points, and anomaly processing are performed to generate a historical mobile phone signaling dataset P for the water conservancy early warning area. Specific information is as follows:

[0091] P = {p1, p2, ..., p} i ,pn |1≤i≤n}, where p i Let n be the mobile signaling dataset for the i-th mobile phone user, where n is the number of mobile phone users.

[0092] p i ={q1,q2,…,q j ,q m |1≤j≤m}, where m is the number of mobile signaling data, q j This refers to the j-th mobile signaling data in the mobile signaling dataset;

[0093] q j = {d, t, s}, where d is the mobile subscriber identification code, t is the timestamp, and s is the base station code;

[0094] In this embodiment, the removal of non-personal ID cards in step S302 specifically involves:

[0095] By utilizing the one-to-one correspondence between SIM cards and mobile user identification codes, mobile signaling data generated by non-human SIM cards such as IoT cards can be identified and eliminated;

[0096] In this embodiment, the deduplication of multiple cards for one person in step S302 specifically involves:

[0097] Calculate the trajectory overlap of multiple mobile user identification codes, perform deduplication of multiple SIM cards for one person within the network, and remove the corresponding mobile signaling data;

[0098] In this embodiment, the abnormal data processing in step S302 specifically involves:

[0099] The base station with the most user connections during the "ping-pong effect" period is designated as the main base station. Records other than those from the main base station are removed from the signaling data generated during the "ping-pong effect" period to complete the data cleaning.

[0100] Step S4: Based on the historical mobile signaling dataset of the water conservancy warning area, use a population analysis model to identify two types of mobile phone user groups related to the water conservancy warning area: permanent residents and short-term transient residents. The permanent residents mobile phone user group includes mobile phone users whose residence or workplace is located in the water conservancy warning area and whose commuting route passes through the water conservancy warning area. The specific steps are as follows:

[0101] S401. Based on the historical mobile signaling dataset of the water conservancy early warning area, obtain the entire activity chain of each mobile phone user, calculate the cumulative daily stay time of the same mobile phone user during the period from 9:00 to 16:30 on weekdays, add mobile phone users whose cumulative daily stay time reaches the threshold of 5 hours to the working population seed set; add mobile phone users whose number of occurrences in the working population seed set reaches the threshold of 6 working stay days to the set of mobile phone users whose workplace is located in the water conservancy early warning area.

[0102] S402. Based on the historical mobile phone signaling dataset of the water conservancy early warning area, calculate the cumulative nighttime stay value of the same mobile phone user during the period from 21:00 to 6:00. Add mobile phone users whose cumulative nighttime stay value reaches the threshold of 7 hours to the seed set of residents. Add mobile phone users whose number of occurrences in the seed set of residents reaches the threshold of 6 days of residence to the set of mobile phone users whose residence is located in the water conservancy early warning area.

[0103] S403. Based on the historical mobile phone signaling dataset of the water conservancy early warning area, remove the mobile phone signaling data of mobile phone users whose residence or workplace is located in the water conservancy early warning area, and generate a temporary mobile phone signaling dataset; based on the temporary mobile phone signaling dataset, calculate the number of days the same mobile phone user appears during the time periods of 6:00-9:00 and 16:00-19:00 on weekdays, and add mobile phone users whose number of days appearing during the weekday commuting period reaches the weekday commuting day threshold of 6 days to the set of mobile phone users commuting through the water conservancy early warning area;

[0104] S404. Combine the sets of mobile phone users whose residences and workplaces are located within the water conservancy warning area and those who commute through the water conservancy warning area, and combine them with the basic information of mobile phone users to generate a dataset of mobile phone numbers of permanent residents related to the water conservancy warning area.

[0105] Step S5: Based on the base station number set data of the water conservancy early warning area and the base station number set data of the adjacent area of ​​the water conservancy early warning area, obtain the real-time mobile phone signaling raw data related to the base stations respectively, preprocess the real-time mobile phone signaling raw data, and generate the real-time mobile phone signaling dataset of the water conservancy early warning area and the real-time mobile phone signaling dataset of the adjacent area of ​​the water conservancy early warning area. The specific steps are as follows:

[0106] S501. Using the base station number set of the water conservancy early warning area and the base station number set of the adjacent area of ​​the water conservancy early warning area as query conditions, the real-time mobile phone signaling data of the water conservancy early warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy early warning area are automatically obtained in a time period of 5 minutes.

[0107] S502. Using a distributed computing framework, the real-time mobile phone signaling data of the water conservancy early warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy early warning area are sorted according to the mobile user identification code and the signaling timestamp. After performing operations such as removing non-personal SIM cards, deduplicating multiple SIM cards for one person, removing duplicate time point data, and processing abnormal data, a real-time mobile phone signaling dataset of the water conservancy early warning area and a real-time mobile phone signaling dataset of the adjacent area of ​​the water conservancy early warning area are generated.

[0108] Step S6: Based on the real-time mobile signaling dataset of the flood warning area, use a population analysis model to identify temporary transit mobile phone users within the flood warning area; based on the real-time mobile signaling dataset of neighboring areas of the flood warning area, use a population trajectory analysis and prediction model to identify temporary transit mobile phone users about to enter the flood warning area. The specific steps are as follows:

[0109] S601. Based on the real-time mobile signaling dataset of the water conservancy early warning area, remove the mobile signaling data records of the mobile phone users of the permanent population and the short-term floating population in S4, and generate the mobile signaling dataset of the temporary transit population of the water conservancy early warning area.

[0110] S602. Based on the mobile phone signaling dataset of temporary transit population in the water conservancy early warning area, and combined with the basic information of mobile phone users, obtain the seed set of mobile phone numbers of temporary transit population in the water conservancy early warning area, and then remove the mobile phone numbers that do not appear for the first time to generate the mobile phone number dataset of temporary transit population in the water conservancy early warning area.

[0111] S603. Based on the real-time mobile signaling dataset of the adjacent area of ​​the water conservancy warning area, remove the mobile signaling data records of mobile user groups that have not appeared for the first time, and generate a temporary transit population mobile signaling dataset of the adjacent area of ​​the water conservancy warning area.

[0112] S604. Based on the mobile phone signaling dataset of temporary transit populations in adjacent areas of the water conservancy warning zone, and using the population travel trajectory analysis and prediction model, determine whether mobile phone users will enter the water conservancy warning zone, and generate a dataset of mobile phone numbers of temporary transit populations who are about to enter the water conservancy warning zone.

[0113] The population travel trajectory analysis and prediction model is built based on LSTM neural network technology, using historical population travel trajectory data extracted from mobile phone signaling data and road network vector data. The model analyzes population travel direction, speed, and trajectory-road correlation to predict future travel trajectories, obtains the probability of mobile phone users in neighboring areas entering flood warning zones, and categorizes mobile phone users whose probabilities exceed a threshold as those about to enter flood warning zones.

[0114] Step S7: Based on preset warning rules, automatically push warning information to three categories of mobile phone users—residents, short-term migrants, and temporary transit users—via SMS or voice call, and track the completion of warning information push. The specific steps are as follows:

[0115] S701. Based on the specific circumstances of water conservancy emergency response, pre-classify and set early warning information for the permanent resident population, short-term floating population and temporary transit population in the water conservancy early warning area, and set parameters such as start push time, push interval and validity period for each category.

[0116] S702. According to the preset warning rules, the system automatically pushes warning information to the mobile phone number datasets of permanent residents, short-term migrants, and temporary transit residents in the water conservancy warning area by sending warning text messages or making voice calls. For mobile phone users marked as "key attention groups" in the mobile phone number dataset of permanent residents, the system pushes warning information by making voice calls as needed.

[0117] S703. Use 0 and 1 to mark mobile phone users whose warning information was successfully pushed and whose warning information was not pushed, respectively, and generate a dataset of mobile phone numbers whose warning information was successfully pushed and a dataset of mobile phone numbers whose warning information was not pushed.

[0118] S704. Resend the warning information to the mobile phone numbers where the warning information failed to be sent, and proceed to step S703 until all relevant mobile phone users in the water conservancy warning area have successfully received the warning information.

[0119] Step S8: Monitor and analyze the population size and spatial distribution within the water conservancy early warning area in real time, and display the analysis results and early warning information push completion status in the form of visual charts. The specific steps are as follows:

[0120] S801. Based on the mobile phone number datasets of permanent residents, short-term migrants, and temporary transit residents in the water conservancy early warning area, count the number of permanent residents, short-term migrants, and temporary transit residents related to the water conservancy early warning area, as well as the number of successful and failed early warning information pushes;

[0121] S802. Spatially locate the base station on the map, and use the number of residents, working people, commuters, short-term migrants, and temporary transit passengers as indicators to generate corresponding population spatial distribution heat maps through GIS spatial interpolation, and use red, green, and blue colors to represent the population size.

[0122] Example 2

[0123] like Figure 2 As shown, this embodiment provides a targeted push system for water conservancy early warning information based on mobile phone signaling big data, including:

[0124] The water conservancy early warning area generation module generates the geographic spatial range of the water conservancy early warning area and the surrounding adjacent areas by manually drawing, buffer analysis, selecting administrative regions, and adding vector files. It also filters and obtains the base station number set data of the water conservancy early warning area and the base station number set data of the adjacent areas of the water conservancy early warning area.

[0125] The mobile signaling data acquisition and processing module uses the base station number set data of the water conservancy early warning area and a specified time as conditions to obtain the raw mobile signaling data related to the base stations in the water conservancy early warning area, and performs data storage and data processing operations through a distributed computing framework to generate a mobile signaling dataset of the water conservancy early warning area.

[0126] The resident population analysis module identifies the residential population, working population, and commuting population related to the water conservancy warning area based on historical mobile phone signaling datasets within a certain time period. Combined with basic mobile phone user information, it generates a dataset of mobile phone numbers of the resident population in the water conservancy warning area.

[0127] The short-term migrant population analysis module identifies the short-term migrant population within the water conservancy warning area based on historical mobile signaling datasets over a certain period of time. It also generates a dataset of mobile phone numbers of the short-term migrant population within the water conservancy warning area by combining basic information of mobile phone users.

[0128] The temporary transit population analysis module is based on the real-time mobile signaling dataset of the water conservancy warning area and the real-time mobile signaling dataset of the adjacent area of ​​the water conservancy warning area. It calculates and analyzes the temporary transit mobile phone user groups located in the water conservancy warning area and those about to enter the water conservancy emergency area. Combined with the basic information of mobile phone users, it generates a dataset of mobile phone numbers of the temporary transit population located in the water conservancy warning area and those about to enter the water conservancy warning area.

[0129] The early warning information push module automatically pushes early warning information to the relevant permanent residents, short-term floating populations, and temporary transit populations in the water conservancy early warning area via SMS or voice call, based on preset early warning information and set parameters, and tracks the completion of early warning information push.

[0130] The real-time population monitoring and analysis module monitors and analyzes the population size and spatial distribution within the water conservancy early warning area in real time, and displays the analysis results and the completion status of early warning SMS push in the form of visual charts.

[0131] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to specific implementations. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for targeted delivery of water conservancy early warning information based on mobile phone signaling big data, characterized in that, Includes the following steps: S1. Based on the early warning requirements of the water conservancy scenario, generate the geographical spatial range of the water conservancy early warning area and the surrounding adjacent area; S2. Based on the geographic spatial range of the water conservancy early warning area and the surrounding adjacent areas, and mobile communication base station data, obtain the base station number set data of the water conservancy early warning area and the base station number set data of the adjacent areas of the water conservancy early warning area, respectively. S3. Based on the base station number set data of the water conservancy early warning area, obtain the historical mobile phone signaling raw data related to the base station in the most recent period, preprocess the historical mobile phone signaling raw data, and generate the historical mobile phone signaling dataset of the water conservancy early warning area. S4. Based on the historical mobile signaling dataset of the water conservancy warning area, use the population analysis model to identify two types of mobile phone users related to the water conservancy warning area: permanent residents and short-term mobile users. The permanent residents mobile phone user group includes mobile phone users whose residence or workplace is located in the water conservancy warning area and whose commuting route passes through the water conservancy warning area. S5. Based on the base station number set data of the water conservancy early warning area and the base station number set data of the adjacent area of ​​the water conservancy early warning area, respectively obtain the real-time mobile phone signaling raw data related to the base station, preprocess the real-time mobile phone signaling raw data, and generate the real-time mobile phone signaling dataset of the water conservancy early warning area and the real-time mobile phone signaling dataset of the adjacent area of ​​the water conservancy early warning area. S6. Based on the real-time mobile signaling dataset of the water conservancy warning area, use a population analysis model to identify the temporary transit mobile phone user group within the water conservancy warning area; based on the real-time mobile signaling dataset of the neighboring areas of the water conservancy warning area, use a population travel trajectory analysis and prediction model to identify the temporary transit mobile phone user group about to enter the water conservancy warning area. S7. Based on preset warning rules, automatically push warning information to three categories of mobile phone users—residents, short-term migrants, and temporary transit users—via SMS or voice call, and track the completion of warning information push. S8. Monitor and analyze the population size and spatial distribution within the water conservancy early warning area in real time, and display the analysis results and early warning information push completion status in the form of visual charts.

2. The method for targeted push of water conservancy early warning information based on mobile phone signaling big data according to claim 1, characterized in that, The steps of step S1 are as follows: S101. Generate the geospatial extent of the water conservancy early warning area using one of the following four methods: manual drawing, buffer analysis, selecting administrative regions, or adding vector files. Manual drawing is based on manually drawing rectangular or polygonal areas on a map to generate the geographic spatial range of the water conservancy early warning area; Buffer analysis involves drawing point, line, and polygon features on a map and automatically generating buffer zones based on buffer radii to create the geographic spatial extent of water conservancy early warning areas. Selecting an administrative region involves choosing the name of the province, city, district / county, street / township, community / village, and using the corresponding administrative boundaries as the geographical spatial range of the water conservancy early warning area. Adding vector files involves uploading vector files or text files containing continuous coordinate values ​​to generate the geospatial extent of the water conservancy early warning area. S102. Perform GIS buffer analysis on the geographic spatial range of the water conservancy early warning area, set the buffer radius, and generate the geographic spatial range of the adjacent area outside the water conservancy early warning area.

3. The method for targeted push of water conservancy early warning information based on mobile phone signaling big data according to claim 1, characterized in that, The steps of step S2 are as follows: S201. Using the signal coverage radius of the mobile communication base station as the buffer radius, perform GIS buffer analysis on the vector points of the mobile communication base station to generate the signal coverage area vector data of the mobile communication base station. S202. Perform GIS spatial overlay analysis on the geographic spatial range of the water conservancy early warning area and the mobile communication base station data to obtain the mobile communication base stations related to the water conservancy early warning area and generate a set of base station number data for the water conservancy early warning area. S203. Perform GIS spatial overlay analysis on the geographic spatial range and mobile communication base station data of the adjacent area outside the water conservancy early warning area to obtain the mobile communication base stations in the adjacent area of ​​the water conservancy early warning area and generate a set of base station number data for the adjacent area of ​​the water conservancy early warning area.

4. The method for targeted push of water conservancy early warning information based on mobile phone signaling big data according to claim 1, characterized in that, The steps of step S3 are as follows: S301. Using the set of base station numbers in the water conservancy early warning area as the query condition, obtain the historical mobile phone signaling data related to the base station within the most recent time period; wherein the mobile phone signaling data contains the mobile user identification code, signaling timestamp, and the encoding information of the current base station; S302. Using a distributed computing framework, historical mobile phone signaling data is sorted according to mobile user identification code and signaling timestamp. Preprocessing operations are performed, including removing non-personal SIM cards, deduplication of multiple SIM cards per person, removal of duplicate time points, and handling of abnormal data, to generate a historical mobile phone signaling dataset P for the water conservancy early warning area. Specific information is as follows: P = {p1, p2, ..., p} i ,p n |1≤i≤n}, where p i Let n be the mobile signaling dataset for the i-th mobile phone user, where n is the number of mobile phone users. p i ={q1,q2,…,q j ,q m |1≤j≤m}, where m is the number of mobile signaling data, q j This refers to the j-th mobile signaling data in the mobile signaling dataset; q j = {d, t, s}, where d is the mobile subscriber identification code, t is the timestamp, and s is the base station code.

5. The method for targeted push of water conservancy early warning information based on mobile phone signaling big data according to claim 1, characterized in that, The steps of step S4 are as follows: S401. Based on the historical mobile signaling dataset of the water conservancy early warning area, obtain the entire activity chain of each mobile phone user, calculate the cumulative daytime stay value of the same mobile phone user on weekdays, add mobile phone users whose cumulative daytime stay value on weekdays reaches the threshold for daytime stay value to the working population seed set; add mobile phone users whose frequency of appearance in the working population seed set reaches the threshold for working stay days to the set of mobile phone users whose workplace is located in the water conservancy early warning area. S402. Based on the historical mobile phone signaling dataset of the water conservancy early warning area, obtain the entire activity chain of each mobile phone user, calculate the cumulative nighttime stay value of the same mobile phone user, add mobile phone users whose cumulative nighttime stay value reaches the threshold of nighttime stay value to the seed set of residents; add mobile phone users whose number of occurrences in the seed set of residents reaches the threshold of number of days of residence to the set of mobile phone users whose residence is located in the water conservancy early warning area. S403. Based on the historical mobile signaling dataset of the water conservancy early warning area, remove the mobile signaling data of mobile phone users whose residence or workplace is located in the water conservancy early warning area, and generate a temporary mobile signaling dataset. Based on the temporary mobile signaling dataset, the entire activity chain of each mobile user is obtained, the number of days the same mobile user appears during the weekday commuting period is calculated, and mobile users whose number of days appearing during the weekday commuting period reaches the weekday commuting day threshold are added to the set of mobile users commuting through the water conservancy warning area. S404. Combine the sets of mobile phone users whose residences and workplaces are located in the water conservancy warning area and those who commute through the water conservancy warning area, and combine them with the basic information of mobile phone users to generate a dataset of mobile phone numbers of permanent residents in the water conservancy warning area; the basic information of mobile phone users includes mobile user identification code, mobile phone number, age, and gender; S405. For mobile phone users who are elderly, they are marked as key groups of concern in the resident population mobile phone number dataset by using labels. S406. Based on the historical mobile signaling dataset of the water conservancy warning area, obtain the entire activity chain of each mobile phone user, calculate the daily cumulative stay time value of the same mobile phone user, and add mobile phone users whose daily cumulative stay time value reaches the daily stay time threshold of the floating population but is less than the daily stay time threshold of the permanent population to the floating population seed set; add mobile phone users whose frequency of appearance in the floating population seed set is less than the floating population stay days threshold and who are still in the water conservancy warning area on the most recent day to the short-term floating population mobile phone user set of the water conservancy warning area, and generate a short-term floating population mobile phone number dataset of the water conservancy warning area by combining the basic information of the mobile phone users.

6. The method for targeted push of water conservancy early warning information based on mobile phone signaling big data according to claim 1, characterized in that, The steps of step S5 are as follows: S501: Using the base station number set of the water conservancy early warning area and the base station number set of the adjacent area of ​​the water conservancy early warning area as query conditions, the real-time mobile phone signaling data of the water conservancy early warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy early warning area are automatically obtained at fixed time intervals. S502. Using a distributed computing framework, the real-time mobile phone signaling data of the water conservancy early warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy early warning area are sorted according to the mobile user identification code and the signaling timestamp. After performing operations such as removing non-personal SIM cards, deduplication of multiple SIM cards for one person, removal of duplicate time point data, and abnormal data processing, the real-time mobile phone signaling dataset of the water conservancy early warning area and the real-time mobile phone signaling dataset of the adjacent area of ​​the water conservancy early warning area are generated.

7. The method for targeted push of water conservancy early warning information based on mobile phone signaling big data according to claim 1, characterized in that, The steps of step S6 are as follows: S601. Based on the real-time mobile signaling dataset of the water conservancy early warning area, combined with the basic information of mobile phone users, obtain the seed set of mobile phone numbers of temporary transit population in the water conservancy early warning area, remove the mobile phone numbers of the permanent population and short-term floating population mobile phone users in S4, and then remove the mobile phone numbers that have not appeared for the first time, and generate the mobile phone number dataset of temporary transit population in the water conservancy early warning area. S602. Based on the real-time mobile phone signaling dataset of the adjacent area of ​​the water conservancy warning area, and using the population travel trajectory analysis and prediction model, obtain the mobile phone user group that is about to enter the water conservancy warning area, exclude mobile phone users that are not appearing for the first time, and generate a temporary transit population mobile phone number dataset that is about to enter the water conservancy warning area.

8. The method for targeted push of water conservancy early warning information based on mobile phone signaling big data according to claim 1, characterized in that, The steps of step S7 are as follows: S701. Based on the specific circumstances of water conservancy emergencies, early warning information shall be set in advance for the permanent residents, short-term floating population and temporary transit population in the water conservancy early warning area, and parameters such as start time, push interval and validity period shall be set. S702. Based on preset warning rules, automatically push warning information to the mobile phone number datasets of permanent residents, short-term migrants, and temporary transit residents in the water conservancy warning area by sending warning text messages or making voice calls; for mobile phone users marked as "key attention groups" in the mobile phone number dataset of permanent residents, push warning information by making voice calls as needed. S703. Mark the mobile phone users whose warning information was successfully pushed and whose warning information was not pushed, and generate a dataset of mobile phone numbers whose warning information was successfully pushed and a dataset of mobile phone numbers whose warning information was not pushed. S704. Resend the warning information to the mobile phone numbers where the warning information failed to be sent, and proceed to step S703 until all relevant mobile phone users in the water conservancy warning area have successfully received the warning information.

9. The method for targeted push of water conservancy early warning information based on mobile phone signaling big data according to claim 1, characterized in that, The steps of step S8 are as follows: S801: Based on the mobile phone number datasets of permanent residents, short-term migrants, and temporary transit residents in the water conservancy early warning area, count the number of permanent residents, short-term migrants, and temporary transit residents related to the water conservancy early warning area, as well as the number of successful and failed early warning information pushes; S802: Spatially locate the base station on the map, and generate corresponding population spatial distribution heat maps by using the number of residents, working people, commuters, short-term migrants, and temporary transit passengers at the base station as indicators.

10. A targeted push system for water conservancy early warning information based on mobile phone signaling big data, characterized in that: include: The water conservancy early warning area generation module generates the geographic spatial range of the water conservancy early warning area and its surrounding neighboring areas by manually drawing, buffer analysis, selecting administrative regions, and adding vector files. It also filters and obtains the base station number set data of the water conservancy early warning area and the base station number set data of the water conservancy early warning area and its neighboring areas. The mobile signaling data acquisition and processing module uses the base station number set data of the water conservancy early warning area and the specified time as conditions to obtain the original mobile signaling data related to the base stations in the water conservancy early warning area, and performs data storage and data processing operations through a distributed computing framework to generate a mobile signaling dataset of the water conservancy early warning area. The resident population analysis module identifies the residential population, working population, and commuting population related to the water conservancy warning area based on historical mobile phone signaling datasets within a certain time period. Combined with basic mobile phone user information, it generates a dataset of mobile phone numbers of the resident population in the water conservancy warning area. The short-term migrant population analysis module identifies the short-term migrant population within the water conservancy warning area based on historical mobile phone signaling datasets over a certain period of time. Combined with basic mobile phone user information, it generates a dataset of mobile phone numbers of the short-term migrant population within the water conservancy warning area. The temporary transit population analysis module, based on the real-time mobile phone signaling dataset of the water conservancy warning area and the real-time mobile phone signaling dataset of the adjacent area of ​​the water conservancy warning area, calculates and analyzes the temporary transit mobile phone user groups located in the water conservancy warning area and about to enter the water conservancy emergency area, respectively. Combined with the basic information of mobile phone users, it generates a dataset of mobile phone numbers of the temporary transit population located in the water conservancy warning area and about to enter the water conservancy warning area. The early warning information push module automatically pushes early warning information to the relevant permanent residents, short-term floating populations, and temporary transit populations in the water conservancy early warning area via SMS or voice call, based on the preset early warning information and set parameters, and tracks the completion of the early warning information push. The real-time population monitoring and analysis module monitors and analyzes the population size and spatial distribution within the water conservancy early warning area in real time, and displays the analysis results and the completion status of early warning SMS push in the form of visual charts.

Images