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

Abstract

The invention belongs to the technical field of water conservancy early warning and big data processing, and particularly relates to a water conservancy early warning information targeted pushing method and system based on mobile phone signaling big data, and the method comprises the steps: generating the geographic space range of a water conservancy early warning region and a peripheral adjacent region; obtaining base stations of a water conservancy early warning area and a peripheral adjacent area; acquiring a historical mobile phone signaling data set related to the base station in a recent time period, and identifying two types of mobile phone user groups of resident population and short-term floating population in the water conservancy early warning area; acquiring a real-time mobile phone signaling data set related to a water conservancy early warning area and a peripheral adjacent area base station, and identifying temporary transit mobile phone user groups in the water conservancy early warning area and about to enter the area; according to a preset early warning rule, early warning information is automatically pushed to the three types of mobile phone user groups in a classified mode; and the population number and the spatial distribution condition in the water conservancy early warning area are monitored in real time. According to the invention, the accuracy and timeliness of water conservancy early warning information pushing are further improved, and the applicability is wider.

Description

Technical Field

[0001] The present invention belongs to the technical field of water conservancy early warning and big data processing, and specifically relates to a method and system for targeted push of water conservancy early warning information based on mobile phone signaling big data. Background Art

[0002] Flood disasters, debris flows, dam breaches, water pollution, and flood discharges in water conservancy are characterized by suddenness, unpredictability, and irregularity. It is of great significance to release early warning information to the public in order to ensure public safety. Traditional ways of releasing early warning information include text messages, television stations, radio stations, and the Internet. However, these methods cannot accurately obtain the scope of early warning information release and the people involved. They only consider local residents, resulting in the omission of foreign tourists, visitors, and temporary transit groups. They fail to cover all people who may be affected, and there are blind spots in the release of early warning information. At the same time, these early warning information release methods have caused many irrelevant people to receive early warning information, resulting in a waste of SMS resources and unnecessary social panic.

[0003] Mobile phone signaling data is the time and location information captured and recorded by the communication base station when the mobile phone user makes a call, sends a text message or moves. It has the advantages of high temporal and spatial accuracy, wide coverage of the population, stability and reliability, and timely updates. Mobile phone signaling data is highly correlated with the temporal and spatial distribution of human activities and is widely used in various fields of natural resources and social development. Mobile phone signaling data is used in the field of water conservancy. After searching:

[0004] A Chinese patent document with publication number CN111711920B discloses a method for constructing an urban rainstorm and flood warning scheme based on mobile phone signaling data. The method uses mobile phone signaling data to identify the user's residence, work place and weekend travel place, and then combines it with a rainstorm spatial distribution map to identify mobile phone users in the rainstorm and flood danger area, and sends rainstorm warning information to the mobile phone users;

[0005] A Chinese patent document with publication number CN117576876B discloses an urban flood disaster warning method for potential people at risk, which uses flood risk data and mobile phone signaling data to identify people at risk and issue warning information to such people. However, the above research only uses mobile phone signaling data for flood disaster scenarios, and cannot be used for dangerous water conservancy emergency scenarios such as mudslides, dam breaches, water pollution, and flood discharge, so the scope of application is limited.

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

[0007] The purpose of the present invention is to overcome the above-mentioned problems existing in traditional technologies, and to provide a method and system for targeted push of water conservancy warning information based on mobile phone signaling big data, which is suitable for most water conservancy warning scenarios, can accurately locate people in the water conservancy warning area and people who are about to go to the area, and accurately push warning information according to different groups of people, thereby significantly improving the accuracy and effectiveness of water conservancy warning information push.

[0008] In order to achieve the above technical objectives and the above technical effects, the present invention is implemented through the following technical solutions:

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

[0010] S1. Generate the geographical space scope of the water conservancy warning area and the surrounding adjacent areas according to the warning needs of the water conservancy scenario;

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

[0012] S3. According to the base station number set data of the water conservancy warning area, the original data of historical mobile phone signaling related to the base station in the recent period is obtained, and the original data of historical mobile phone signaling is preprocessed to generate a historical mobile phone signaling data set of the water conservancy warning area;

[0013] S4. Based on the historical mobile phone signaling data set of the water conservancy warning area, the population analysis model is used to identify two types of mobile phone user groups related to the water conservancy warning area: the permanent population and the short-term floating population. The permanent population mobile phone user group includes mobile phone users whose residence or work place is located in the water conservancy warning area and whose commuting routes pass through the water conservancy warning area;

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

[0015] S6. Based on the real-time mobile phone signaling data set of the water conservancy warning area, a population analysis model is used to identify the temporary transit mobile phone user group in the water conservancy warning area; based on the real-time mobile phone signaling data set of the area adjacent to the water conservancy warning area, a population travel trajectory analysis and prediction model is used to identify the temporary transit mobile phone user group that is about to enter the water conservancy warning area;

[0016] S7. According to the preset warning rules, the warning information is automatically pushed to three types of mobile phone users, namely, permanent residents, short-term floating population, and temporary transit population, by SMS or voice calls, and the completion of the warning information push is tracked;

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

[0018] Furthermore, the sub-steps of step S1 are as follows:

[0019] S101. Generate the geographic spatial scope of the water conservancy warning area by one of the four methods: manual drawing, buffer zone analysis, selecting administrative areas, and adding vector files, where:

[0020] Manual drawing is to draw a rectangular or polygonal area manually based on the map to generate the geographical spatial scope of the water conservancy warning area;

[0021] Buffer analysis is to draw points, lines, and surface elements on the map, automatically generate buffers based on the buffer radius, and generate the geographic spatial scope of the water conservancy warning area;

[0022] The selected administrative area is to select the name of the province, city, district, county, street town, community and village, and use the corresponding administrative boundary as the geographical space scope of the water conservancy warning area;

[0023] Adding vector files is to generate the geographic spatial scope of the water conservancy warning area by uploading vector files in formats such as shp, json, kml, etc. or text files containing continuous coordinate values;

[0024] S102. Perform GIS buffer zone analysis on the geographic space scope of the water conservancy warning area, set a buffer radius, and generate the geographic space scope of the peripheral adjacent area of ​​the water conservancy warning area.

[0025] Furthermore, the sub-steps of step S2 are as follows:

[0026] S201, taking the signal coverage radius of the mobile communication base station as the buffer radius, performing GIS buffer analysis on the mobile communication base station vector points to generate planar vector data of the mobile communication base station signal coverage;

[0027] S202, performing GIS spatial overlay analysis on the geographic space scope of the water conservancy warning area and the mobile communication base station data, obtaining the mobile communication base stations related to the water conservancy warning area, and generating the base station number set data of the water conservancy warning area;

[0028] S203, perform GIS spatial overlay analysis on the geographic space scope and mobile communication base station data of the peripheral neighboring area of ​​the water conservancy warning area, obtain the mobile communication base stations in the neighboring area of ​​the water conservancy warning area, and generate base station number set data of the neighboring area of ​​the water conservancy warning area.

[0029] Furthermore, the sub-steps of step S3 are as follows:

[0030] S301, using the base station number set in the water conservancy early warning area as a query condition, obtaining historical mobile phone signaling data related to the base station within a certain period of time in the recent period; wherein the mobile phone signaling data contains a mobile user identification code, a signaling timestamp, and coding information of the current base station;

[0031] S302, using a distributed computing framework, sort the historical mobile phone signaling data according to the mobile user identification code and signaling timestamp, remove the non-personal number cards, remove the duplicates of multiple cards for one person, remove the repeated time point data, and perform abnormal data processing pre-processing operations to generate a historical mobile phone signaling data set P for the water conservancy warning area. The specific information is:

[0032] P={p1,p2,…,p i ,p n |1≤i≤n}, where p i is the mobile phone signaling data set of the i-th mobile phone user, and 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 phone signaling data, q j is the jth mobile phone signaling data in the mobile phone signaling data set;

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

[0035] Furthermore, the sub-steps of step S4 are as follows:

[0036] S401. Based on the historical mobile phone signaling data set of the water conservancy warning area, the whole process activity chain of each mobile phone user is obtained, and the cumulative daytime stay time value of the same mobile phone user on working days is calculated. The mobile phone users whose cumulative daytime stay time value on working days reaches the single-day daytime stay time threshold are added to the working population seed set; if the number of times the same mobile phone user appears in the working population seed set reaches the working stay days threshold, the mobile phone user is added to the set of mobile phone users whose work place is located in the water conservancy warning area;

[0037] S402. Based on the historical mobile phone signaling data set of the water conservancy warning area, the whole process activity chain of each mobile phone user is obtained, and the cumulative nighttime stay time value of the same mobile phone user in a single day is calculated. The mobile phone users whose cumulative nighttime stay time value in a single day reaches the nighttime stay time threshold in a single day are added to the resident population seed set; if the number of times the same mobile phone user appears in the resident population seed set reaches the resident stay days threshold, the mobile phone user is added to the set of mobile phone users whose residence is located in the water conservancy warning area;

[0038] S403, based on the historical mobile phone signaling data set of the water conservancy warning area, the mobile phone signaling data of the mobile phone users whose residence or work place is located in the water conservancy warning area are eliminated to generate a temporary mobile phone signaling data set; based on the temporary mobile phone signaling data set, the whole process activity chain of each mobile phone user is obtained, and the number of days that the same mobile phone user appears in the commuting period on weekdays is calculated, and if the number of days that the same mobile phone user appears in the commuting period on weekdays reaches the threshold of the number of commuting days on weekdays, the mobile phone user is added to the set of mobile phone users commuting through the water conservancy warning area;

[0039] S404, merging a set of mobile phone users whose residence or work place is in the water conservancy warning area and whose commute passes through the water conservancy warning area, and combining the basic information of the mobile phone users, to generate a mobile phone number dataset of the permanent population in the water conservancy warning area; wherein the basic information of the mobile phone users includes mobile user identification code, mobile phone number, age, and gender;

[0040] S405: If the mobile phone user is elderly, they are marked with a label in the resident population mobile phone number dataset and marked as a key focus group;

[0041] S406. Based on the historical mobile phone signaling dataset of the water conservancy warning area, the full-process activity chain of each mobile phone user is obtained, and the daily cumulative stay time value of the same mobile phone user is calculated. Mobile phone users whose daily cumulative stay time value reaches the daily stay time threshold of the floating population and is less than the daily stay time threshold of the resident population are added to the floating population seed set; if the same mobile phone user appears in the floating population seed set less than the floating population stay days threshold and is still in the water conservancy warning area in the most recent day, he is added to the short-term floating population mobile phone user set in the water conservancy warning area, and combined with the basic information of the mobile phone user, a short-term floating population mobile phone number dataset in the water conservancy warning area is generated.

[0042] Furthermore, the sub-steps of step S5 are as follows:

[0043] S501, using the base station number set of the water conservancy warning area and the base station number set of the adjacent area of ​​the water conservancy warning area as query conditions, automatically acquiring the real-time mobile phone signaling data of the water conservancy warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy warning area at fixed time intervals;

[0044] S502. Using a distributed computing framework, the real-time mobile phone signaling data of the water conservancy warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy warning area are sorted according to the mobile user identification code and the signaling timestamp, and after removing the non-personal number cards, removing the duplicate time point data, and performing abnormal data processing operations, a real-time mobile phone signaling data set of the water conservancy warning area and a real-time mobile phone signaling data set of the adjacent area of ​​the water conservancy warning area are generated.

[0045] Further, the sub-steps of step S6 are as follows:

[0046] S601. Based on the real-time mobile phone signaling data set of the water conservancy 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 warning area, remove the mobile phone numbers of the resident population and short-term floating population mobile phone user groups in S4, and then remove the mobile phone numbers that do not appear for the first time, and generate the mobile phone number data set of temporary transit population in the water conservancy warning area;

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

[0048] Furthermore, the population travel trajectory analysis and prediction model in step S602 is based on the population travel historical trajectory data and road network vector data extracted from mobile phone signaling data, and analyzes the population travel direction, speed and trajectory-road correlation through deep learning technology, predicts the population's future travel trajectory, and obtains the probability of mobile phone users in the vicinity of the water conservancy warning area entering 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, the sub-steps of step S7 are as follows:

[0050] S701. According to the specific situation of water conservancy emergency, classify and set early warning information for the permanent population, short-term floating population, and temporary transit population in the water conservancy early warning area in advance, and set the start push time, push interval, and validity period parameters;

[0051] S702. According to the preset warning rules, the warning information is automatically pushed to the mobile phone number dataset of the permanent population, the mobile phone number dataset of the short-term floating population, and the mobile phone number dataset of the temporary transit population in the water conservancy warning area by sending warning text messages or making voice calls; for the mobile phone users marked as "key focus groups" in the mobile phone number dataset of the permanent population, the warning information is pushed by making voice calls according to actual needs;

[0052] S703, marking the mobile phone users to which the warning information is pushed successfully and the mobile phone users to which the warning information is pushed unsuccessfully, and generating a data set of mobile phone numbers to which the warning information is pushed successfully and a data set of mobile phone numbers to which the warning information is pushed unsuccessfully respectively;

[0053] S704: Re-send the warning information to the mobile phone numbers to which the warning information has failed to be pushed, and proceed to step S703 until all relevant mobile phone users in the water conservancy warning area have successfully received the warning information.

[0054] Further, the sub-steps of step S8 are as follows:

[0055] S801: Based on the mobile phone number dataset of permanent residents, short-term floating population and temporary transit population in the water warning area, count the number of permanent residents, short-term floating population and temporary transit population in the water warning area and the number of successful and failed warning information pushes;

[0056] S802: The base stations are spatially located on the map, and the number of residents, working population, commuting population, short-term floating population, and temporary transit population at the base stations are used as index values, and corresponding population spatial distribution heat maps are generated through GIS spatial interpolation.

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

[0058] The water conservancy warning area generation module generates the geographical space scope of the water conservancy warning area and the surrounding adjacent areas by manual drawing, buffer zone analysis, selection of administrative areas, and adding vector files, and obtains the base station number set data of the water conservancy warning area and the base station number set data of the adjacent areas of the water conservancy warning area by screening;

[0059] The mobile phone signaling data collection and processing module obtains the original mobile phone signaling data related to the base stations in the water conservancy warning area based on the base station number set data and the specified time in the water conservancy warning area, and performs data storage and data processing operations through the distributed computing framework to generate the mobile phone signaling data set of the water conservancy warning area;

[0060] The permanent population analysis module identifies the residents, workers, and commuters in the water conservancy warning area based on the historical mobile phone signaling data set within a certain period of time, and generates a mobile phone number data set of the permanent population in the water conservancy warning area based on the basic information of mobile phone users;

[0061] The short-term floating population analysis module identifies the short-term floating population in the water conservancy warning area based on the historical mobile phone signaling data set within a certain period of time, and generates a mobile phone number data set of the short-term floating population in the water conservancy warning area based on the basic information of mobile phone users;

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

[0063] The early warning information push module automatically pushes early warning information to the permanent population, short-term floating population, and temporary transit population related to the water conservancy early warning area through SMS or voice calls according to the preset early warning information and setting 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 in the water conservancy warning area in real time, and displays the analysis results and the completion status of warning SMS push in the form of visual charts.

[0065] The beneficial effects of the present invention are:

[0066] 1. Based on mobile phone signaling big data and GIS technology, the present invention accurately identifies the resident population, working population, commuting population, and temporary transit population related to the water conservancy emergency area, and automatically classifies and pushes warning information. Specifically, the population in the water conservancy warning area is accurately identified and classified, and different warning information is pushed to different mobile phone user groups, thereby improving the accuracy and timeliness of water conservancy warnings. The present invention further improves water conservancy safety and reduces the cost, manpower, and material costs of pushing warning information by text message to a certain extent.

[0067] 2. The present invention can be used in various water conservancy scenarios such as flood disasters, mudslides, dam breaches, water pollution, and flood discharge, and has wide applicability.

[0068] 3. The present invention identifies mobile phone users who are about to enter the water conservancy warning area through a population travel trajectory analysis and prediction model, and pushes warning information in advance to avoid or reduce personnel safety issues caused by entering the water conservancy warning area.

[0069] 4. The present invention identifies the elderly in the water conservancy warning area, achieves focused attention on this key group of people, and reduces the risks faced by this group of people.

[0070] Of course, any product implementing the present invention does not necessarily need to achieve all of the above advantages at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

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

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

[0074] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] Embodiment 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: Generate the geographical space range of the water conservancy warning area and the surrounding adjacent areas according to the warning requirements of the water conservancy scene. The specific steps are as follows:

[0078] S101. Generate the geographic spatial scope of the water conservancy warning area by one of the four methods: manual drawing, buffer zone analysis, selecting administrative areas, and adding vector files. Specifically:

[0079] The manual drawing is based on manually drawing a rectangular or polygonal area on a map to generate the geographical space scope of the water conservancy warning area;

[0080] The buffer zone analysis is to draw points, lines and surface elements on the map, automatically generate buffer zones according to the buffer radius, and generate the geographic spatial scope of the water conservancy warning area;

[0081] The selected administrative area is to select the name of the province, city, district, county, street town, community village, and use the corresponding administrative boundary as the geographical space scope of the water conservancy warning area;

[0082] The adding of vector files is to generate the geographic space scope of the water conservancy warning area by uploading vector files in formats such as shp, json, kml, etc. or text files containing continuous coordinate values.

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

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

[0085] S201, taking the signal coverage radius of the mobile communication base station as the buffer radius, performing GIS buffer analysis on the mobile communication base station vector points to generate planar vector data of the mobile communication base station signal coverage;

[0086] S202, performing GIS spatial overlay analysis on the geographic space scope of the water conservancy warning area and the mobile communication base station data, obtaining the mobile communication base stations related to the water conservancy warning area, and generating the base station number set data of the water conservancy warning area;

[0087] S203, performing GIS spatial overlay analysis on the geographic space scope and mobile communication base station data of the peripheral neighboring area of ​​the water conservancy warning area, obtaining the mobile communication base stations in the neighboring area of ​​the water conservancy warning area, and generating base station number set data of the neighboring area of ​​the water conservancy warning area.

[0088] Step S3, according to the base station number set data of the water conservancy early warning area, obtain the original data of historical mobile phone signaling related to the base station within a certain period, pre-process the original data of historical mobile phone signaling, and generate a historical mobile phone signaling data set of the water conservancy early warning area, the specific steps are as follows:

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

[0090] S302, using Hadoop+Spark to build a distributed computing framework, sort the historical mobile phone signaling data according to the mobile user identification code and signaling timestamp, perform pre-processing operations such as removing non-personal number cards, removing duplicates of multiple cards for one person, removing duplicate time point data, and abnormal data processing, and generate a historical mobile phone signaling data set P for the water conservancy warning area. The specific information is:

[0091] P={p1,p2,…,p i ,pn |1≤i≤n}, where p i is the mobile phone signaling data set of the i-th mobile phone user, and 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 phone signaling data, q j is the jth mobile phone signaling data in the mobile phone signaling data set;

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

[0094] In this embodiment, the removal of non-personal number cards in step S302 is specifically as follows:

[0095] Using the one-to-one correspondence between the number card and the mobile user identification code, the mobile phone signaling data generated by non-personal number cards such as the Internet of Things card can be identified and eliminated;

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

[0097] Calculate the overlap of multiple mobile user identification codes, remove duplicates of multiple cards in the network, and remove the corresponding mobile phone signaling data;

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

[0099] The base station to which the user connects most times during the "ping-pong effect" is taken as the main base station, and the records other than the main base station in the signaling data generated during the "ping-pong effect" are removed to complete the data cleaning.

[0100] Step S4: Based on the historical mobile phone signaling data set of the water conservancy warning area, using the population analysis model, identify two types of mobile phone user groups related to the water conservancy warning area, namely, the permanent population and the short-term floating population. The permanent population mobile phone user group includes mobile phone users whose residence or work place is located in the water conservancy warning area and whose commuting routes pass through the water conservancy warning area. The specific steps are as follows:

[0101] S401. Based on the historical mobile phone signaling data set of the water conservancy warning area, the whole process activity chain of each mobile phone user is obtained, and the single-day cumulative daytime stay time value of the same mobile phone user in the time period of 9:00-16:30 on working days is calculated. Mobile phone users whose single-day daytime cumulative stay time value reaches the single-day daytime stay time threshold of 5 hours are added to the working population seed set; if the number of times the same mobile phone user appears in the working population seed set reaches the working stay day threshold of 6 days, the mobile phone user is added to the set of mobile phone users whose work place is located in the water conservancy warning area;

[0102] S402, based on the historical mobile phone signaling data set of the water conservancy warning area, calculate the cumulative night stay time of the same mobile phone user in the time period of 21:00-6:00, and add the mobile phone users whose cumulative night stay time reaches the threshold of 7 hours to the resident population seed set; if the number of times the same mobile phone user appears in the resident population seed set reaches the threshold of 6 days of residence, add him to the set of mobile phone users whose residence is located in the water conservancy warning area;

[0103] S403, based on the historical mobile phone signaling data set of the water conservancy warning area, the mobile phone signaling data of the mobile phone users whose residence or work place is located in the water conservancy warning area are eliminated to generate a temporary mobile phone signaling data set; based on the temporary mobile phone signaling data set, the number of days that the same mobile phone user appears in the time periods of 6:00-9:00 and 16:00-19:00 on weekdays is calculated, and if the number of days that the same mobile phone user appears in the commuting period on weekdays reaches the threshold of 6 days of commuting days on weekdays, the mobile phone user is added to the set of mobile phone users commuting through the water conservancy warning area;

[0104] S404. Combine the mobile phone users whose residence or work place is in the water warning area and whose commute passes through the water warning area, and combine the basic information of the mobile phone users to generate a mobile phone number dataset of the permanent population related to the water warning area.

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

[0106] S501, using the base station number set of the water conservancy warning area and the base station number set of the adjacent area of ​​the water conservancy warning area as query conditions, automatically acquiring the real-time mobile phone signaling data of the water conservancy warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy warning area in a 5-minute period;

[0107] S502. Utilize a distributed computing framework to sort the real-time mobile phone signaling data of the water conservancy warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy warning area according to the mobile user identification code and the signaling timestamp, remove non-personal number cards, remove duplicate time point data, process abnormal data, and generate a real-time mobile phone signaling dataset of the water conservancy warning area and a real-time mobile phone signaling dataset of the adjacent area of ​​the water conservancy warning area after performing operations such as removing non-personal number cards, removing duplicate time point data, and processing abnormal data.

[0108] Step S6: Based on the real-time mobile phone signaling data set of the water conservancy warning area, a population analysis model is used to identify the temporary transit mobile phone user group in the water conservancy warning area; based on the real-time mobile phone signaling data set of the adjacent area of ​​the water conservancy warning area, a population trajectory analysis prediction model is used to identify the temporary transit mobile phone user group that is about to enter the water conservancy warning area. The specific steps are as follows:

[0109] S601, based on the real-time mobile phone signaling data set of the water conservancy warning area, remove the mobile phone signaling data records of the mobile phone user groups of the permanent population and the short-term floating population in S4, and generate a mobile phone signaling data set of the temporary transit population in the water conservancy warning area;

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

[0111] S603, based on the real-time mobile phone signaling data set of the area adjacent to the water conservancy warning area, the mobile phone signaling data records of the mobile phone user groups that are not appearing for the first time are eliminated to generate a mobile phone signaling data set of the temporary transit population in the area adjacent to the water conservancy warning area;

[0112] S604. Based on the mobile phone signaling dataset of temporary transit population in the vicinity of the water conservancy warning area, a population travel trajectory analysis and prediction model is used to determine whether the mobile phone user will enter the water conservancy warning area, and a mobile phone number dataset of temporary transit population who are about to enter the water conservancy warning area is generated.

[0113] The population travel trajectory analysis and prediction model is built based on the population travel history trajectory data and road network vector data extracted from mobile phone signaling data, based on LSTM neural network technology. The model is used to analyze the population travel direction, speed and trajectory-road correlation, predict the population's future travel trajectory, obtain the probability of mobile phone users in the neighboring area entering the water conservancy warning area, and classify mobile phone users whose probability exceeds the threshold as mobile phone users who are about to enter the water conservancy warning area.

[0114] Step S7: According to the preset warning rules, the warning information is automatically pushed to three types of mobile phone user groups, namely, permanent population, short-term floating population, and temporary transit population, by SMS or voice call, and the completion of the warning information push is tracked. The specific steps are as follows:

[0115] S701. According to the specific situation of water conservancy emergency, early warning information is set for the permanent population, short-term floating population, and temporary transit population in the water conservancy early warning area in advance, and the parameters such as the start push time, push interval, and validity period are set respectively;

[0116] S702. According to the preset warning rules, the warning information is automatically pushed to the mobile phone number dataset of the permanent population, the mobile phone number dataset of the short-term floating population, and the mobile phone number dataset of the temporary transit population in the water conservancy warning area by sending warning text messages or making voice calls; for the mobile phone users marked as "key focus groups" in the mobile phone number dataset of the permanent population, the warning information is pushed by making voice calls according to actual needs;

[0117] S703, use 0 and 1 to mark the mobile phone users to which the warning information is pushed successfully and to which it fails, respectively, to generate a data set of mobile phone numbers to which the warning information is pushed successfully and a data set of mobile phone numbers to which the warning information is pushed failed;

[0118] S704: Re-send the warning information to the mobile phone number to which the warning information has failed to be pushed, and enter 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 and spatial distribution in the water conservancy warning area in real time, and display the analysis results and the completion of warning information push in the form of visual charts. The specific steps are as follows:

[0120] S801. Based on the mobile phone number dataset of the permanent population, the mobile phone number dataset of the short-term floating population and the mobile phone number dataset of the temporary transit population in the water conservancy warning area, count the number of permanent population, the number of short-term floating population, the number of temporary transit population related to the water conservancy warning area and the number of successful and failed warning information pushes;

[0121] S802. The base stations are spatially located on the map. The number of residents, working population, commuting population, short-term floating population, and temporary transit population at the base stations are used as index values. The corresponding population spatial distribution heat maps are generated through GIS spatial interpolation, and red, green, and blue colors are used to represent the population.

[0122] Embodiment 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] A water conservancy warning area generation module generates the geographical space scope of the water conservancy warning area and the surrounding areas of the water conservancy warning area by means of manual drawing, buffer zone analysis, selection of administrative areas, and addition of vector files, and obtains the base station number set data of the water conservancy warning area and the base station number set data of the surrounding areas of the water conservancy warning area by screening;

[0125] Mobile phone signaling data collection and processing module, which uses the base station number set data and specified time in the water conservancy warning area as conditions to obtain the original mobile phone signaling data related to the base stations in the water conservancy warning area, and performs data storage and data processing operations through a distributed computing framework to generate a mobile phone signaling data set for the water conservancy warning area;

[0126] Permanent population analysis module: This module identifies the resident population, working population, and commuting population related to the water conservancy warning area based on the historical mobile phone signaling data set within a certain period of time, and generates a mobile phone number data set of the permanent population in the water conservancy warning area based on the basic information of mobile phone users;

[0127] Short-term floating population analysis module, which identifies the short-term floating population in the water conservancy warning area based on the historical mobile phone signaling data set within a certain period of time, and generates a data set of mobile phone numbers of the short-term floating population in the water conservancy warning area based on the basic information of mobile phone users;

[0128] Temporary transit population analysis module: This module calculates and analyzes the temporary transit mobile phone user groups in the water conservancy warning area and the water conservancy emergency area 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. Combined with the basic information of mobile phone users, it generates a mobile phone number dataset of the temporary transit population in the water conservancy warning area and the water conservancy warning area.

[0129] Warning information push module, which automatically pushes warning information to the permanent population, short-term floating population, and temporary transit population related to the water conservancy warning area by SMS or voice call according to the preset warning information and setting parameters, and tracks the completion of warning information push;

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

[0131] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for targeted push of water conservancy early warning information based on mobile phone signaling big data, characterized in that: The steps include: S1. Generate the geographical space scope of the water conservancy warning area and the surrounding adjacent areas according to the warning needs of the water conservancy scenario; S2. Based on the geographic space range of the water conservancy warning area and the peripheral adjacent areas, and the mobile communication base station data, respectively obtain the base station number set data of the water conservancy warning area and the base station number set data of the adjacent areas of the water conservancy warning area; S3. According to the base station number set data of the water conservancy warning area, the original data of historical mobile phone signaling related to the base station in the recent period is obtained, and the original data of historical mobile phone signaling is preprocessed to generate a historical mobile phone signaling data set of the water conservancy warning area; S4. Based on the historical mobile phone signaling data set of the water conservancy warning area, the population analysis model is used to identify two types of mobile phone user groups related to the water conservancy warning area: the permanent population and the short-term floating population. The permanent population mobile phone user group includes mobile phone users whose residence or work place is located in the water conservancy warning area and whose commuting routes pass through the water conservancy warning area; S5. Based on the base station number set data of the water conservancy warning area and the base station number set data of the adjacent area of ​​the water conservancy warning area, respectively obtain the real-time mobile phone signaling original data related to the base station, pre-process the real-time mobile phone signaling original data, and generate the real-time mobile phone signaling data set of the water conservancy warning area and the real-time mobile phone signaling data set of the adjacent area of ​​the water conservancy warning area; S6. Based on the real-time mobile phone signaling data set of the water conservancy warning area, a population analysis model is used to identify the temporary transit mobile phone user group in the water conservancy warning area; based on the real-time mobile phone signaling data set of the area adjacent to the water conservancy warning area, a population travel trajectory analysis and prediction model is used to identify the temporary transit mobile phone user group that is about to enter the water conservancy warning area; S7. According to the preset warning rules, the warning information is automatically pushed to three types of mobile phone users, namely, permanent residents, short-term floating population, and temporary transit population, by SMS or voice calls, and the completion of the warning information push is tracked; S8. Monitor and analyze the population size and spatial distribution in the water conservancy warning area in real time, and display the analysis results and 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 is characterized in that: The sub-steps of step S1 are as follows: S101. Generate the geographic spatial scope of the water conservancy warning area by one of the four methods: manual drawing, buffer zone analysis, selecting administrative areas, and adding vector files, where: Manual drawing is to draw a rectangular or polygonal area manually based on the map to generate the geographical spatial scope of the water conservancy warning area; Buffer analysis is to draw points, lines, and surface elements on the map, automatically generate buffers based on the buffer radius, and generate the geographic spatial scope of the water conservancy warning area; The selected administrative area is to select the name of the province, city, district, county, street town, community and village, and use the corresponding administrative boundary as the geographical space scope of the water conservancy warning area; Adding a vector file is to generate the geographic spatial scope of the water conservancy warning area by uploading a vector file or a text file containing continuous coordinate values; S102. Perform GIS buffer zone analysis on the geographic space scope of the water conservancy warning area, set a buffer radius, and generate the geographic space scope of the peripheral adjacent area of ​​the water conservancy 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 is characterized in that: The sub-steps of step S2 are as follows: S201, taking the signal coverage radius of the mobile communication base station as the buffer radius, performing GIS buffer analysis on the mobile communication base station vector points to generate planar vector data of the mobile communication base station signal coverage; S202, performing GIS spatial overlay analysis on the geographic space scope of the water conservancy warning area and the mobile communication base station data, obtaining the mobile communication base stations related to the water conservancy warning area, and generating the base station number set data of the water conservancy warning area; S203, perform GIS spatial overlay analysis on the geographic space scope and mobile communication base station data of the peripheral neighboring area of ​​the water conservancy warning area, obtain the mobile communication base stations in the neighboring area of ​​the water conservancy warning area, and generate base station number set data of the neighboring area of ​​the water conservancy 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 is characterized in that: The sub-steps of step S3 are as follows: S301, using the base station number set in the water conservancy early warning area as a query condition, obtaining historical mobile phone signaling data related to the base station within a certain period of time in the recent period; wherein the mobile phone signaling data contains a mobile user identification code, a signaling timestamp, and coding information of the current base station; S302, using a distributed computing framework, sort the historical mobile phone signaling data according to the mobile user identification code and signaling timestamp, remove the non-personal number cards, remove the duplicates of multiple cards for one person, remove the repeated time point data, and perform abnormal data processing pre-processing operations to generate a historical mobile phone signaling data set P for the water conservancy warning area. The specific information is: P={p1,p2,…,p i ,p n |1≤i≤n}, where p i is the mobile phone signaling data set of the i-th mobile phone user, and 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 phone signaling data, q j is the jth mobile phone signaling data in the mobile phone signaling data set; q j ={d,t,s}, where d is the mobile user 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 is characterized in that: The sub-steps of step S4 are as follows: S401. Based on the historical mobile phone signaling data set of the water conservancy warning area, the whole process activity chain of each mobile phone user is obtained, and the cumulative daytime stay time value of the same mobile phone user on working days is calculated. The mobile phone users whose cumulative daytime stay time value on working days reaches the single-day daytime stay time threshold are added to the working population seed set; if the number of times the same mobile phone user appears in the working population seed set reaches the working stay days threshold, the mobile phone user is added to the set of mobile phone users whose work place is located in the water conservancy warning area; S402. Based on the historical mobile phone signaling data set of the water conservancy warning area, the whole process activity chain of each mobile phone user is obtained, and the cumulative nighttime stay time value of the same mobile phone user in a single day is calculated. The mobile phone users whose cumulative nighttime stay time value in a single day reaches the nighttime stay time threshold in a single day are added to the resident population seed set; if the number of times the same mobile phone user appears in the resident population seed set reaches the resident stay days threshold, the mobile phone user is added to the set of mobile phone users whose residence is located in the water conservancy warning area; S403, based on the historical mobile phone signaling data set of the water conservancy warning area, excluding the mobile phone signaling data of mobile phone users whose residence or work place is located in the water conservancy warning area, to generate a temporary mobile phone signaling data set; Based on the temporary mobile phone signaling data set, the full process activity chain of each mobile phone user is obtained, and the number of days that the same mobile phone user appears during the weekday commuting period is calculated. If the number of days that the same mobile phone user appears during the weekday commuting period reaches the weekday commuting day threshold, the user is added to the set of mobile phone users commuting through the water conservancy warning area; S404, merging a set of mobile phone users whose residence or work place is in the water conservancy warning area and whose commute passes through the water conservancy warning area, and combining the basic information of the mobile phone users, to generate a mobile phone number dataset of the permanent population in the water conservancy warning area; wherein the basic information of the mobile phone users includes mobile user identification code, mobile phone number, age, and gender; S405: If the mobile phone user is elderly, they are marked with a label in the resident population mobile phone number dataset and marked as a key focus group; S406. Based on the historical mobile phone signaling dataset of the water conservancy warning area, the full-process activity chain of each mobile phone user is obtained, and the daily cumulative stay time value of the same mobile phone user is calculated. Mobile phone users whose daily cumulative stay time value reaches the daily stay time threshold of the floating population and is less than the daily stay time threshold of the resident population are added to the floating population seed set; if the same mobile phone user appears in the floating population seed set less than the floating population stay days threshold and is still in the water conservancy warning area in the most recent day, he is added to the short-term floating population mobile phone user set in the water conservancy warning area, and combined with the basic information of the mobile phone user, a short-term floating population mobile phone number dataset in the water conservancy warning area is generated.

6. The method for targeted push of water conservancy early warning information based on mobile phone signaling big data according to claim 1 is characterized in that: The sub-steps of step S5 are as follows: S501, using the base station number set of the water conservancy warning area and the base station number set of the adjacent area of ​​the water conservancy warning area as query conditions, automatically acquiring the real-time mobile phone signaling data of the water conservancy warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy warning area at fixed time intervals; S502. Using a distributed computing framework, the real-time mobile phone signaling data of the water conservancy warning area and the real-time mobile phone signaling data of the adjacent area of ​​the water conservancy warning area are sorted according to the mobile user identification code and the signaling timestamp, and after removing the non-personal number cards, removing the duplicate time point data, and performing abnormal data processing operations, a real-time mobile phone signaling data set of the water conservancy warning area and a real-time mobile phone signaling data set of the adjacent area of ​​the water conservancy 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 is characterized in that: The sub-steps of step S6 are as follows: S601. Based on the real-time mobile phone signaling data set of the water conservancy 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 warning area, remove the mobile phone numbers of the resident population and short-term floating population mobile phone user groups in S4, and then remove the mobile phone numbers that do not appear for the first time, and generate the mobile phone number data set of temporary transit population in the water conservancy warning area; S602. Based on the real-time mobile phone signaling dataset of the area adjacent to the water conservancy warning area, a population travel trajectory analysis and prediction model is used to obtain the mobile phone user group that is about to enter the water conservancy warning area, and mobile phone users who are not appearing for the first time are eliminated to generate a mobile phone number dataset of the temporary transit population 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 is characterized in that: The sub-steps of step S7 are as follows: S701. According to the specific situation of water conservancy emergency, classify and set early warning information for the permanent population, short-term floating population, and temporary transit population in the water conservancy early warning area in advance, and set the start push time, push interval, and validity period parameters; S702. According to the preset warning rules, the warning information is automatically pushed to the mobile phone number dataset of the permanent population, the mobile phone number dataset of the short-term floating population, and the mobile phone number dataset of the temporary transit population in the water conservancy warning area by sending warning text messages or making voice calls; for the mobile phone users marked as "key focus groups" in the mobile phone number dataset of the permanent population, the warning information is pushed by making voice calls according to actual needs; S703, marking the mobile phone users to which the warning information is pushed successfully and the mobile phone users to which the warning information is pushed unsuccessfully, and generating a data set of mobile phone numbers to which the warning information is pushed successfully and a data set of mobile phone numbers to which the warning information is pushed unsuccessfully respectively; S704: Re-send the warning information to the mobile phone numbers to which the warning information has failed to be pushed, 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 is characterized in that: The sub-steps of step S8 are as follows: S801: Based on the mobile phone number dataset of permanent residents, short-term floating population and temporary transit population in the water warning area, count the number of permanent residents, short-term floating population and temporary transit population in the water warning area and the number of successful and failed warning information pushes; S802: The base stations are spatially located on the map, and the number of residents, working population, commuting population, short-term floating population, and temporary transit population at the base stations are used as index values, and corresponding population spatial distribution heat maps are generated through GIS spatial interpolation.

10. The targeted push system for water conservancy early warning information based on mobile phone signaling big data is characterized by: include: The water conservancy warning area generation module generates the geographical space scope of the water conservancy warning area and the surrounding adjacent areas by manual drawing, buffer zone analysis, selection of administrative areas, and adding vector files, and obtains the base station number set data of the water conservancy warning area and the base station number set data of the adjacent areas of the water conservancy warning area by screening; The mobile phone signaling data collection and processing module obtains the original mobile phone signaling data related to the base stations in the water conservancy warning area based on the base station number set data and the specified time in the water conservancy warning area, and performs data storage and data processing operations through the distributed computing framework to generate the mobile phone signaling data set of the water conservancy warning area; The permanent population analysis module identifies the residents, workers, and commuters in the water conservancy warning area based on the historical mobile phone signaling data set within a certain period of time, and generates a mobile phone number data set of the permanent population in the water conservancy warning area based on the basic information of mobile phone users; The short-term floating population analysis module identifies the short-term floating population in the water conservancy warning area based on the historical mobile phone signaling data set within a certain period of time, and generates a mobile phone number data set of the short-term floating population in the water conservancy warning area based on the basic information of mobile phone users; The temporary transit population analysis module calculates and analyzes the temporary transit mobile phone user groups in the water conservancy warning area and the water conservancy emergency area based on the real-time mobile phone signaling data set of the water conservancy warning area and the real-time mobile phone signaling data set of the adjacent area of ​​the water conservancy warning area. Combined with the basic information of mobile phone users, it generates a mobile phone number data set of the temporary transit population in the water conservancy warning area and the water conservancy warning area. The early warning information push module automatically pushes early warning information to the permanent population, short-term floating population, and temporary transit population related to the water conservancy early warning area through SMS or voice calls according to the preset early warning information and setting 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 in the water conservancy warning area in real time, and displays the analysis results and the completion status of warning SMS push in the form of visual charts.

Images