Geographic information management system and management method for emergency disposal

Abstract

The invention relates to the technical field of geographic information management, in particular to a geographic information management system for emergency disposal, which comprises an equipment end and a cloud platform, the equipment end collects data and transmits the data to the cloud platform; the cloud platform carries out storage and analysis according to the data collected by the equipment end, and the cloud platform comprises a data processing and analysis module, a data storage management module and an emergency alarm module; according to the invention, through real-time data acquisition and data processing and analysis, the fluctuation condition of indexes is monitored in real time, whether a critical event occurs under the current condition is judged in combination with a built-in threshold value alarm algorithm rule, and when the critical events such as leakage and fire occur, an emergency alarm is automatically sent to a related unit of a region where the system is located; therefore, response time of related units can be shortened, personnel life and property safety loss can be reduced, and emergency management department information in a certain range of a management area is daily updated, so that emergency alarm information can be accurately sent.

Description

Technical Field

[0001] This invention relates to the field of geographic information management technology, and in particular to a geographic information management system and management method for emergency response. Background Technology

[0002] Oil pipeline leaks are a common occurrence in oil and gas extraction, transportation, and storage. If not handled promptly, they can cause serious environmental pollution and threaten people's lives and property. Fires also pose a serious threat to people's lives and property. Therefore, it is crucial to develop comprehensive emergency response plans to enable rapid response and handling in the event of leaks, fires, or other emergencies, thereby minimizing accident damage and property losses and reducing negative impacts.

[0003] Geographic Information Management Systems (GIS) provide users with the information and insights needed for decision-making through data analysis and simulation, thereby helping users make informed decisions. However, existing GIS systems can only send alarm messages to system users when a crisis occurs, and then send emergency alarms to relevant units (departments) in the local area through manual intervention. Therefore, they cannot react quickly at the first moment when a crisis occurs, delaying the best time for emergency response and causing great harm to people's lives and property.

[0004] Therefore, to address the aforementioned problem of the inability to automatically send emergency alarms to relevant units in the local area, a system and management method that can send emergency alarms to relevant units in the local area can be designed. Summary of the Invention

[0005] In order to overcome the problem of not being able to automatically send emergency alarms to relevant units in the local area.

[0006] The technical solution of this invention is: a geographic information management system for emergency response, including a device terminal and a cloud platform; The device is equipped with a display screen to show the user interface. It collects data through sensors installed on site and transmits the collected data to the cloud platform using 5G / 4G, WIFI, and VPN communication methods. The cloud platform stores and analyzes the data collected from the devices. The cloud platform includes a data processing and analysis module, a data storage and management module, and an emergency alarm module. The data processing and analysis module processes and analyzes the collected information and feeds the results back to the emergency alarm module. The data processing and analysis includes the following steps: (1) Data cleaning process: The raw data is processed to remove noise, errors and redundant information; (2) Data normalization processing, which transforms data of different scales and units into the same standard form; (3) Data discretization and continuous processing Discretization transforms continuous data into discrete data to enable classification and grouping analysis; Continuous transformation converts discrete data into continuous data to enable regression and predictive analysis. (4) Data analysis: statistical analysis of the cleaned data to extract valuable information and conclusions. Data analysis methods include descriptive statistics, inferential statistics, data mining and visualization analysis. (5) Feature engineering, including feature selection, feature construction, feature transformation and feature encoding; (6) Model construction, including linear regression, logistic regression, decision tree, random forest, support vector machine, and neural network. The formula for the linear regression model is expressed as follows: ,in It is the dependent variable. It is the matrix of independent variables. It is a vector of regression coefficients. It is the error term; the formula for the logistic regression model is... , Let represent the probability that the output y=1 given the input x, w be the feature weight vector, b be the bias term, and e be the base of the natural logarithm. This represents a statistic; the formula for the decision tree model is: Let the dataset D = ( ),( ),…( ),in It is the input feature vector, It is the output category; the formula for the random forest model is: Where y represents the target variable and x represents the independent variable. Let M represent the output structure of each tree, and M represent the total number of trees in the forest; the model formula for the Support Vector Machine is... x is the input feature vector. 'b' is the normal vector of the hyperplane, and 'b' is the bias term. Representative variables; the formula for the neural network model is... ,in , Indicates the input to the neuron, , This represents the weight corresponding to each input, and b is the bias term. y represents the activation function, and y represents the output of the neuron. When building the model, the training dataset is used for training, and the validation dataset is used for validation and adjustment of the model parameters. The data storage management module stores the processed and analyzed data into the database, including storage units and management units. The storage units store the data in a structured manner, and the management units maintain and manage the data in the database. The emergency alarm module includes an emergency management department information management unit and an alarm unit. The emergency management department information management unit can manually update the information of the emergency management department in the area. The alarm unit sets the warning threshold for each indicator, monitors the fluctuation of the indicators in real time based on the data processing and analysis results, and determines whether it is a critical event based on the built-in threshold alarm algorithm rules. When a critical event occurs, it sends an emergency alarm to the relevant units in the area. The threshold alarm algorithm rules are as follows: Set warning thresholds for pipeline pressure, pipeline liquid temperature, pipeline flow rate, air temperature, carbon monoxide concentration, carbon dioxide concentration, and methane concentration, and use [methods / methods] respectively. express, These are real-time values. The result is recorded as 1. The result is recorded as 0. Then, the results are compared sequentially and accumulated. When the result is 0, it is considered normal. When the result is between 1 and d, it is considered an unstable state. The result is sent to the system administrator, who will invite technical personnel to make a judgment and troubleshoot. When the result is greater than d, it is considered a critical event, and an emergency alarm is sent to the relevant units in the area.

[0007] Preferably, the display screen is an LCD splicing screen, a DLP rear projection screen, or a small-pitch LED display screen; the data collected by the sensor includes the pressure inside the pipe, the temperature of the liquid inside the pipe, the flow rate inside the pipe, the air temperature, the concentration of carbon monoxide gas, the concentration of carbon dioxide gas, and the concentration of methane gas.

[0008] Preferably, data cleaning includes handling missing values, duplicate values, and outliers. Methods for handling missing values ​​include deleting missing values, filling missing values, interpolating, or extrapolating. Methods for handling duplicate values ​​include deleting duplicate rows, keeping the first or last row, or using a custom method to determine which copies to keep or delete. Methods for handling outliers include removing, pruning extreme values, or replacing extreme values ​​with specified values ​​that are closer to other data points.

[0009] Preferred methods for data normalization include min-max normalization, Z-score normalization, and fractional scaling normalization; methods for data discretization include equal-width discretization, equal-frequency discretization, and cluster discretization; and methods for data continuity include interpolation, regression, and smoothing.

[0010] As preferred methods, descriptive statistics perform basic statistical analysis on data, including counting, summation, mean, and median; inferential statistics perform inferential analysis based on sample data, including hypothesis testing and confidence interval estimation; data mining utilizes data mining algorithms and techniques to uncover hidden association rules, trends, and patterns in data; and visualization analysis uses graphical tools or programming languages ​​to present the analysis results in a visual way, helping users to understand and interpret the data more intuitively.

[0011] Preferably, feature selection uses filtering, wrapping, or embedding methods; feature construction can create ratio, difference, product, or composite features; feature transformation methods include multinomial regression, logarithmic transformation, power transformation, and principal component analysis; feature encoding methods include one-hot encoding, label encoding, and ordinal encoding.

[0012] Preferably, the storage unit uses disks to store data during the data storage process, records data modification operations through log files, employs data compression technology to save storage space, and uses partitioning and sharding technology to improve data storage efficiency.

[0013] Preferably, the management unit includes data backup and recovery, data security, data integrity, and data auditing. Data backup and recovery periodically backs up the data in the database to a secure storage medium; data security protects the confidentiality, integrity, and availability of data through access control, encryption, and auditing; data integrity ensures the consistency and accuracy of data through constraints, triggers, and stored procedures; and data auditing records and analyzes database operation logs, monitors database usage, and identifies and prevents potential security threats.

[0014] A geographic information management method for emergency response, comprising a geographic information management system for emergency response as described in any one of the above-mentioned methods, comprising the following steps: Step 1: The sensor collects data and transmits the collected data to the cloud platform using 5G / 4G, WIFI, or VPN communication methods; Step Two: Data Processing and Analysis (1) Data cleaning process: The raw data is processed to remove noise, errors and redundant information; (2) Data normalization processing, which transforms data of different scales and units into the same standard form; (3) Data discretization and continuous processing Discretization transforms continuous data into discrete data to enable classification and grouping analysis; Continuous transformation converts discrete data into continuous data to enable regression and predictive analysis. (4) Data analysis: statistical analysis of the cleaned data to extract valuable information and conclusions. Data analysis methods include descriptive statistics, inferential statistics, data mining and visualization analysis. (5) Feature engineering, including feature selection, feature construction, feature transformation and feature encoding; (6) Model construction, including linear regression, logistic regression, decision tree, random forest, support vector machine, and neural network. The formula for the linear regression model is expressed as follows: ,in It is the dependent variable. It is the matrix of independent variables. It is a vector of regression coefficients. It is the error term; the formula for the logistic regression model is... , Let represent the probability that the output y=1 given the input x, w be the feature weight vector, b be the bias term, and e be the base of the natural logarithm. This represents a statistic; the formula for the decision tree model is: Let the dataset D = ( ),( ),…( ),in It is the input feature vector, It is the output category; the formula for the random forest model is: Where y represents the target variable and x represents the independent variable. Let M represent the output structure of each tree, and M represent the total number of trees in the forest; the model formula for the Support Vector Machine is... x is the input feature vector. 'b' is the normal vector of the hyperplane, and 'b' is the bias term. Representative variables; the formula for the neural network model is... ,in , Indicates the input to the neuron, , This represents the weight corresponding to each input, and b is the bias term. y represents the activation function, and y represents the output of the neuron. When building the model, the training dataset is used for training, and the validation dataset is used for validation and adjustment of the model parameters. Step 3: Process and analyze the data and store it in the database; Step 4: Based on the data processing and analysis results, monitor the fluctuation of indicators in real time, and combine the built-in threshold alarm algorithm rules to determine whether it is a critical event. When a critical event occurs, issue an emergency alarm to the relevant units in the area. The threshold alarm algorithm rules are as follows: Set warning thresholds for pipeline pressure, pipeline liquid temperature, pipeline flow rate, air temperature, carbon monoxide concentration, carbon dioxide concentration, and methane concentration, and use [methods / methods] respectively. express, These are real-time values. The result is recorded as 1. The result is recorded as 0. Then, the results are compared sequentially and accumulated. When the result is 0, it is considered normal. When the result is between 1 and d, it is considered an unstable state. The result is sent to the system administrator, who will invite technical personnel to make a judgment and troubleshoot. When the result is greater than d, it is considered a critical event, and an emergency alarm is sent to the relevant units in the area.

[0015] The beneficial effects of this invention are as follows: By collecting, processing, and analyzing real-time data, the fluctuation of indicators can be monitored in real time. Combined with the built-in threshold alarm algorithm rules, it can determine whether the current situation is a critical event. In the event of a critical event such as a leak or fire, it can automatically send an emergency alarm to the relevant units in the area, thereby shortening the response time of the relevant units, reducing the loss of life and property, and updating the emergency management information of the emergency management departments within a certain range of the area on a daily basis, thereby ensuring that the emergency alarm information can be delivered accurately. Attached Figure Description

[0016] Figure 1 The diagram shown is a schematic representation of the framework structure of the geographic information management system for emergency response according to the present invention. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0018] Please see Figure 1 The present invention provides an embodiment of an emergency response geographic information management system, including a device terminal and a cloud platform; The device is equipped with a display screen to show the user interface. It collects data through sensors installed on site and transmits the collected data to the cloud platform using 5G / 4G, WIFI, and VPN communication methods. The cloud platform stores and analyzes the data collected from the devices. The cloud platform includes a data processing and analysis module, a data storage and management module, and an emergency alarm module. The data processing and analysis module processes and analyzes the collected information and feeds the results back to the emergency alarm module. The data processing and analysis includes the following steps: (1) Data cleaning process: The raw data is processed to remove noise, errors and redundant information; (2) Data normalization processing, which transforms data of different scales and units into the same standard form; (3) Data discretization and continuous processing Discretization transforms continuous data into discrete data to enable classification and grouping analysis; Continuous transformation converts discrete data into continuous data to enable regression and predictive analysis. (4) Data analysis: statistical analysis of the cleaned data to extract valuable information and conclusions. Data analysis methods include descriptive statistics, inferential statistics, data mining and visualization analysis. (5) Feature engineering, including feature selection, feature construction, feature transformation and feature encoding; (6) Model construction, including linear regression, logistic regression, decision tree, random forest, support vector machine, and neural network. The formula for the linear regression model is expressed as follows: ,in It is the dependent variable. It is the matrix of independent variables. It is a vector of regression coefficients. It is the error term; the formula for the logistic regression model is... , Let represent the probability that the output y=1 given the input x, w be the feature weight vector, b be the bias term, and e be the base of the natural logarithm. This represents a statistic; the formula for the decision tree model is: Let the dataset D = ( ),( ),…( ),in It is the input feature vector, It is the output category; the formula for the random forest model is: Where y represents the target variable and x represents the independent variable. Let M represent the output structure of each tree, and M represent the total number of trees in the forest; the model formula for the Support Vector Machine is... x is the input feature vector. 'b' is the normal vector of the hyperplane, and 'b' is the bias term. Representative variables; the formula for the neural network model is... ,in , Indicates the input to the neuron, , This represents the weight corresponding to each input, and b is the bias term. y represents the activation function, and y represents the output of the neuron. When building the model, the training dataset is used for training, and the validation dataset is used for validation and adjustment of the model parameters. The data storage management module stores the processed and analyzed data into the database, including storage units and management units. The storage units store the data in a structured manner, and the management units maintain and manage the data in the database. The emergency alarm module includes an emergency management department information management unit and an alarm unit. The emergency management department information management unit can manually update the information of the emergency management department in the area. The alarm unit sets the warning threshold for each indicator, monitors the fluctuation of the indicators in real time based on the data processing and analysis results, and determines whether it is a critical event based on the built-in threshold alarm algorithm rules. When a critical event occurs, it sends an emergency alarm to the relevant units in the area. The threshold alarm algorithm rules are as follows: Set warning thresholds for pipeline pressure, pipeline liquid temperature, pipeline flow rate, air temperature, carbon monoxide concentration, carbon dioxide concentration, and methane concentration, and use [methods / methods] respectively. express, These are real-time values. The result is recorded as 1. The result is recorded as 0. Then, the results are compared sequentially and accumulated. When the result is 0, it is considered normal. When the result is between 1 and d, it is considered an unstable state. The result is sent to the system administrator, who will invite technical personnel to make a judgment and troubleshoot. When the result is greater than d, it is considered a critical event, and an emergency alarm is sent to the relevant units in the area.

[0019] Preferably, the display screen is an LCD splicing screen, a DLP rear projection screen, or a small-pitch LED display screen; the data collected by the sensor includes the pressure inside the pipe, the temperature of the liquid inside the pipe, the flow rate inside the pipe, the air temperature, the concentration of carbon monoxide gas, the concentration of carbon dioxide gas, and the concentration of methane gas.

[0020] Preferably, data cleaning includes handling missing values, duplicate values, and outliers. Methods for handling missing values ​​include deleting missing values, filling missing values, interpolating, or extrapolating. Methods for handling duplicate values ​​include deleting duplicate rows, keeping the first or last row, or using a custom method to determine which copies to keep or delete. Methods for handling outliers include removing, pruning extreme values, or replacing extreme values ​​with specified values ​​that are closer to other data points.

[0021] Preferred methods for data normalization include min-max normalization, Z-score normalization, and fractional scaling normalization; methods for data discretization include equal-width discretization, equal-frequency discretization, and cluster discretization; and methods for data continuity include interpolation, regression, and smoothing.

[0022] As preferred methods, descriptive statistics perform basic statistical analysis on data, including counting, summation, mean, and median; inferential statistics perform inferential analysis based on sample data, including hypothesis testing and confidence interval estimation; data mining utilizes data mining algorithms and techniques to uncover hidden association rules, trends, and patterns in data; and visualization analysis uses graphical tools or programming languages ​​to present the analysis results in a visual way, helping users to understand and interpret the data more intuitively.

[0023] Preferably, feature selection uses filtering, wrapping, or embedding methods; feature construction can create ratio, difference, product, or composite features; feature transformation methods include multinomial regression, logarithmic transformation, power transformation, and principal component analysis; feature encoding methods include one-hot encoding, label encoding, and ordinal encoding.

[0024] Preferably, the storage unit uses disks to store data during the data storage process, records data modification operations through log files, employs data compression technology to save storage space, and uses partitioning and sharding technology to improve data storage efficiency.

[0025] Preferably, the management unit includes data backup and recovery, data security, data integrity, and data auditing. Data backup and recovery periodically backs up the data in the database to a secure storage medium; data security protects the confidentiality, integrity, and availability of data through access control, encryption, and auditing; data integrity ensures the consistency and accuracy of data through constraints, triggers, and stored procedures; and data auditing records and analyzes database operation logs, monitors database usage, and identifies and prevents potential security threats.

[0026] Example 1 The steps of geographic information management methods for emergency response are as follows: Step 1: The sensor collects data and transmits the collected data to the cloud platform using 5G / 4G, WIFI, or VPN communication methods; Step Two: Data Processing and Analysis (1) Data cleaning process: The raw data is processed to remove noise, errors and redundant information; (2) Data normalization processing, which transforms data of different scales and units into the same standard form; (3) Data discretization and continuous processing Discretization transforms continuous data into discrete data to enable classification and grouping analysis; Continuous transformation converts discrete data into continuous data to enable regression and predictive analysis. (4) Data analysis: statistical analysis of the cleaned data to extract valuable information and conclusions. Data analysis methods include descriptive statistics, inferential statistics, data mining and visualization analysis. (5) Feature engineering, including feature selection, feature construction, feature transformation and feature encoding; (6) Model construction: A linear regression model is adopted, and the formula is expressed as follows: ,in It is the dependent variable. It is the matrix of independent variables. It is a vector of regression coefficients. It is the error term. When building a model, the training dataset is used for training, and the validation dataset is used for validation and adjustment of model parameters. Step 3: Process and analyze the data and store it in the database; Step 4: Based on the data processing and analysis results, monitor the fluctuation of indicators in real time, and combine the built-in threshold alarm algorithm rules to determine whether it is a critical event. When a critical event occurs, issue an emergency alarm to the relevant units in the area. The threshold alarm algorithm rules are as follows: Set warning thresholds for pipeline pressure, pipeline liquid temperature, pipeline flow rate, air temperature, carbon monoxide concentration, carbon dioxide concentration, and methane concentration, and use [methods / methods] respectively. express, These are real-time values. The result is recorded as 1. The result is recorded as 0. Then, the results are compared sequentially and accumulated. When the result is 0, it is considered normal. When the result is 1 to 3, it is considered an unstable state. The result is sent to the system administrator, who invites technical personnel to make a judgment and troubleshoot. When the result is > 3, it is considered a critical event, and an emergency alarm is sent to the relevant units in the area.

[0027] Example 2 The steps of geographic information management methods for emergency response are as follows: Step 1: The sensor collects data and transmits the collected data to the cloud platform using 5G / 4G, WIFI, or VPN communication methods; Step Two: Data Processing and Analysis (1) Data cleaning process: The raw data is processed to remove noise, errors and redundant information; (2) Data normalization processing, which transforms data of different scales and units into the same standard form; (3) Data discretization and continuous processing Discretization transforms continuous data into discrete data to enable classification and grouping analysis; Continuous transformation converts discrete data into continuous data to enable regression and predictive analysis. (4) Data analysis: statistical analysis of the cleaned data to extract valuable information and conclusions. Data analysis methods include descriptive statistics, inferential statistics, data mining and visualization analysis. (5) Feature engineering, including feature selection, feature construction, feature transformation and feature encoding; (6) Model construction: The model adopts the decision tree model. The formula for the decision tree model is: Let the dataset D = ( ),( ),…( ),in It is the input feature vector, The output category is the model type. When building the model, the training dataset is used for training, and the validation dataset is used for validation and tuning of the model parameters. Step 3: Process and analyze the data and store it in the database; Step 4: Based on the data processing and analysis results, monitor the fluctuation of indicators in real time, and combine the built-in threshold alarm algorithm rules to determine whether it is a critical event. When a critical event occurs, issue an emergency alarm to the relevant units in the area. The threshold alarm algorithm rules are as follows: Set warning thresholds for pipeline pressure, pipeline liquid temperature, pipeline flow rate, air temperature, carbon monoxide concentration, carbon dioxide concentration, and methane concentration, and use [methods / methods] respectively. express, These are real-time values. The result is recorded as 1. The result is recorded as 0. Then, the results are compared sequentially and accumulated. When the result is 0, it is considered normal. When the result is 1 to 4, it is considered an unstable state and the result is sent to the system administrator, who will invite technical personnel to make a judgment and troubleshoot. When the result is > 4, it is considered a critical event and an emergency alarm is sent to the relevant units in the area.

[0028] Example 3 The steps of geographic information management methods for emergency response are as follows: Step 1: The sensor collects data and transmits the collected data to the cloud platform using 5G / 4G, WIFI, or VPN communication methods; Step Two: Data Processing and Analysis (1) Data cleaning process: The raw data is processed to remove noise, errors and redundant information; (2) Data normalization processing, which transforms data of different scales and units into the same standard form; (3) Data discretization and continuous processing Discretization transforms continuous data into discrete data to enable classification and grouping analysis; Continuous transformation converts discrete data into continuous data to enable regression and predictive analysis. (4) Data analysis: statistical analysis of the cleaned data to extract valuable information and conclusions. Data analysis methods include descriptive statistics, inferential statistics, data mining and visualization analysis. (5) Feature engineering, including feature selection, feature construction, feature transformation and feature encoding; (6) Model construction: The model adopts the random forest model, and the formula is as follows: Where y represents the target variable and x represents the independent variable. This represents the output structure of each tree, and M represents the total number of trees in the forest. When building the model, the training dataset is used for training, and the validation dataset is used for validation and adjustment of the model parameters. Step 3: Process and analyze the data and store it in the database; Step 4: Based on the data processing and analysis results, monitor the fluctuation of indicators in real time, and combine the built-in threshold alarm algorithm rules to determine whether it is a critical event. When a critical event occurs, issue an emergency alarm to the relevant units in the area. The threshold alarm algorithm rules are as follows: Set warning thresholds for pipeline pressure, pipeline liquid temperature, pipeline flow rate, air temperature, carbon monoxide concentration, carbon dioxide concentration, and methane concentration, and use [methods / methods] respectively. express, These are real-time values. The result is recorded as 1. The result is recorded as 0. Then, the results are compared sequentially and accumulated. When the result is 0, it is considered normal. When the result is 1 to 3, it is considered an unstable state. The result is sent to the system administrator, who invites technical personnel to make a judgment and troubleshoot. When the result is > 3, it is considered a critical event, and an emergency alarm is sent to the relevant units in the area.

[0029] Example 4 The steps of geographic information management methods for emergency response are as follows: Step 1: The sensor collects data and transmits the collected data to the cloud platform using 5G / 4G, WIFI, or VPN communication methods; Step Two: Data Processing and Analysis (1) Data cleaning process: The raw data is processed to remove noise, errors and redundant information; (2) Data normalization processing, which transforms data of different scales and units into the same standard form; (3) Data discretization and continuous processing Discretization transforms continuous data into discrete data to enable classification and grouping analysis; Continuous transformation converts discrete data into continuous data to enable regression and predictive analysis. (4) Data analysis: statistical analysis of the cleaned data to extract valuable information and conclusions. Data analysis methods include descriptive statistics, inferential statistics, data mining and visualization analysis. (5) Feature engineering, including feature selection, feature construction, feature transformation and feature encoding; (6) Model construction: The model adopts a neural network model, and the formula for the neural network model is as follows: ,in , Indicates the input to the neuron, , This represents the weight corresponding to each input, and b is the bias term. y represents the activation function, and y represents the output of the neuron. When building the model, the training dataset is used for training, and the validation dataset is used for validation and adjustment of the model parameters. Step 3: Process and analyze the data and store it in the database; Step 4: Based on the data processing and analysis results, monitor the fluctuation of indicators in real time, and combine the built-in threshold alarm algorithm rules to determine whether it is a critical event. When a critical event occurs, issue an emergency alarm to the relevant units in the area. The threshold alarm algorithm rules are as follows: Set warning thresholds for pipeline pressure, pipeline liquid temperature, pipeline flow rate, air temperature, carbon monoxide concentration, carbon dioxide concentration, and methane concentration, and use [methods / methods] respectively. express, These are real-time values. The result is recorded as 1. The result is recorded as 0. Then, the results are compared sequentially and accumulated. When the result is 0, it is considered normal. When the result is 1 to 4, it is considered an unstable state and the result is sent to the system administrator, who will invite technical personnel to make a judgment and troubleshoot. When the result is > 4, it is considered a critical event and an emergency alarm is sent to the relevant units in the area.

[0030] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A geographic information management system for emergency response; characterized in that: Including both device-side and cloud platform; The device is equipped with a display screen to show the user interface. It collects data through sensors installed on site and transmits the collected data to the cloud platform using 5G / 4G, WIFI, and VPN communication methods. The cloud platform stores and analyzes the data collected from the devices. The cloud platform includes a data processing and analysis module, a data storage and management module, and an emergency alarm module. The data processing and analysis module processes and analyzes the collected information and feeds the results back to the emergency alarm module. The data processing and analysis includes the following steps: (1) Data cleaning process: The raw data is processed to remove noise, errors and redundant information; (2) Data normalization processing, which transforms data of different scales and units into the same standard form; (3) Data discretization and continuous processing Discretization transforms continuous data into discrete data to enable classification and grouping analysis; Continuous transformation converts discrete data into continuous data to enable regression and predictive analysis. (4) Data analysis: statistical analysis of the cleaned data to extract valuable information and conclusions. Data analysis methods include descriptive statistics, inferential statistics, data mining and visualization analysis. (5) Feature engineering, This includes feature selection, feature construction, feature transformation, and feature encoding; (6) Model construction, including linear regression, logistic regression, decision tree, random forest, support vector machine, and neural network. The formula for the linear regression model is expressed as follows: ,in It is the dependent variable. It is the matrix of independent variables. It is a vector of regression coefficients. It is the error term; the formula for the logistic regression model is... , Let represent the probability that the output y=1 given the input x, w be the feature weight vector, b be the bias term, and e be the base of the natural logarithm. This represents a statistic; the formula for the decision tree model is: Let the dataset D = ( ),( ),…( ),in It is the input feature vector, It is the output category; the formula for the random forest model is: Where y represents the target variable and x represents the independent variable. Let M represent the output structure of each tree, and M represent the total number of trees in the forest; the model formula for the Support Vector Machine is... x is the input feature vector. 'b' is the normal vector of the hyperplane, and 'b' is the bias term. Representative variables; the formula for the neural network model is... ,in , Indicates the input to the neuron. , This represents the weight corresponding to each input, and b is the bias term. y represents the activation function, and y represents the output of the neuron. When building the model, the training dataset is used for training, and the validation dataset is used for validation and adjustment of the model parameters. The data storage management module stores the processed and analyzed data into the database, including storage units and management units. The storage units store the data in a structured manner, and the management units maintain and manage the data in the database. The emergency alarm module includes an emergency management department information management unit and an alarm unit. The emergency management department information management unit can manually update the information of the emergency management department in the area. The alarm unit sets the warning threshold for each indicator, monitors the fluctuation of the indicators in real time based on the data processing and analysis results, and determines whether it is a critical event based on the built-in threshold alarm algorithm rules. When a critical event occurs, it sends an emergency alarm to the relevant units in the area. The threshold alarm algorithm rules are as follows: Set warning thresholds for pipeline pressure, pipeline liquid temperature, pipeline flow rate, air temperature, carbon monoxide concentration, carbon dioxide concentration, and methane concentration, and use [methods / methods] respectively. express, These are real-time values. The result is recorded as 1. The result is recorded as 0. Then, the results are compared sequentially and accumulated. When the result is 0, it is considered normal. When the result is between 1 and d, it is considered an unstable state. The result is sent to the system administrator, who will invite technical personnel to make a judgment and troubleshoot. When the result is greater than d, it is considered a critical event, and an emergency alarm is sent to the relevant units in the area.

2. The geographic information management system for emergency response according to claim 1, characterized in that: The display screen can be an LCD splicing screen, a DLP rear projection screen, or a small-pitch LED display screen; the data collected by the sensors include the pressure inside the pipe, the temperature of the liquid inside the pipe, the flow rate inside the pipe, the air temperature, the concentration of carbon monoxide gas, the concentration of carbon dioxide gas, and the concentration of methane gas.

3. The geographic information management system for emergency response according to claim 1, characterized in that: Data cleaning includes handling missing values, duplicate values, and outliers. Methods for handling missing values ​​include deleting missing values, filling missing values, interpolation, or extrapolation. Methods for handling duplicate values ​​include deleting duplicate rows, keeping the first or last row, or using custom methods to determine which copies to keep or delete. Methods for handling outliers include removing, pruning extreme values, and replacing extreme values ​​with specified values ​​that are closer to other data points.

4. The geographic information management system for emergency response according to claim 1, characterized in that: Data normalization methods include min-max normalization, Z-score normalization, and fractional scaling normalization; data discretization methods include equal-width discretization, equal-frequency discretization, and cluster discretization; data continuity methods include interpolation, regression, and smoothing.

5. The geographic information management system for emergency response according to claim 1, characterized in that: Descriptive statistics performs basic statistical analysis on data, including counting, summation, mean, and median; inferential statistics performs inferential analysis based on sample data, including hypothesis testing and confidence interval estimation; data mining uses data mining algorithms and techniques to uncover hidden association rules, trends, and patterns in data; visualization analysis uses graphical tools or programming languages ​​to present the analysis results in a visual way, helping users to understand and interpret the data more intuitively.

6. The geographic information management system for emergency response according to claim 1, characterized in that: Feature selection can be performed using filtering, wrapping, or embedding methods; feature construction can create ratio, difference, product, or composite features. Feature transformation methods include multinomial regression, logarithmic transformation, power transformation, and principal component analysis; feature encoding methods include one-hot encoding, label encoding, and ordinal encoding.

7. The geographic information management system for emergency response according to claim 1, characterized in that: During the data storage process, the storage unit uses disks to store data and logs data modification operations through log files. It also employs data compression technology to save storage space and partitioning and sharding technology to improve data storage efficiency.

8. The geographic information management system for emergency response according to claim 1, characterized in that: The management unit includes data backup and recovery, data security, data integrity, and data auditing. Data backup and recovery periodically backs up data in the database to secure storage media. Data security protects the confidentiality, integrity, and availability of data through access control, encryption, and auditing. Data integrity ensures data consistency and accuracy through constraints, triggers, and stored procedures. Data auditing records and analyzes database operation logs, monitors database usage, and identifies and prevents potential security threats.

9. A geographic information management method for emergency response, characterized in that... The emergency response geographic information management system, including any one of claims 1-8, comprises the following steps: Step 1: The sensor collects data and transmits the collected data to the cloud platform using 5G / 4G, WIFI, or VPN communication methods; Step Two: Data Processing and Analysis (1) Data cleaning process: The raw data is processed to remove noise, errors and redundant information; (2) Data normalization processing, which transforms data of different scales and units into the same standard form; (3) Data discretization and continuous processing Discretization transforms continuous data into discrete data to enable classification and grouping analysis; Continuous transformation converts discrete data into continuous data to enable regression and predictive analysis. (4) Data analysis: statistical analysis of the cleaned data to extract valuable information and conclusions. Data analysis methods include descriptive statistics, inferential statistics, data mining and visualization analysis. (5) Feature engineering, This includes feature selection, feature construction, feature transformation, and feature encoding; (6) Model construction, including linear regression, logistic regression, decision tree, random forest, support vector machine, and neural network. The formula for the linear regression model is expressed as follows: ,in It is the dependent variable. It is the matrix of independent variables. It is a vector of regression coefficients. It is the error term; the formula for the logistic regression model is... , Let represent the probability that the output y=1 given the input x, w be the feature weight vector, b be the bias term, and e be the base of the natural logarithm. This represents a statistic; the formula for the decision tree model is: Let the dataset D = ( ),( ),…( ),in It is the input feature vector, It is the output category; the formula for the random forest model is: Where y represents the target variable and x represents the independent variable. Let M represent the output structure of each tree, and M represent the total number of trees in the forest; the model formula for the Support Vector Machine is... x is the input feature vector. 'b' is the normal vector of the hyperplane, and 'b' is the bias term. Representative variable; The formula for the neural network model is: ,in , Indicates the input to the neuron. , This represents the weight corresponding to each input, and b is the bias term. y represents the activation function, and y represents the output of the neuron. When building the model, the training dataset is used for training, and the validation dataset is used for validation and adjustment of the model parameters. Step 3: Process and analyze the data and store it in the database; Step 4: Based on the data processing and analysis results, monitor the fluctuation of indicators in real time, and combine the built-in threshold alarm algorithm rules to determine whether it is a critical event. When a critical event occurs, issue an emergency alarm to the relevant units in the area. The threshold alarm algorithm rules are as follows: Set warning thresholds for pipeline pressure, pipeline liquid temperature, pipeline flow rate, air temperature, carbon monoxide concentration, carbon dioxide concentration, and methane concentration, and use [methods / methods] respectively. express, These are real-time values. The result is recorded as 1. The result is recorded as 0. Then, the results are compared sequentially and accumulated. When the result is 0, it is considered normal. When the result is between 1 and d, it is considered an unstable state. The result is sent to the system administrator, who will invite technical personnel to make a judgment and troubleshoot. When the result is greater than d, it is considered a critical event, and an emergency alarm is sent to the relevant units in the area.

Images