Emergency rescue management system and method based on similar neural network

Abstract

The invention discloses an emergency rescue management system and method based on a similar neural network, and relates to the technical field of emergency rescue management.The method comprises the steps that historical emergency rescue data are obtained, and key features are extracted from the cleaned historical emergency rescue data; using historical emergency rescue data to train the neural network-like model; extracting corresponding accident key features from the real-time accident data; substituting the accident key features into a trained neural network model; carrying out rescue resource transportation according to an output result of the neural network-like model; analyzing a transportation influence coefficient according to the rescue resource supply condition of the real-time accident occurrence point; according to the historical emergency rescue data with the secondary transportation rescue resources, a transportation influence coefficient threshold value is analyzed; and carrying out marking and visual alarm display according to the real-time accident occurrence point transportation influence coefficient. The rescue scheme is automatically adjusted, and the rescue efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of emergency rescue management, and in particular to an emergency rescue management system and method based on a quasi-neural network. Background Art

[0002] Urban water supply is the lifeline of the city. The safety of urban water supply is directly related to people's lives and industrial production, and even affects life safety. In order to take effective measures in time when water supply is abnormal and to ensure the safety of urban water supply, it is necessary to formulate an emergency rescue management plan.

[0003] At present, in emergency situations, rescue resources such as personnel and equipment are deployed manually, but insufficient or misallocated resources often occur. The existing technology does not further analyze the real-time rescue resource supply situation at the accident site, and cannot automatically adjust the rescue plan, which can easily cause further property losses.

[0004] Therefore, the present invention discloses an emergency rescue management system and method based on a neural network to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to provide an emergency rescue management system and method based on a neural network to solve the problems raised in the prior art.

[0006] To achieve the above object, the present invention provides the following technical solution: an emergency rescue management method based on a neural network, the method comprising the following steps: S1: Acquire historical emergency rescue data, clean the historical emergency rescue data, and extract key features from the cleaned historical emergency rescue data; S2: Use historical emergency rescue data to train the neural network model; S3: Obtain the real-time accident data of the real-time accident point, extract the corresponding key features of the accident from the real-time accident data; substitute the key features of the accident into the trained neural network model; transport rescue resources according to the output results of the neural network model; analyze the transportation impact coefficient according to the rescue resource supply situation at the real-time accident point; S4: Analyze the threshold of the transport impact coefficient based on the historical emergency rescue data with secondary transport rescue resources; mark and visualize the alarm according to the transport impact coefficient of the real-time accident occurrence point.

[0007] According to the above scheme, in S1, the key features include accident data, resource demand data corresponding to the accident data, and resource supply data corresponding to the accident data; the accident data includes the accident type, the location of the accident occurrence point, the scope of the accident impact, the time of occurrence, the monitoring data of the accident occurrence point, and the sensor data of the accident occurrence point; the resource demand data corresponding to the accident data includes the name of the rescue resource, the estimated delivery time of the rescue resource, and the demand for rescue resources; the resource supply data corresponding to the accident data includes the resource supply point attributes and the rescue resource transportation log; the resource supply point attributes include the location of the resource supply point and the inventory of rescue resources; the rescue resource transportation log refers to the transportation record of all rescue resource transportation tasks, each rescue resource transportation task corresponds to a transportation record, and the transportation record includes the rescue resource supply, the rescue resource delivery time, and the damage information; the damage information refers to the rescue resource information damaged during transportation, including the rescue resource name and the damaged quantity.

[0008] The present invention extracts key features from historical emergency rescue data, wherein the key features include accident data, resource demand data corresponding to the accident data, and resource supply data corresponding to the accident data; and can effectively provide effective data support for subsequent analysis.

[0009] According to the above scheme, S2 includes the following contents: S201: extracting accident key features and rescue key features from historical emergency rescue data; the accident key features include accident type, accident location, accident impact range, accident time, accident monitoring data and accident sensor data; the rescue key features include rescue resource name, rescue resource expected delivery time, rescue resource demand and rescue resource supply at each resource supply point; S202: Manually score the accident based on the key features of each historical emergency rescue data; S203: dividing the accident key features and rescue key features of each historical emergency rescue data after manual scoring into a training set and a validation set according to the proportion, selecting a ZF-Net network structure and a cross entropy function as a loss function, and inputting the scored training set data into a neural network model for training; S204: Analyze the accuracy of the output results of the neural network model according to the validation set data; if the accuracy is greater than the preset accuracy threshold, stop training; the output results include the score value, rescue resource name, rescue resource expected delivery time, rescue resource demand and rescue resource supply of each resource supply point.

[0010] Neural networks can quickly process large amounts of data and respond quickly to emergencies; through training with historical data, they can provide a reliable basis for command decisions.

[0011] According to the above scheme, S3 includes the following: S301: Acquire real-time accident data at the real-time accident occurrence point, extract corresponding key features of the accident from the real-time accident data; substitute the key features of the accident into the trained neural network model; S302: In the output result of the quasi-neural network model, the rescue resource names corresponding to the i-th resource supply point are grouped into a set and recorded as: RES i ={RES （i，1） , RES （i，2） ,…,RES （i，j） ,…,RES （i，J）}; where RES （i，j） represents the jth rescue resource in the rescue resource name set corresponding to the i-th resource supply point; i∈[1,I], I represents the total number of resource supply points providing rescue resources; j∈[1,J], J represents the total number of rescue resources in the rescue resource name set corresponding to the i-th resource supply point; rescue resources are transported according to the output result of the quasi-neural network model; S303: Analyze the transportation impact coefficient TIC according to the rescue resource supply situation at the real-time accident occurrence point: TIC=log a [∑I i=1∑J j=1DR （i，j） +a]×|S(PDT i -DT)|; Among them, DR （i，j） Rescue resources RES （i，j） Corresponding damage rate, PDT i represents the rescue resource delivery time corresponding to the i-th resource supply point, DT represents the estimated rescue resource delivery time at the accident point; S() represents the sample standard deviation function; a is a constant preset by the system; Rescue Resources RES （i，j） Corresponding damage rate DR （i，j） Equal to rescue resources RES （i，j） The corresponding amount of damage divided by the rescue resources RES （i，j） The corresponding supply of rescue resources.

[0012] The present invention analyzes the transportation impact coefficient according to the rescue resource supply situation at the real-time accident occurrence point, can intuitively judge the transportation situation of rescue resources, improve the rescue efficiency, and provide more effective historical emergency rescue data for subsequent accidents.

[0013] According to the above scheme, S4 includes the following contents: S401: Mark the historical emergency rescue data with secondary transport rescue resources, analyze the transportation impact coefficient of the marked historical emergency rescue data, and form a set, which is recorded as CTIC; S402: Analyze the transport impact coefficient threshold CTIC based on the set of transport impact coefficients of the marked historical emergency rescue data T : CTIC T =min{(CTIC min +CTIC max )÷2, CTIC mean}; Among them, CTIC min Represents the minimum value of the transport impact coefficient in the set CTIC; CTIC max Indicates the maximum value of the transport impact coefficient in the set CTIC; CTIC mean It represents the average value of the transport impact coefficient in the set CTIC; min{} represents the minimum value calculation; By analyzing the transport impact coefficient threshold based on the corresponding data of accidents involving secondary transport of rescue resources in historical emergency rescue data, the accident situation can be judged more accurately and the accuracy of the system can be improved.

[0014] S403: If the transport impact coefficient TIC of the real-time accident occurrence point is less than the transport impact coefficient threshold CTIC T , no action will be taken; if the transport impact coefficient TIC of the real-time accident occurrence point is greater than or equal to the transport impact coefficient threshold CTIC T , then mark the real-time accident occurrence point and display a visual alarm to the administrator; S404: Substitute the latest real-time accident data of the real-time accident occurrence point marked into the trained neural network model, and transport rescue resources again according to the output result of the neural network model; after the real-time accident is over, substitute the latest real-time accident data and the corresponding rescue key features into S203, and retrain the neural network model.

[0015] The present invention can compare the transport influence coefficient with the transport influence coefficient threshold, automatically adjust the rescue plan, and improve the rescue efficiency.

[0016] Another aspect of the present application provides an emergency rescue management system based on a neural network, the system is applied to the implementation of the above-mentioned emergency rescue management method based on a neural network, the system includes a rescue data feature analysis module, a neural network model construction module, a real-time accident analysis module and a threshold calculation marking module; The rescue data feature analysis module is used to obtain historical emergency rescue data, perform data cleaning on the historical emergency rescue data, and extract key features from the cleaned historical emergency rescue data; The neural network model building module is used to train the neural network model using historical emergency rescue data; The real-time accident analysis module is used to obtain real-time accident data at the real-time accident point, extract corresponding key features of the accident from the real-time accident data; substitute the key features of the accident into the trained neural network model; transport rescue resources according to the output results of the neural network model; and analyze the transportation impact coefficient according to the rescue resource supply situation at the real-time accident point; The threshold calculation and marking module is used to analyze the transport impact coefficient threshold according to the historical emergency rescue data with secondary transport rescue resources; and mark and visually display the transport impact coefficient according to the real-time accident occurrence point.

[0017] According to the above scheme, the rescue data feature analysis module includes a rescue data preprocessing unit and a feature analysis unit; The rescue data preprocessing unit is used to obtain historical emergency rescue data and perform data cleaning on the historical emergency rescue data; the data cleaning includes removing noise and abnormal values ​​in the data; The feature analysis unit is used to extract key features from historical emergency rescue data, where the key features include accident data, resource demand data corresponding to the accident data, and resource supply data corresponding to the accident data.

[0018] According to the above scheme, the neural network model construction module includes a feature scoring unit and a neural network model training unit; The feature scoring unit is used to extract key accident features and rescue key features from historical emergency rescue data; and manually score the accident based on the key accident features of each historical emergency rescue data; The neural network model training unit is used to divide the accident key features and rescue key features of each historical emergency rescue data after manual scoring into a training set and a verification set according to the proportion, select the ZF-Net network structure, and use the cross entropy function as the loss function. The scored training set data is input into the neural network model for training; the accuracy of the output results of the neural network model is analyzed according to the verification set data; if the accuracy is greater than the preset accuracy threshold, the training is stopped.

[0019] According to the above scheme, the real-time accident analysis module includes a rescue resource calculation unit and a transportation impact coefficient analysis unit; The rescue resource calculation unit is used to obtain real-time accident data at the real-time accident occurrence point, extract corresponding key features of the accident from the real-time accident data; substitute the key features of the accident into the trained neural network model; and transport rescue resources according to the output results of the neural network model; The transport impact coefficient analysis unit is used to analyze the transport impact coefficient according to the rescue resource supply situation at the real-time accident occurrence point.

[0020] According to the above scheme, the threshold calculation and marking module includes a transportation impact coefficient threshold calculation unit and a marking alarm unit; The transport impact coefficient threshold value calculation unit is used to mark the historical emergency rescue data with secondary transport rescue resources, analyze the transport impact coefficient of the marked historical emergency rescue data, and form a set; analyze the transport impact coefficient threshold value according to the set of transport impact coefficients of the marked historical emergency rescue data; The marking alarm unit is used to not process if the transportation impact coefficient of the real-time accident point is less than the transportation impact coefficient threshold; if the transportation impact coefficient of the real-time accident point is greater than or equal to the transportation impact coefficient threshold, then mark the real-time accident point and display a visual alarm to the administrator; substitute the latest real-time accident data of the marked real-time accident point into the trained neural network model, and transport rescue resources again according to the output result of the neural network model; after the real-time accident is over, re-train the neural network model using the latest real-time accident data and corresponding rescue key features.

[0021] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention extracts key features from historical emergency rescue data, wherein the key features include accident data, resource demand data corresponding to the accident data, and resource supply data corresponding to the accident data; it can effectively provide effective data support for subsequent analysis; the neural network can quickly process a large amount of data to achieve a rapid response to emergencies; through the training of historical data, a reliable basis is provided for command decisions; the present invention analyzes the transportation influence coefficient according to the rescue resource supply situation at the real-time accident occurrence point, can intuitively judge the transportation situation of rescue resources, improve the rescue efficiency, and provide more effective historical emergency rescue data for subsequent accidents; according to the corresponding data of accidents with secondary transportation of rescue resources in the historical emergency rescue data, the transportation influence coefficient threshold is analyzed, so as to more accurately judge the accident situation and improve the accuracy of the system; the present invention can compare the transportation influence coefficient with the transportation influence coefficient threshold, automatically adjust the rescue plan, and improve the rescue efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 A schematic diagram of a flow chart of an emergency rescue management method based on a neural network-like system according to the present invention; Figure 2 The present invention is a schematic structural diagram of an emergency rescue management system based on a neural network. DETAILED DESCRIPTION

[0023] 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.

[0024] See also Figure 1 The present invention provides a technical solution: an emergency rescue management method based on a neural network, the method comprising the following steps: S1: Acquire historical emergency rescue data, clean the historical emergency rescue data, and extract key features from the cleaned historical emergency rescue data; In S1, the key features include accident data, resource demand data corresponding to the accident data, and resource supply data corresponding to the accident data; the accident data includes the accident type, the location of the accident occurrence point, the scope of the accident impact, the time of occurrence, the monitoring data of the accident occurrence point, and the sensor data of the accident occurrence point; the resource demand data corresponding to the accident data includes the name of the rescue resource, the estimated delivery time of the rescue resource, and the demand for rescue resources; the resource supply data corresponding to the accident data includes the resource supply point attributes and the rescue resource transportation log; the resource supply point attributes include the location of the resource supply point and the inventory of rescue resources; the rescue resource transportation log refers to the transportation record of all rescue resource transportation tasks, each rescue resource transportation task corresponds to a transportation record, and the transportation record includes the rescue resource supply, the rescue resource delivery time, and the damage information; the damage information refers to the rescue resource information damaged during the transportation process, including the rescue resource name and the damaged quantity.

[0025] S2: Use historical emergency rescue data to train the neural network model; In S2, the following are included: S201: extracting accident key features and rescue key features from historical emergency rescue data; accident key features include accident type, accident location, accident impact range, accident time, accident monitoring data and accident sensor data; rescue key features include rescue resource name, rescue resource expected delivery time, rescue resource demand and rescue resource supply at each resource supply point; S202: Manually score the accident based on the key features of each historical emergency rescue data; S203: dividing the accident key features and rescue key features of each historical emergency rescue data after manual scoring into a training set and a validation set according to the proportion, selecting a ZF-Net network structure and a cross entropy function as a loss function, and inputting the scored training set data into a neural network model for training; S204: Analyze the accuracy of the output results of the neural network model according to the validation set data; if the accuracy is greater than the preset accuracy threshold, stop training; the output results include the score value, rescue resource name, rescue resource expected delivery time, rescue resource demand and rescue resource supply of each resource supply point.

[0026] S3: Obtain the real-time accident data of the real-time accident point, extract the corresponding key features of the accident from the real-time accident data; substitute the key features of the accident into the trained neural network model; transport rescue resources according to the output results of the neural network model; analyze the transportation impact coefficient according to the rescue resource supply situation at the real-time accident point; In S3, include the following: S301: Acquire real-time accident data at the real-time accident occurrence point, extract corresponding key features of the accident from the real-time accident data; substitute the key features of the accident into the trained neural network model; S302: In the output result of the quasi-neural network model, the rescue resource names corresponding to the i-th resource supply point are grouped into a set and recorded as: RES i ={RES （i，1） , RES （i，2） ,…,RES （i，j） ,…,RES （i，J）}; where RES （i，j） represents the jth rescue resource in the rescue resource name set corresponding to the i-th resource supply point; i∈[1,I], I represents the total number of resource supply points providing rescue resources; j∈[1,J], J represents the total number of rescue resources in the rescue resource name set corresponding to the i-th resource supply point; rescue resources are transported according to the output result of the quasi-neural network model; S303: Analyze the transportation impact coefficient TIC according to the rescue resource supply situation at the real-time accident occurrence point: TIC=log a [∑I i=1∑J j=1DR （i，j） +a]×|S(PDT i -DT)|; Among them, DR （i，j） Rescue resources RES （i，j） Corresponding damage rate, PDT i represents the rescue resource delivery time corresponding to the i-th resource supply point, DT represents the estimated rescue resource delivery time at the accident point; S() represents the sample standard deviation function; a is a constant preset by the system; Rescue Resources RES （i，j） Corresponding damage rate DR （i，j） Equal to rescue resources RES （i，j） The corresponding amount of damage divided by the rescue resources RES （i，j） The corresponding supply of rescue resources.

[0027] S4: Analyze the threshold of the transport impact coefficient based on the historical emergency rescue data with secondary transport rescue resources; mark and visualize the alarm according to the transport impact coefficient of the real-time accident occurrence point.

[0028] In S4, the following are included: S401: Mark the historical emergency rescue data with secondary transport rescue resources, analyze the transportation impact coefficient of the marked historical emergency rescue data, and form a set, which is recorded as CTIC; S402: Analyze the transport impact coefficient threshold CTIC based on the set of transport impact coefficients of the marked historical emergency rescue data T : CTIC T =min{(CTIC min +CTIC max )÷2, CTIC mean}; Among them, CTIC min Represents the minimum value of the transport impact coefficient in the set CTIC; CTIC max Indicates the maximum value of the transport impact coefficient in the set CTIC; CTIC mean It represents the average value of the transport impact coefficient in the set CTIC; min{} represents the minimum value calculation; Example 1: In this example, CTIC={5, 8, 8, 7, 9, 6, 10, 12}; Therefore, in this embodiment, CTIC min =5; CTIC max =12, then (CTIC min +CTIC max )÷2=8.5; CTIC mean = (5+8+8+7+9+6+10+12) ÷ 8 = 8.125; CTIC T =min{8.5,8.125}=8.125; S403: If the transport impact coefficient TIC of the real-time accident occurrence point is less than the transport impact coefficient threshold CTIC T , no action will be taken; if the transport impact coefficient TIC of the real-time accident occurrence point is greater than or equal to the transport impact coefficient threshold CTIC T , then mark the real-time accident occurrence point and display a visual alarm to the administrator; S404: Substitute the latest real-time accident data of the marked real-time accident occurrence point into the trained neural network model, and transport rescue resources again according to the output result of the neural network model; after the real-time accident is over, substitute the latest real-time accident data and the corresponding rescue key features into S203, and retrain the neural network model.

[0029] See also Figure 2 ,The present invention provides a technical solution: an emergency rescue management system based on a quasi-neural network, the system includes a rescue data feature analysis module, a quasi-neural network model construction module, a real-time accident analysis module and a threshold calculation marking module; The rescue data feature analysis module is used to obtain historical emergency rescue data, clean the historical emergency rescue data, and extract key features from the cleaned historical emergency rescue data; The neural network model building module is used to train the neural network model using historical emergency rescue data; The real-time accident analysis module is used to obtain the real-time accident data of the real-time accident point, extract the corresponding key features of the accident from the real-time accident data; substitute the key features of the accident into the trained neural network model; transport rescue resources according to the output results of the neural network model; analyze the transportation impact coefficient according to the rescue resource supply situation at the real-time accident point; The threshold calculation and marking module is used to analyze the transport impact coefficient threshold based on historical emergency rescue data with secondary transport rescue resources; and to mark and visually display the transport impact coefficient based on the real-time accident occurrence point.

[0030] The rescue data feature analysis module includes a rescue data preprocessing unit and a feature analysis unit; The rescue data preprocessing unit is used to obtain historical emergency rescue data and perform data cleaning on the historical emergency rescue data; data cleaning includes removing noise and outliers in the data; The feature analysis unit is used to extract key features from historical emergency rescue data, and the key features include accident data, resource demand data corresponding to the accident data, and resource supply data corresponding to the accident data.

[0031] The neural network model building module includes a feature scoring unit and a neural network model training unit; The feature scoring unit is used to extract the key features of accidents and rescue key features from the historical emergency rescue data; and manually score the accidents according to the key features of accidents in each historical emergency rescue data; The neural network model training unit is used to divide the accident key features and rescue key features of each historical emergency rescue data after manual scoring into training set and verification set according to the proportion, select the ZF-Net network structure and the cross entropy function as the loss function, and input the scored training set data into the neural network model for training; analyze the accuracy of the output results of the neural network model according to the verification set data; if the accuracy is greater than the preset accuracy threshold, stop training.

[0032] The real-time accident analysis module includes a rescue resource calculation unit and a transportation impact coefficient analysis unit; The rescue resource calculation unit is used to obtain the real-time accident data of the real-time accident occurrence point, extract the corresponding key features of the accident from the real-time accident data; substitute the key features of the accident into the trained neural network model; and transport rescue resources according to the output results of the neural network model; The transport impact coefficient analysis unit is used to analyze the transport impact coefficient according to the rescue resource supply situation at the real-time accident occurrence point.

[0033] The threshold calculation and marking module includes a transport impact coefficient threshold calculation unit and a marking alarm unit; The transport impact coefficient threshold calculation unit is used to mark the historical emergency rescue data with secondary transport rescue resources, analyze the transport impact coefficient of the marked historical emergency rescue data, and form a set; analyze the transport impact coefficient threshold according to the set of transport impact coefficients of the marked historical emergency rescue data; The marking alarm unit is used to not process if the transportation impact coefficient of the real-time accident point is less than the transportation impact coefficient threshold; if the transportation impact coefficient of the real-time accident point is greater than or equal to the transportation impact coefficient threshold, the real-time accident point is marked and a visual alarm is displayed to the administrator; the latest real-time accident data of the marked real-time accident point is substituted into the trained neural network model, and rescue resources are transported again according to the output results of the neural network model; after the real-time accident is over, the neural network model is retrained using the latest real-time accident data and the corresponding rescue key features.

[0034] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0035] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An emergency rescue management method based on a neural network, characterized in that: The method comprises the following steps: S1: Acquire historical emergency rescue data, clean the historical emergency rescue data, and extract key features from the cleaned historical emergency rescue data; S2: Use historical emergency rescue data to train the neural network model; S3: Acquire real-time accident data at the real-time accident occurrence point, and extract corresponding key features of the accident from the real-time accident data; Substitute the key features of the accident into the trained neural network model; transport rescue resources according to the output of the neural network model; analyze the transportation impact coefficient according to the rescue resource supply situation at the real-time accident point; S4: Analyze the threshold of the transport impact coefficient based on the historical emergency rescue data with secondary transport rescue resources; mark and visualize the alarm according to the transport impact coefficient of the real-time accident occurrence point.

2. The method for emergency rescue management based on a neural network according to claim 1, characterized in that: In S1, the key features include accident data, resource demand data corresponding to the accident data, and resource supply data corresponding to the accident data; the accident data includes accident type, accident location, accident impact range, accident time, accident monitoring data, and accident sensor data; The resource demand data corresponding to the accident data includes the name of the rescue resources, the estimated delivery time of the rescue resources and the demand for rescue resources; the resource supply data corresponding to the accident data includes the resource supply point attributes and the rescue resource transportation log; the resource supply point attributes include the resource supply point location and the rescue resource inventory; the rescue resource transportation log refers to the transportation record of all rescue resource transportation tasks, each rescue resource transportation task corresponds to a transportation record, and the transportation record includes the rescue resource supply, the rescue resource delivery time and the damage information; the damage information refers to the information of rescue resources damaged during transportation, including the name of the rescue resource and the damaged quantity.

3. The emergency rescue management method based on a neural network according to claim 2, characterized in that: In S2, the following are included: S201: extracting accident key features and rescue key features from historical emergency rescue data; The key features of the accident include the type of accident, the location of the accident, the scope of the accident, the time of occurrence, the monitoring data of the accident point and the sensor data of the accident point; the key features of the rescue include the name of the rescue resource, the estimated time of rescue resource delivery, the demand for rescue resources and the supply of rescue resources at each resource supply point; S202: Manually score the accident based on the key features of each historical emergency rescue data; S203: dividing the accident key features and rescue key features of each historical emergency rescue data after manual scoring into a training set and a validation set according to the proportion, selecting a ZF-Net network structure and a cross entropy function as a loss function, and inputting the scored training set data into a neural network model for training; S204: Analyze the accuracy of the output results of the neural network model according to the validation set data; if the accuracy is greater than the preset accuracy threshold, stop training; the output results include the score value, rescue resource name, rescue resource expected delivery time, rescue resource demand and rescue resource supply of each resource supply point.

4. The method for emergency rescue management based on a neural network according to claim 3, characterized in that: In S3, include the following: S301: Acquire real-time accident data at the real-time accident occurrence point, and extract corresponding key features of the accident from the real-time accident data; Substitute the key features of the accident into the trained neural network model; S302: In the output result of the quasi-neural network model, the rescue resource names corresponding to the i-th resource supply point are grouped into a set and recorded as: RES i ={RES （i，1） , RES （i，2） ,…,RES （i，j） ,…,RES （i，J） }; where RES （i，j） Indicates the jth rescue resource in the rescue resource name set corresponding to the i-th resource supply point; i∈[1,I], where I represents the total number of resource supply points providing rescue resources; j∈[1, J], J represents the total number of rescue resources in the rescue resource name set corresponding to the i-th resource supply point; rescue resources are transported according to the output results of the neural network model; S303: Analyze the transportation impact coefficient TIC according to the rescue resource supply situation at the real-time accident occurrence point: TIC=log a [∑I i=1∑J j=1DR （i，j） +a]×|S(PDT i -DT)|; Among them, DR （i，j） Rescue resources RES （i，j） Corresponding damage rate, PDT i represents the rescue resource delivery time corresponding to the i-th resource supply point, DT represents the estimated rescue resource delivery time at the accident point; S() represents the sample standard deviation function; a is a constant preset by the system; Rescue Resources RES （i，j） Corresponding damage rate DR （i，j） Equal to rescue resources RES （i，j） The corresponding amount of damage divided by the rescue resources RES （i，j） The corresponding supply of rescue resources.

5. The method for emergency rescue management based on a neural network according to claim 4, characterized in that: In S4, the following are included: S401: Mark the historical emergency rescue data with secondary transport rescue resources, analyze the transportation impact coefficient of the marked historical emergency rescue data, and form a set, which is recorded as CTIC; S402: Analyze the transport impact coefficient threshold CTIC based on the set of transport impact coefficients of the marked historical emergency rescue data T : CTIC T =min{(CTIC min +CTIC max )÷2,CTIC mean }; Among them, CTIC min Represents the minimum value of the transport impact coefficient in the set CTIC; CTIC max Indicates the maximum value of the transport impact coefficient in the set CTIC; CTIC mean It represents the average value of the transport impact coefficient in the set CTIC; min{} represents the minimum value calculation; S403: If the transport impact coefficient TIC of the real-time accident occurrence point is less than the transport impact coefficient threshold CTIC T , no action will be taken; if the transport impact coefficient TIC of the real-time accident occurrence point is greater than or equal to the transport impact coefficient threshold CTIC T , then mark the real-time accident occurrence point and display a visual alarm to the administrator; S404: Substitute the latest real-time accident data of the real-time accident occurrence point marked into the trained neural network model, and transport rescue resources again according to the output result of the neural network model; after the real-time accident is over, substitute the latest real-time accident data and the corresponding rescue key features into S203, and retrain the neural network model.

6. A neural network-based emergency rescue management system, the system is applied to the neural network-based emergency rescue management method according to any one of claims 1 to 5, characterized in that: The system includes a rescue data feature analysis module, a neural network model building module, a real-time accident analysis module and a threshold calculation marking module; The rescue data feature analysis module is used to obtain historical emergency rescue data, perform data cleaning on the historical emergency rescue data, and extract key features from the cleaned historical emergency rescue data; The neural network model building module is used to train the neural network model using historical emergency rescue data; The real-time accident analysis module is used to obtain real-time accident data at the real-time accident occurrence point and extract corresponding key accident features from the real-time accident data; Substitute the key features of the accident into the trained neural network model; transport rescue resources according to the output of the neural network model; analyze the transportation impact coefficient according to the rescue resource supply situation at the real-time accident point; The threshold calculation marking module is used to analyze the transport impact coefficient threshold based on the historical emergency rescue data with secondary transport rescue resources; Mark and display visual alarms based on the transport impact coefficient of the real-time accident occurrence point.

7. The neural network-based emergency rescue management system according to claim 6, characterized in that: The rescue data feature analysis module includes a rescue data preprocessing unit and a feature analysis unit; The rescue data preprocessing unit is used to obtain historical emergency rescue data and perform data cleaning on the historical emergency rescue data; the data cleaning includes removing noise and abnormal values ​​in the data; The feature analysis unit is used to extract key features from historical emergency rescue data, where the key features include accident data, resource demand data corresponding to the accident data, and resource supply data corresponding to the accident data.

8. The neural network-based emergency rescue management system according to claim 6, characterized in that: The neural network model building module includes a feature scoring unit and a neural network model training unit; The feature scoring unit is used to extract key accident features and rescue key features from historical emergency rescue data; and manually score the accident based on the key accident features of each historical emergency rescue data; The neural network model training unit is used to divide the accident key features and rescue key features of each historical emergency rescue data after manual scoring into a training set and a verification set according to the proportion, select the ZF-Net network structure, and use the cross entropy function as the loss function. The scored training set data is input into the neural network model for training; the accuracy of the output results of the neural network model is analyzed according to the verification set data; if the accuracy is greater than the preset accuracy threshold, the training is stopped.

9. The neural network-based emergency rescue management system according to claim 6, characterized in that: The real-time accident analysis module includes a rescue resource calculation unit and a transportation impact coefficient analysis unit; The rescue resource calculation unit is used to obtain real-time accident data at the real-time accident occurrence point and extract corresponding key accident features from the real-time accident data; Substitute the key features of the accident into the trained neural network model; transport rescue resources according to the output results of the neural network model; The transport impact coefficient analysis unit is used to analyze the transport impact coefficient according to the rescue resource supply situation at the real-time accident occurrence point.

10. The neural network-based emergency rescue management system according to claim 6, characterized in that: The threshold calculation and marking module includes a transport impact coefficient threshold calculation unit and a marking alarm unit; The transport impact coefficient threshold value calculation unit is used to mark the historical emergency rescue data with secondary transport rescue resources, analyze the transport impact coefficient of the marked historical emergency rescue data, and form a set; analyze the transport impact coefficient threshold value according to the set of transport impact coefficients of the marked historical emergency rescue data; The marking alarm unit is used to not process if the transportation impact coefficient of the real-time accident occurrence point is less than the transportation impact coefficient threshold; If the transport impact coefficient of the real-time accident occurrence point is greater than or equal to the transport impact coefficient threshold, the real-time accident occurrence point is marked, and a visual alarm is displayed to the administrator; the latest real-time accident data of the marked real-time accident occurrence point is substituted into the trained neural network model, and rescue resources are transported again according to the output result of the neural network model; After the real-time accident is over, the neural network model is retrained using the latest real-time accident data and the corresponding rescue key features.

Images