Emergency response plan and task generation system based on generative artificial intelligence

Abstract

The invention provides an emergency response plan and task generation system based on generative artificial intelligence, and the system comprises a knowledge base search system which receives emergency event information inputted by a user through an emergency event information input interface, the emergency event information is processed, matched emergency plan text fragments are screened out and input into an emergency response plan generation system and an emergency response task generation system, and then the emergency response plan generation system generates a targeted emergency response plan according to the emergency event information and the emergency plan text fragments; and the emergency response task generation system generates an emergency response task list and description according to the emergency response plan, the emergency event information and the emergency plan text fragment. Through cooperative work of the three components, intelligent retrieval of emergency plan files, and automatic generation of specific emergency response plans and tasks are realized.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence and emergency management technology, and in particular to an emergency response plan and task generation system based on generative artificial intelligence, an electronic device and a readable storage medium. Background Art

[0002] Public emergencies refer to emergencies that occur suddenly and may cause heavy casualties, property losses, ecological environmental damage, and serious social harm, endangering public safety. According to the type and severity, these events can be divided into natural disasters, accidental disasters, public health events, and social security events, and are divided into Level I (especially serious), Level II (serious), Level III (relatively serious), and Level IV (general) according to the severity level.

[0003] Existing emergency command systems usually use manual analysis and judgment to handle emergency situations. When an emergency occurs, the system needs to use manual comprehensive analysis to determine the severity of the situation, and formulate emergency response plans and tasks to be executed based on the judgment results and relevant emergency plans. This method has some significant technical defects in practical applications.

[0004] First, the complexity of emergencies and the diversity of emergency plans make the analysis process extremely complicated. In an emergency, the system needs to complete analysis and decision-making in a very short time to prevent the situation from getting worse. However, the speed of manual analysis often cannot meet this time requirement, resulting in delayed response and possible unnecessary losses.

[0005] Secondly, it is difficult to combine the actual situation with the preset emergency plan. The specific situation of each emergency may be different from the plan, and it is necessary to make quick adjustments based on the actual situation. Manual analysis is prone to information omission or misjudgment in this process, affecting the accuracy and effectiveness of emergency response.

[0006] Finally, emergency rescue usually involves the collaboration of multiple departments and units, and the division of tasks is complex. Emergency response plans formulated manually in a short period of time may not fully consider the resource allocation and coordination issues of all aspects, affecting the overall rescue efficiency.

[0007] In view of the above problems, the existing technical solutions are difficult to ensure the timeliness, accuracy and comprehensiveness of emergency response when facing complex and changeable public emergencies. These technical defects directly affect the effectiveness of emergency management and may lead to serious consequences such as casualties and property losses. Therefore, there is an urgent need for a new method that can quickly, accurately and comprehensively generate emergency response plans and tasks to improve the ability to respond to public emergencies. Summary of the invention

[0008] In view of the above problems, the present invention is proposed to provide an emergency response plan and task generation system based on generative artificial intelligence, an electronic device and a readable storage medium that overcome the above problems or at least partially solve the above problems.

[0009] The present invention provides an emergency response plan and task generation system based on generative artificial intelligence, comprising:

[0010] A knowledge base search system is used to receive emergency event information input by a user through an emergency event information input interface, process the emergency event information and filter out matching emergency plan text fragments, and input the matching emergency plan text fragments into an emergency response plan generation system and an emergency response task generation system;

[0011] An emergency response plan generation system is used to respond to a user's emergency response plan generation request, obtain the emergency event information and receive the emergency plan text fragment input by the knowledge base search system, and generate an emergency response plan according to the emergency event information and the emergency plan text fragment;

[0012] The emergency response task generation system is used to respond to the user's emergency response task generation request, obtain the sudden emergency event information, the emergency response plan and receive the emergency plan text fragment input by the knowledge base search system, and generate an emergency response task based on the emergency response plan, the sudden emergency event information and the emergency plan text fragment.

[0013] Optionally, the knowledge base search system includes a text segmentation and embedding system, a vector data storage system, and a vector query system.

[0014] The text segmentation and embedding system is used to segment the emergency event information into a plurality of emergency event information blocks by using a fixed-length segmentation method based on line breaks or a semantic segmentation method, and convert the emergency event information blocks into corresponding emergency event information embedding vectors by using a text embedding model;

[0015] The vector data storage system is used to pre-store the emergency plan file name, the emergency plan text segment corresponding to the emergency plan file name, and the emergency plan text segment embedding vector corresponding to the emergency plan text segment;

[0016] The vector query system is used to calculate the semantic similarity between each sudden emergency event information embedding vector and all the emergency plan text fragment embedding vectors in the vector data storage system, and screen out a preset number of emergency plan text fragment embedding vectors whose semantic similarity ranking is at the top with respect to the sudden emergency event information embedding vector, extract the emergency plan text fragments corresponding to the preset number of emergency plan text fragment embedding vectors, use the text rearrangement semantic analysis model to sort the extracted preset number of emergency plan text fragments by relevance according to the sudden emergency event information, determine the emergency plan text fragment with the highest relevance as the emergency plan text fragment matching the sudden emergency event information, summarize the emergency plan text fragments with the highest relevance of all sudden emergency event information embedding vectors to obtain all the emergency plan text fragments matching the sudden emergency event information, and input all the emergency plan text fragments matching the sudden emergency event information into the emergency response plan generation system and the emergency response task generation system.

[0017] Optionally, the knowledge base search system includes a text segmentation and embedding system and a vector data storage system.

[0018] The text segmentation and embedding system is also used to process emergency plan files, using a fixed-length segmentation method based on line breaks or a semantic-based segmentation method to segment the file content of the emergency plan file into multiple emergency plan text segments, and using a text embedding model to convert the emergency plan text segments into corresponding emergency plan text segment embedding vectors, and input the emergency plan file name, the emergency plan text segment corresponding to the emergency plan file name, and the emergency plan text segment embedding vector corresponding to the emergency plan text segment into a vector data storage system;

[0019] The vector data storage system is used to store and manage the received emergency plan file name, the emergency plan text segment corresponding to the emergency plan file name, and the emergency plan text segment embedding vector corresponding to the emergency plan text segment.

[0020] Optionally, the emergency response plan generation system includes a generative artificial intelligence role defined by prompt words; the emergency response task generation system includes a generative artificial intelligence role defined by prompt words and a fixed predefined task format output; the generative artificial intelligence role defined by the prompt words is set to play the role of an emergency response commander through preset large language model prompt words.

[0021] Optionally,

[0022] The emergency response plan generation system is used to parse the emergency event information to obtain the time, location, event type, impact scope, and number of people affected, and adapt, integrate and output the time, location, event type, impact scope, number of people affected, and the measures and resource requirements in the emergency plan text segment to obtain the emergency response plan.

[0023] Optionally,

[0024] The emergency response task generation system is used to parse each plan item of the emergency response plan, and combine the sudden emergency event information and the emergency plan text fragment to determine the task item information corresponding to each plan item, and output it in a predefined task format to obtain the emergency response task; the task item information includes the execution unit, task objectives, required manpower, required materials, and estimated completion time.

[0025] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the electronic device executes the computer program, it loads the emergency response plan and task generation system based on generative artificial intelligence as described in any one of the embodiments of the present invention.

[0026] The present invention also provides a readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the emergency response plan and task generation system based on generative artificial intelligence as described in any one of the embodiments of the present invention is loaded.

[0027] The present invention includes the following advantages:

[0028] The emergency response plan and task generation system based on generative artificial intelligence of the present invention receives the emergency event information input by the user through the emergency event information input interface through the knowledge base search system, processes it and selects the matching emergency plan text fragments to input into the emergency response plan generation system and the emergency response task generation system, and then the emergency response plan generation system generates an emergency response plan according to the emergency event information and the emergency plan text fragments, and the emergency response task generation system generates an emergency response task according to the emergency response plan, the emergency event information and the emergency plan text fragments. Through the collaborative work of these three components, the intelligent retrieval of emergency plan files to the automatic generation of specific emergency response plans and tasks is realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a structural block diagram of an emergency response plan and task generation system based on generative artificial intelligence provided by an embodiment of the present invention;

[0030] Figure 2It is a flow chart of emergency plan preprocessing and screening matching emergency plan text fragments provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] Reference Figure 1 , shows a structural block diagram of an emergency response plan and task generation system based on generative artificial intelligence provided in an embodiment of the present invention, which may specifically include the following modules:

[0033] A knowledge base search system is used to receive emergency event information input by a user through an emergency event information input interface, process the emergency event information and filter out matching emergency plan text fragments, and input the matching emergency plan text fragments into an emergency response plan generation system and an emergency response task generation system;

[0034] An emergency response plan generation system is used to respond to a user's emergency response plan generation request, obtain the emergency event information and receive the emergency plan text fragment input by the knowledge base search system, and generate an emergency response plan according to the emergency event information and the emergency plan text fragment;

[0035] The emergency response task generation system is used to respond to the user's emergency response task generation request, obtain the sudden emergency event information, the emergency response plan and receive the emergency plan text fragment input by the knowledge base search system, and generate an emergency response task based on the emergency response plan, the sudden emergency event information and the emergency plan text fragment.

[0036] In one embodiment of the present invention, the knowledge base search system includes a text segmentation and embedding system, a vector data storage system, and a vector query system.

[0037] The text segmentation and embedding system is used to segment the emergency event information into a plurality of emergency event information blocks by using a fixed-length segmentation method based on line breaks or a semantic segmentation method, and convert the emergency event information blocks into corresponding emergency event information embedding vectors by using a text embedding model;

[0038] The vector data storage system is used to pre-store the emergency plan file name, the emergency plan text segment corresponding to the emergency plan file name, and the emergency plan text segment embedding vector corresponding to the emergency plan text segment;

[0039] The vector query system is used to calculate the semantic similarity between each sudden emergency event information embedding vector and all the emergency plan text fragment embedding vectors in the vector data storage system, and screen out a preset number of emergency plan text fragment embedding vectors whose semantic similarity ranking is at the top with respect to the sudden emergency event information embedding vector, extract the emergency plan text fragments corresponding to the preset number of emergency plan text fragment embedding vectors, use the text rearrangement semantic analysis model to sort the extracted preset number of emergency plan text fragments by relevance according to the sudden emergency event information, determine the emergency plan text fragment with the highest relevance as the emergency plan text fragment matching the sudden emergency event information, summarize the emergency plan text fragments with the highest relevance of all sudden emergency event information embedding vectors to obtain all the emergency plan text fragments matching the sudden emergency event information, and input all the emergency plan text fragments matching the sudden emergency event information into the emergency response plan generation system and the emergency response task generation system.

[0040] In one embodiment of the present invention, the knowledge base search system includes a text segmentation and embedding system, a vector data storage system,

[0041] The text segmentation and embedding system is also used to obtain an emergency plan file, segment the file content of the emergency plan file into multiple emergency plan text segments using a fixed-length segmentation method based on line breaks or a semantic-based segmentation method, and convert the emergency plan text segments into corresponding emergency plan text segment embedding vectors using a text embedding model, and input the emergency plan file name, the emergency plan text segment corresponding to the emergency plan file name, and the emergency plan text segment embedding vector corresponding to the emergency plan text segment into a vector data storage system;

[0042] The vector data storage system is used to store and manage the received emergency plan file name, the emergency plan text segment corresponding to the emergency plan file name, and the emergency plan text segment embedding vector corresponding to the emergency plan text segment.

[0043] In one embodiment of the present invention, the emergency response plan generation system includes a generative artificial intelligence role defined by prompt words; the emergency response task generation system includes a generative artificial intelligence role defined by prompt words and a fixed predefined task format output; the generative artificial intelligence role defined by the prompt words is set to play the role of an emergency response commander through a preset large language model prompt word.

[0044] In one embodiment of the present invention,

[0045] The emergency response plan generation system is used to parse the sudden emergency event information to obtain event elements such as time, location, event type, impact range, and number of people affected, and to adapt, integrate, and output the elements such as time, location, event type, impact range, and number of people affected with the measures and resource requirements in the emergency plan text segment to obtain the emergency response plan.

[0046] In one embodiment of the present invention,

[0047] The emergency response task generation system is used to parse each plan item of the emergency response plan, and combine the sudden emergency event information and the emergency plan text fragment to determine the task item information corresponding to each plan item, and output it in a predefined task format to obtain the emergency response task; the task item information includes the execution unit, task objectives, required manpower, required materials, estimated completion time, etc.

[0048] The present invention provides a system for generating emergency response plans and tasks based on generative artificial intelligence. The system is mainly composed of three core components: a knowledge base search system, an emergency response plan generation system, and an emergency response task generation system. These three components work together to achieve the intelligent retrieval of emergency plan documents and the automatic generation of specific emergency response plans and tasks.

[0049] Among them, the knowledge base search system includes text segmentation and embedding system, vector data storage system and vector query system.

[0050] The emergency response plan generation system includes predefined generative artificial intelligence "roles".

[0051] The emergency response task generation system includes predefined generative artificial intelligence "roles" and a fixed output format.

[0052] Furthermore, the knowledge base search system is the basis of the present invention, which includes three key subsystems: a text segmentation and embedding system, a vector data storage system, and a vector query system. The file segmentation and embedding system adopts a fixed-length segmentation method based on line breaks or a semantic-based segmentation method, which not only retains the file content, but also ensures that the segmented files can maintain certain format information. This design adapts to the length limit of the text embedding artificial intelligence model. Subsequently, the text embedding model converts these segments into corresponding embedding vectors. The vector data storage system is responsible for storing these embedded vectors together with the segmented file information and file names into the database. The vector query system provides a text query function based on the semantic similarity of natural language text. Combined with text materials such as laws and regulations, emergency plans, etc. in the field of emergency, this system can be used as a comprehensive emergency expert knowledge base to provide services for other modules in the system.

[0053] Furthermore, the vector query system includes a database query system and a text reordering system. The database query system calculates the semantic similarity between the embedded vector of the input emergency event information and the embedded vector of the pre-stored emergency plan, obtains the semantic similarity, and queries the top N knowledge fragments (N is an adjustable variable) that are most relevant to the input embedded vector from the database. The text reordering system based on the text reordering model reorders these N knowledge fragments, selects the knowledge text that is most relevant to the input event text, and uses it as the input of the emergency response plan generation system and the emergency response task generation system.

[0054] Furthermore, the emergency response plan generation system includes a predefined "role" of generative artificial intelligence, which is composed of a large language model and prompt words. The large language model has the ability to understand, reason and generate natural language, and can perform the specified text generation task according to the "role" played by the corresponding prompt words. In this system, the prompt words are set to let the large language model play the role of emergency response command, and generate emergency plans and tasks based on the provided emergency knowledge text and emergencies. This process gives full play to the advantages of generative artificial intelligence in information processing and decision-making speed, and greatly shortens the time required for traditional manual planning.

[0055] Furthermore, the output format of generative artificial intelligence is predefined to ensure a fixed output format of generative artificial intelligence, thereby ensuring that the plan for the output of generative artificial intelligence can use a fixed parsing system to generate a file in the format required by the scenario.

[0056] Furthermore, the emergency response plan generation system uses generative artificial intelligence with preset "roles" to comprehensively consider the information of emergency events and related emergency plan documents output by the file query system. The generative artificial intelligence outputs an overall emergency response plan. Since generative artificial intelligence has a natural speed advantage over manual labor while being able to comprehensively process text information, it greatly reduces the time consumption caused by manual emergency response plan formulation.

[0057] The present invention combines advanced text processing technology, intelligent search algorithms and generative artificial intelligence to create a comprehensive and efficient emergency response system. Compared with traditional methods, this system not only significantly improves the speed and accuracy of emergency plan matching, but also generates detailed and executable emergency response plans and tasks in a very short time, providing strong technical support for quickly and effectively responding to various public emergencies.

[0058] Specifically, refer to Figure 2 The specific implementation process of the present invention mainly includes the following steps:

[0059] 1. System initialization and data preparation

[0060] Before the system is put into use, it is necessary to first establish an emergency plan knowledge base. This process involves processing emergency laws and regulations, emergency plans, emergency experience summaries, etc. through file segmentation and embedding system processing, generating corresponding text segment embedding vectors, and storing them in the vector database together with the original text. This step is usually completed by the system administrator in the background without user participation.

[0061] 2. User interface operation

[0062] When an emergency occurs, the user enters the event information through the page interface provided by the system. This includes but is not limited to the following main event information:

[0063] Time: The specific date and time when the event occurred.

[0064] Location: The detailed geographical location where the incident occurred, including province, city, district, county and specific location.

[0065] Event description: A detailed description of the event, including key information such as event type, scope of impact, and number of people affected.

[0066] 3. Emergency response plan generation

[0067] After the user fills in the information, click the "Generate Plan" button or automatically, the system will perform the following steps:

[0068] a) The file query system converts the input event information into an embedding vector.

[0069] b) The system performs similarity calculations in the database and preliminarily selects relevant emergency plan texts.

[0070] c) The document re-ranking system further ranks the screening results to determine the most relevant knowledge text fragments.

[0071] d) Generative AI comprehensively analyzes event information and related plans to generate an overall emergency response plan.

[0072] 4. Emergency response task generation

[0073] After the user clicks the "Generate Task" button, the system will perform similar steps, but the focus is on generating specific executable task items:

[0074] a) The system repeats the document search and re-ranking process to ensure that the latest and most relevant knowledge text information is used.

[0075] b) Generative AI generates a detailed list of emergency response tasks based on the preset emergency task output format.

[0076] c) Each task item includes specific information such as the executing unit, required materials, and estimated completion time.

[0077] As an example of the present invention:

[0078] In order to more specifically illustrate the use method and effect of the system of the present invention, a hypothetical emergency event is taken as an example:

[0079] Assume that on x month x day in 2024, a large reservoir in a certain area of ​​a city breaks its dam. The user enters the following information in the system interface:

[0080] Time: 2024-03-01 14:30

[0081] Location: A certain reservoir in a certain district of a certain city

[0082] Description of the incident: The dam on the west side of a certain reservoir broke, the length of the breach was about 50 meters, and the water level of the reservoir was about 3 meters higher than the lowest point of the breach. There are 3 villages within 2 kilometers nearby, with a total population of about 5,000 people, and they need to be evacuated urgently.

[0083] After the user enters the information and clicks the "Generate Plan" button, the system generates a detailed emergency response plan within a few seconds, including the following main contents:

[0084] Immediately activate the flood prevention emergency plan and set up an on-site command center.

[0085] Organize fire fighters, other rescue personnel and other forces to urgently seal the breach.

[0086] Mobilize bulldozers, loaders and other heavy machinery to support blocking the mouth.

[0087] Organize emergency evacuation of residents in nearby villages, giving priority to the elderly, children and the sick.

[0088] Set up temporary shelters and prepare food, drinking water and basic daily necessities.

[0089] Medical teams were put on standby to deal with possible casualties.

[0090] Notify downstream areas to make precautionary preparations.

[0091] Then, the user clicks the "Generate Tasks" button, and the system generates a series of specific tasks within a few seconds, such as:

[0092] A city’s Emergency Management Bureau: Immediately launch the city-level flood prevention emergency plan and coordinate the actions of various departments.

[0093] The Public Security Bureau of a certain district: Organize 300 rescue personnel to evacuate 5,000 residents in the surrounding area, and the task is expected to take 2 hours.

[0094] A city's fire rescue detachment dispatched 10 fire trucks and 50 firefighters to the scene for blockade and rescue operations.

[0095] The Transportation Bureau of a certain city: mobilized 20 buses to assist in the evacuation work.

[0096] The Water Resources Department of a certain province dispatched a team of 5 water resources experts to the scene to assess the dangerous situation and provide technical support.

[0097] The Civil Affairs Bureau of a certain city: Set up three temporary resettlement points in safe areas and prepare 5,000 emergency food and drinking water.

[0098] Through this example, we can see that this system can generate comprehensive and specific emergency response plans and tasks in a very short time, providing decision makers and executors with valuable time and clear action guidelines, greatly improving the efficiency and effectiveness of responding to emergencies.

[0099] The emergency response plan and task generation system based on generative artificial intelligence of the present invention has the following advantages:

[0100] First of all, this system uses an advanced text embedding artificial intelligence network neural model to pre-process emergency plan documents. By dividing emergency laws and regulations and emergency plan documents into blocks and converting them into embedded vectors, they are stored in a vector database to form an emergency expert knowledge base. The system can retrieve the emergency vector database by semantic similarity to obtain highly relevant emergency plan knowledge, thereby improving the efficiency of knowledge retrieval. In practical applications, when an emergency event occurs, the system can quickly convert the event information into an embedded vector and quickly match the relevant emergency plan documents through similarity calculation. Compared with traditional manual retrieval, this method not only ensures the accuracy of retrieval, but also significantly improves the retrieval speed.

[0101] Secondly, the present invention introduces an innovative semantic-based text reordering method. By using a text reordering semantic analysis model, the system can sort the retrieved emergency plan documents by relevance according to the specific circumstances of the emergency event. This technological breakthrough ensures that emergency plan documents with higher relevance and stronger guidance are given priority, thereby greatly improving the accuracy and efficiency of subsequent generative artificial intelligence in formulating emergency response plans.

[0102] Finally, the system adopts a unique approach to emergency response plan and task generation. By presetting the "roles" and output formats of generative AI, the system can ensure that the generated emergency response tasks are highly relevant to the emergency and maintain a logical sequence. This approach not only improves the quality of response plans and response tasks, but also enables the format of output data to be directly processed by subsequent code, greatly improving the automation and intelligence level of the overall emergency response process.

[0103] In summary, the present invention significantly improves the efficiency, accuracy and operability of the emergency response system through an innovative combination of technologies. These improvements can not only speed up emergency response and reduce human errors, but also more effectively utilize the existing emergency knowledge base, thereby playing a key role in handling public emergencies and potentially reducing the risk of casualties and property losses.

[0104] Based on the same inventive concept, another embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory, wherein the processor executes the computer program to load the emergency response plan and task generation system based on generative artificial intelligence as described in any one of the embodiments of the present invention.

[0105] Specifically, the electronic device includes: a memory and a processor, the memory and the processor are connected via a bus communication, a computer program is stored in the memory, and the computer program can be run on the processor, thereby loading the emergency response plan and task generation system based on generative artificial intelligence as described in any one of the first aspects of the embodiments of the present invention.

[0106] The memory may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0107] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a graphics processing unit or a computing power processor (GPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0108] Based on the same inventive concept, another embodiment of the present invention provides a computer-readable storage medium having a computer program / instruction stored thereon. When the computer program / instruction is executed by a processor, it loads the emergency response plan and task generation system based on generative artificial intelligence as described in any one of the first aspects of the embodiments of the present invention.

[0109] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0110] Those skilled in the art will appreciate that embodiments of the present invention may be provided as systems, electronic devices, storage media or computer program products. Therefore, embodiments of the present invention may take the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, embodiments of the present invention may take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CDROM, optical storage, etc.) containing computer-usable program codes.

[0111] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0112] Finally, 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 terminal 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 terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0113] The above is a detailed introduction to an emergency response plan and task generation system, electronic device and readable storage medium based on generative artificial intelligence provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application; at the same time, for those of ordinary skill in the art, according to the idea of ​​the present application, there will be changes in the specific implementation methods and scope of application. In summary, the content of this specification should not be understood as a limitation on the present application. The above embodiments are only preferred embodiments cited to fully illustrate the present invention, and the scope of protection of the present invention is not limited thereto. Equivalent substitutions or changes made by those skilled in the art on the basis of the present invention are all within the scope of protection of the present invention.

Claims

1. An emergency response plan and task generation system based on generative artificial intelligence, characterized in that: include: A knowledge base search system is used to receive emergency event information input by a user through an emergency event information input interface, process the emergency event information and filter out matching emergency plan text fragments from an emergency knowledge base, and input the matching emergency plan text fragments into an emergency response plan generation system and an emergency response task generation system; An emergency response plan generation system is used to respond to a user's emergency response plan generation request, obtain the emergency event information and receive the emergency plan text fragment input by the knowledge base search system, and generate an emergency response plan according to the emergency event information and the emergency plan text fragment; The emergency response task generation system is used to respond to the user's emergency response task generation request, obtain the sudden emergency event information, the emergency response plan and receive the emergency plan text fragment input by the knowledge base search system, and generate an emergency response task based on the emergency response plan, the sudden emergency event information and the emergency plan text fragment.

2. The emergency response plan and task generation system based on generative artificial intelligence according to claim 1 is characterized in that: The knowledge base search system includes a text segmentation and embedding system, a vector data storage system, and a vector query system. The text segmentation and embedding system is used to convert the emergency event information into a corresponding emergency event information embedding vector by using a text embedding model; The vector data storage system is used to pre-store the emergency plan file name, the emergency plan text segment corresponding to the emergency plan file name, and the emergency plan text segment embedding vector corresponding to the emergency plan text segment; The vector query system is used to calculate the semantic similarity between each sudden emergency event information embedding vector and all the emergency plan text fragment embedding vectors in the vector data storage system, and screen out a preset number of emergency plan text fragment embedding vectors that are ranked at the top in semantic similarity with the sudden emergency event information embedding vector, extract the emergency plan text fragments corresponding to the preset number of emergency plan text fragment embedding vectors, use a text rearrangement semantic analysis model to sort the extracted preset number of emergency plan text fragments by relevance according to the sudden emergency event information, determine the preset number of emergency plan text fragments with the highest relevance as the emergency plan text fragments matching the sudden emergency event information, summarize the emergency plan text fragments with the highest relevance of all sudden emergency event information embedding vectors to obtain all the emergency plan text fragments matching the sudden emergency event information, and input all the emergency plan text fragments matching the sudden emergency event information into an emergency response plan generation system and an emergency response task generation system.

3. The emergency response plan and task generation system based on generative artificial intelligence according to claim 1 is characterized in that: The knowledge base search system includes a text segmentation and embedding system and a vector data storage system. The text segmentation and embedding system is also used to process emergency plan files, using a fixed-length segmentation method based on line breaks or a semantic-based segmentation method to segment the file content of the emergency plan file into multiple emergency plan text segments, and using a text embedding model to convert the emergency plan text segments into corresponding emergency plan text segment embedding vectors, and input the emergency plan file name, the emergency plan text segment corresponding to the emergency plan file name, and the emergency plan text segment embedding vector corresponding to the emergency plan text segment into a vector data storage system; The vector data storage system is used to store and manage the received emergency plan file name, the emergency plan text segment corresponding to the emergency plan file name, and the emergency plan text segment embedding vector corresponding to the emergency plan text segment.

4. The emergency response plan and task generation system based on generative artificial intelligence according to claim 1 is characterized in that: The emergency response plan generation system includes a generative artificial intelligence role defined by prompt words; the emergency response task generation system includes a generative artificial intelligence role defined by prompt words and a fixed predefined task format output; the generative artificial intelligence role defined by the prompt words is set to play the role of an emergency response commander through a preset large language model prompt word.

5. The emergency response plan and task generation system based on generative artificial intelligence according to claim 1 is characterized in that: The emergency response plan generation system is used to parse the emergency event information to obtain the time, location, event type, impact scope, and number of people affected, and adapt, integrate and output the time, location, event type, impact scope, number of people affected, and the measures and resource requirements in the emergency plan text segment to obtain the emergency response plan.

6. The emergency response plan and task generation system based on generative artificial intelligence according to claim 1 is characterized in that: The emergency response task generation system is used to parse each plan item of the emergency response plan, and combine the sudden emergency event information and the emergency plan text fragment to determine the task item information corresponding to each plan item, and output it in a predefined task format to obtain the emergency response task; the task item information includes the execution unit, task objectives, required manpower, required materials, and estimated completion time.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the electronic device executes the computer program, it loads the emergency response plan and task generation system based on generative artificial intelligence as described in any one of claims 1 to 6.

8. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the emergency response plan and task generation system based on generative artificial intelligence as described in any one of claims 1 to 6 is loaded.

Images