Intelligent alarm receiving and processing guiding method and system

Abstract

The invention provides an intelligent alarm receiving and processing guiding method and system, and the method comprises the steps: recognizing the case type and the guiding type of a current alarm case, and obtaining a locally cached question tree and a guiding template based on the case type and the guiding type; based on the question tree and the guide template, obtaining answer information of an alarm person; and aggregating the answer information of different types of alarm personnel to generate an alarm receiving and processing guidance summary, and pushing the summary to the alarm client and the alarm receiving client. The invention provides an intelligent alarm receiving and processing method and system combining a structured question and answer tree, AI induction and summarization and knowledge recommendation, and rapid, accurate and standardized guidance is realized.

Description

Technical Field

[0001] This invention relates to the fields of fire protection information technology and emergency command technology, and in particular to an intelligent alarm response guidance method and system. Background Technology

[0002] In the existing emergency response process, dispatchers typically rely on their experience to question callers over the phone and determine the nature of the incident and the appropriate dispatch plan based on limited information. Especially in scenarios involving multiple calls or complex, sudden incidents, traditional methods are no longer sufficient to meet the demands for efficient information aggregation and intelligent handling guidance. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent alarm handling guidance method and system to solve the above-mentioned problems.

[0004] The technical solution provided by this invention is as follows: In a first aspect, the present invention provides an intelligent alarm handling guidance method, comprising: Identify the case type and guidance type of the current alarm case, and obtain the locally cached problem tree and guidance template based on the case type and guidance type; Based on the question tree and the guidance template, obtain the answer information from the person who made the alarm; The system aggregates the responses from different types of alarm users to generate a summary of alarm handling guidelines, which is then pushed to both the alarm client and the alarm receiving client.

[0005] In one implementation, aggregating the answer information from the different types of alarm users to generate a summary of alarm handling guidance and response strategies includes: Based on the case type, the guidance type, and the type of the person reporting the incident, different prompt words are constructed; The large language model is invoked to perform semantic summarization on the answer information of the alarm user and the prompt words, and to generate a summary of alarm handling guidance. The key information in the answer information of the alarm caller is extracted through the RAG process. Based on the key information in the answer information of the alarm caller, the knowledge base is searched to obtain knowledge recommendation information to be added to the alarm handling guidance summary.

[0006] In one implementation, the step of retrieving key information from the answer information of the person making the alarm to obtain knowledge recommendation information from the knowledge base includes: Vector retrieval is used to match the key information to generate knowledge recommendation information, which includes laws and regulations, emergency response plans, historical cases, text and image materials, and video tutorials.

[0007] In one implementation, obtaining the answer information of the person making the alarm based on the question tree and the guidance template includes: Based on the response status of the person who made the alarm, the question nodes and corresponding question content of the question tree are adjusted in real time.

[0008] In one implementation, it also includes: The alarm handling client obtains the alarm handling instructions summary for the current alarm user and the alarm handling instructions summary for the current alarm case, and performs real-time analysis and comparison.

[0009] Secondly, this application also provides an intelligent alarm handling guidance system, including: The problem tree management module is used to identify the case type and guidance type of the current alarm case, and to obtain the locally cached problem tree and guidance template based on the case type and guidance type; The acquisition module is used to obtain the answer information of the alarm user based on the question tree and the guidance template; The guidance module is used to aggregate the answer information of different types of alarm users to generate a summary of alarm handling guidance and push it to the alarm client and the alarm receiving client.

[0010] In one implementation, the guidance module is further configured to: Based on the case type, the guidance type, and the type of the person reporting the incident, different prompt words are constructed; The large language model is invoked to perform semantic summarization on the answer information of the alarm user and the prompt words, and to generate a summary of alarm handling guidance. The key information in the answer information of the alarm caller is extracted through the RAG process. Based on the key information in the answer information of the alarm caller, the knowledge base is searched to obtain knowledge recommendation information to be added to the alarm handling guidance summary.

[0011] In one implementation, the guidance module is further configured to: Vector retrieval is used to match the key information to generate knowledge recommendation information, which includes laws and regulations, emergency response plans, historical cases, text and image materials, and video tutorials.

[0012] In one implementation, the acquisition module is further configured to: Based on the response status of the person who made the alarm, the question nodes and corresponding question content of the question tree are adjusted in real time.

[0013] In one implementation, it also includes: a comparison module, used for: The alarm handling client obtains the alarm handling instructions summary for the current alarm user and the alarm handling instructions summary for the current alarm case, and performs real-time analysis and comparison.

[0014] This invention utilizes a three-pronged approach—a question-and-answer tree for emergency calls, Large Language Model (LLM) summarization, and Retrieval-Enhanced Generation (RAG) knowledge base recommendation—to achieve standardized questioning and intelligent generation of response guidelines. It also incorporates local and remote multi-level caching to improve response performance. Through this system, dispatchers can quickly ask and record questions guided by a tree structure. The system automatically synthesizes and provides feedback, pushing professional response suggestions to both the dispatcher and on-site firefighters. The system can be widely applied in emergency scenarios such as firefighting, traffic emergencies, and hazardous materials emergencies. Attached Figure Description

[0015] The preferred embodiments will be described below in a clear and easy-to-understand manner, with reference to the accompanying drawings, to further explain the above-mentioned characteristics, technical features, advantages, and implementation methods of an intelligent alarm handling guidance method and system.

[0016] Figure 1 This is a schematic diagram of an embodiment of an intelligent alarm handling guidance system of the present invention; Figure 2 This is a schematic diagram of another embodiment of the intelligent emergency response guidance system of the present invention.

[0017] Figure 3 This is a flowchart of an embodiment of an intelligent alarm handling guidance method of the present invention.

[0018] Figure 4 This is a flowchart of another embodiment of the intelligent alarm handling guidance method of the present invention. Detailed Implementation

[0019] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0020] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or sets.

[0021] To keep the drawings concise, only the parts relevant to the invention are shown schematically in each figure, and they do not represent the actual structure of the product. Furthermore, for ease of understanding, in some figures, only one of components with the same structure or function is shown schematically, or only one is labeled. In this document, "one" can mean not only "only one" but also "more than one".

[0022] It should also be further understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0023] Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the specific implementation methods of the present invention will be described below with reference to the accompanying drawings. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings and other implementation methods can be obtained based on these drawings without any creative effort.

[0025] Because dispatchers rely on their human experience to question callers and determine the nature of the incident and dispatch plan based on limited information, this model has the following problems: 1. Inconsistent inquiry process: Different dispatchers ask different questions, which may lead to the omission of key information.

[0026] 2. Delayed knowledge acquisition: When encountering special scenarios, it is necessary to look up information, resulting in slow response speed.

[0027] 3. High reliance on experience: Novice dispatchers need a long period of training to become proficient.

[0028] 4. Low data utilization: Historical case data has not been effectively accumulated and cannot provide reference for new cases.

[0029] 5. Low handling efficiency: There is a lack of effective handling guidelines before arrival and a lack of historical experience to support on-site handling, which affects the corresponding efficiency.

[0030] In view of the shortcomings of existing technologies, this invention provides an intelligent emergency response system that combines structured question-and-answer trees, AI inductive summarization, and knowledge recommendation to achieve fast, accurate, and standardized guidance.

[0031] Reference Appendix Figure 1 The overall architecture diagram of the intelligent alarm handling guidance system of the present invention is shown below. First, the intelligent alarm handling guidance system of the present invention adopts a layered architecture, including: 1) The alarm response question-and-answer engine layer functions as follows: a question tree-structured model that supports multi-branch conditional navigation. Questions are associated with case types and guidance types, and are dynamically loaded.

[0032] 2) The intelligent analysis layer's function is to invoke the LLM (Local Management Model) to semantically summarize the alarm user's responses, generating a question-and-answer summary and actionable handling suggestions. Using the RAG (Relevant Information Group) model, it retrieves relevant regulations and contingency plans from the knowledge base to supplement the guidelines.

[0033] 3) Data access and caching layer: Its functions are as follows: Local cache (Caffeine) stores frequently accessed question-answer trees and guidance templates, reducing database access. Remote cache (Redis) serves as a cluster-shared cache, ensuring data consistency across multiple instances.

[0034] 4) Interface and Presentation Layer: Its functions include providing web-based and command terminal interfaces, supporting Q&A visualization, exporting guidance documents, and viewing recommended knowledge. It also provides a REST API for third-party access.

[0035] Among them, LLM (Large Language Model) is a giant artificial intelligence model that learns the statistical laws of language by being trained on massive amounts of text data, thereby being able to understand and generate natural language.

[0036] RAG stands for Retrieval-Augmented Generation. It is a technical framework that combines information retrieval systems with the generative capabilities of Large Language Models (LLM). Its core purpose is to enable LLM to reference authoritative and up-to-date information from external knowledge bases when generating answers, thereby producing more accurate and reliable responses.

[0037] The following description is in conjunction with the accompanying drawings: Reference Appendix Figure 2 This invention provides an intelligent alarm handling guidance system, comprising: The problem tree management module 101 is used to identify the case type and guidance type of the current alarm case, and to obtain the locally cached problem tree and guidance template based on the case type and guidance type.

[0038] The question tree management module also supports configuring question nodes and jump conditions according to case type and handling object, and supports import / export and version management.

[0039] The acquisition module 102 is used to obtain the answer information of the alarm personnel based on the question tree and the guidance template.

[0040] In this embodiment, the acquisition module 102 serves as a real-time alarm response module, capable of guiding the dispatcher to ask questions and recording answers according to a question-and-answer tree. The questions guided by the question-and-answer tree support hierarchical progression, conditional branching, and mandatory / optional answer configurations. Simultaneously, multiple callers can submit questions and answers for the same case, and the alarm response guidance system of this invention aggregates all feedback answer information. This real-time alarm response module also supports triggering immediate warnings and prompts with key answers (such as "Someone is trapped" triggering priority for life-saving rescue).

[0041] In one implementation, the acquisition module is further configured to: adjust the question nodes and corresponding question content of the question tree in real time based on the response status of the alarm person.

[0042] Specifically, the acquisition module can automatically load customized question trees and guidance templates for different types of cases, and adjust the order and content of subsequent questions in real time based on each step of the complainant's response, avoiding redundant questioning from a fixed script. In other words, this acquisition module can achieve dynamic loading of the question-and-answer tree driven by case type.

[0043] The guidance module 103 is used to aggregate the answer information of different types of alarm users to generate a summary of alarm handling guidance and push it to the alarm client and the alarm receiving client.

[0044] Specifically, the guidance module is also used for: Based on the case type, the guidance type, and the type of the person reporting the incident, different prompt words are constructed; a large language model is invoked to semantically summarize the answer information of the person reporting the incident and the prompt words, and a summary of the emergency response guidance is generated; key information in the answer information of the person reporting the incident is extracted through the RAG process, and a knowledge base is retrieved based on the key information in the answer information of the person reporting the incident to obtain knowledge recommendation information to be added to the summary of the emergency response guidance.

[0045] Specifically, the guidance module includes: an AI-based induction and guidance summary generation module: through multiple callers' answers, the system automatically aggregates different information points; constructs a prompt template, categorizing and organizing it according to the caller's role (informed person, injured person, passerby); and calls an LLM model to generate a structured "question and answer summary" for subsequent dispatch reference.

[0046] The guidance module is further configured to: use vector retrieval to match the key information to generate knowledge recommendation information, which includes laws and regulations, contingency plans, historical cases, graphic materials, and video tutorials.

[0047] In this embodiment, the guidance module also includes a knowledge base and a RAG recommendation module. The knowledge base stores laws and regulations, operational plans, historical cases, textual and graphic materials, video tutorials, etc. The knowledge base and the RAG recommendation module use vector retrieval to match the current case context and generate knowledge recommendations. For example, in the case of trapped personnel, it can also generate evacuation suggestions or emergency escape plans.

[0048] In one implementation, it further includes a comparison module, used to: obtain the summary of emergency response instructions for the current alarm user and the summary of emergency response instructions for the current alarm case through the alarm handling client, and perform real-time analysis and comparison.

[0049] Specifically, this comparison module serves as a statistics and analysis module, used to statistically analyze case types, the number of guidance generated, and the knowledge base hit rate. Through this module, dispatchers can obtain guidance summaries from the current caller and others, enabling real-time information analysis and comparison.

[0050] Based on the above functional modules, the alarm handling guidance system of the present invention further includes: Multi-level caching modules: Caffeine local cache: millisecond-level access speed. Redis remote cache: supports sharing across multiple instances.

[0051] This multi-level caching module achieves high performance and low latency: the dual-caching architecture ensures millisecond-level response, meeting the extremely high real-time requirements of alarm receiving scenarios. Furthermore, the system is easily expandable: it supports the rapid integration of multiple case types and new business modules.

[0052] Dispatch module: Summary content can be pushed to the caller's mobile phone (SMS / WeChat, etc.); it can also be pushed to the mobile terminal of the dispatching team to provide information for on-site decision-making.

[0053] This invention provides an intelligent fire emergency response guidance system. It utilizes a three-pronged approach—a question-and-answer tree, Large Language Model (LLM) summarization, and Retrieval-Enhanced Generation (RAG) knowledge base recommendation—to achieve standardized questioning and intelligent generation of response guidance. Furthermore, it incorporates local and remote multi-level caching to improve response performance. Through this system, dispatchers can quickly ask and record questions guided by a tree structure. The system automatically synthesizes and provides feedback, pushing professional handling suggestions to both the dispatcher and firefighters on the scene. Moreover, this intelligent emergency response guidance system can be widely applied to emergency scenarios such as fire fighting, traffic emergencies, and hazardous materials emergencies.

[0054] Based on the above embodiments, the corresponding method flow will be described below with reference to the accompanying drawings: Please refer to Figure 3 This is a flowchart of an intelligent alarm handling guidance method provided in an embodiment of this application. Figure 3 As shown, the method includes the following steps: S101 identifies the case type and guidance type of the current alarm case, and obtains the locally cached problem tree and guidance template based on the case type and guidance type.

[0055] In this embodiment, the case types specifically include: fire fighting, disaster relief, and social assistance. Different case types correspond to different guidance types. The local cache (Caffeine) stores frequently accessed question-and-answer trees and guidance templates for later use, reducing database access. Simultaneously, a dual-layer local and remote caching system is implemented, significantly improving system response speed while ensuring data consistency.

[0056] For example, the guidelines for fire fighting include: 1. Fires in street-front shops; 2. Fires in high-rise buildings; 3. Fires involving electric bicycles; 4. Fires in factories and warehouses; 5. ... For example, the corresponding guidance types for emergency rescue include: 1. High-altitude rescue; 2. Mountain rescue; 3. Traffic accidents; 4. Elevator malfunctions; 5. ... In this embodiment, since there are different guidance types for different case types, there will be different question trees and guidance modules for different guidance types.

[0057] S102 obtains the answer information of the alarm personnel based on the question tree and the guidance template.

[0058] In this embodiment, a question tree and guidance template are used to efficiently guide the caller to ask questions, reducing the omission of key information.

[0059] The following are some issues related to fires in ordinary residential buildings: If the person reporting the incident is a passerby: 1. Did you see anyone calling for help? 2. Did you see any open flame or smoke? If the answer is: there is an open flame, then the next question is: is there a tendency for it to spread? If the answer is: there is smoke, then the next question is: is it black smoke or white smoke? 3. On what floor did the fire likely start? You can choose to fill in a number or answer "I don't know".

[0060] 4. How many basements does the building have? You can choose to fill in a number or answer "I don't know".

[0061] In this embodiment, a hierarchical question guidance template enables the timely acquisition of accurate information, avoiding some existing problems associated with manual experience-based questions. Simultaneously, it standardizes questioning, reduces human oversights, and improves the quality of alarm reception.

[0062] S103 aggregates the answer information from different types of alarm users to generate a summary of alarm handling guidance and pushes it to the alarm client and the alarm receiving client.

[0063] Specifically, different types of guidance have different roles, for example: 1. In the case of a fire in a typical residential building, the roles of the person reporting the fire include: passersby, homeowners, property management staff, and neighbors.

[0064] 2. When trapped in an elevator, the roles of the person who calls the police include: the trapped person, and someone outside the elevator who knows the situation.

[0065] In this embodiment, the alarm caller information is tagged (informed person, injured person, passerby, property management, etc.), and different Prompt templates are built according to roles. Multiple summaries are generated for manual selection or fusion, so as to have a more comprehensive understanding of the situation on site, realize the aggregation of multiple alarms and role differentiation, and improve the accuracy of decision-making by fusion and judgment of information from multiple parties.

[0066] This invention can automatically push the summary of emergency response guidelines to the alarm caller and rescue personnel, allowing rescue personnel to obtain sufficient situation prediction before arriving at the scene, improving pre-arrival preparation and reducing the risk of casualties and misjudgments.

[0067] In one implementation, aggregating the answer information from the different types of alarm users to generate a summary of alarm handling guidance and response strategies includes: Based on the case type, the guidance type, and the type of the person reporting the incident, different prompt words are constructed; a large language model is invoked to semantically summarize the answer information of the person reporting the incident and the prompt words, and a summary of the emergency response guidance is generated; key information in the answer information of the person reporting the incident is extracted through the RAG process, and a knowledge base is retrieved based on the key information in the answer information of the person reporting the incident to obtain knowledge recommendation information to be added to the summary of the emergency response guidance.

[0068] Among them, the Large Language Model (LLM) can be prompted to directly generate explicit tree structures because tree structures can be clearly represented in text form (such as JSON, YAML, indentation format).

[0069] In the practical scenario of this embodiment, the specific combination of the tree-structured problem model and the large language model involves constructing corresponding prompt words based on the case type and guidance type, and then sending the question-and-answer records and prompt words from the alarm guidance to the large model to generate corresponding question-and-answer summaries. During this process, some laws and regulations and historical experience are also incorporated to generate intelligent supplementary information.

[0070] In this embodiment, intelligent guidance is implemented. After completing the question and answer, a concise and structured alarm guidance summary is automatically generated using LLM+RAG, which supports the duty officer to make quick decisions and improves the response speed.

[0071] This invention combines three technologies—structured question answering, LLM induction, and RAG recommendation—in a unified alarm response mode for the first time, ensuring standardized questioning while providing intelligent analysis and knowledge supplementation.

[0072] The step of retrieving key information from the answer information of the person who made the alarm to obtain knowledge recommendation information from the knowledge base includes: Vector retrieval is used to match the key information to generate knowledge recommendation information, which includes laws and regulations, emergency response plans, historical cases, text and image materials, and video tutorials.

[0073] For example, in real-world scenarios, there are alarm situations involving hazardous materials: The person who called the police described: "I was on the road next to a chemical plant when I saw a tanker truck leaking liquid. There was a very pungent smell, and some people around me felt dizzy!" RAG will extract key information such as chemical plant, liquid leak, and personnel discomfort.

[0074] And perform a knowledge base search: For example, the provisions on emergency response to leaks in the "Regulations on the Safety Management of Hazardous Chemicals" were found; The search revealed the golden rules for handling "unknown chemical leaks" in the "Fire Response Plan": evacuation, isolation, fire prevention, and reporting procedures. A search of chemical databases revealed that a "pungent odor" may be associated with irritating and hazardous chemicals such as ammonia and chlorine, along with preliminary protective measures. This information will be displayed on the emergency call client's screen and will alert the dispatcher.

[0075] This invention also provides a continuously evolving knowledge base, supporting the automatic accumulation of experience in handling new cases into the knowledge base and its participation in subsequent retrieval and recommendation. This embodiment achieves real-time knowledge dissemination, eliminating the need for manual searching, as the system automatically supplements relevant regulations and contingency plans.

[0076] In one implementation, obtaining the answer information of the person making the alarm based on the question tree and the guidance template includes: Based on the response status of the person who made the alarm, the question nodes and corresponding question content of the question tree are adjusted in real time.

[0077] In this embodiment, dynamic loading of question-and-answer trees driven by case type enables the automatic loading of exclusive question trees and guidance templates for different types of cases. The order and content of subsequent questions are adjusted in real time according to each step of the caller's answer, avoiding redundant questioning from a fixed script.

[0078] In one implementation, it also includes: The alarm handling client obtains the alarm handling instructions summary for the current alarm user and the alarm handling instructions summary for the current alarm case, and performs real-time analysis and comparison.

[0079] In this embodiment, the summary of the emergency response guidelines will be automatically pushed to the person who reported the incident and the rescue personnel, thereby improving the preparedness before arrival and reducing the risk of casualties and misjudgment.

[0080] Reference Appendix Figure 4 The alarm handling flowchart described above involves the following steps: When an alarm is received, the case type is identified, and the corresponding question tree (local + Redis cache) is obtained; the question-and-answer process begins, with hierarchical question guidance; multiple alarm responses (from different roles) are aggregated; prompt words are constructed, and a summary is generated using LLM; RAG is invoked to obtain recommended solutions for similar historical cases; and guidance content in text, image, and video formats is generated and pushed to the terminals of the alarm caller and rescue personnel.

[0081] This invention can efficiently guide alarm receivers to ask questions, reducing the omission of key information; integrate information from multiple sources to improve decision-making accuracy; provide rescuers with sufficient situational predictions before they arrive at the scene; provide intelligent guidance content to assist in rescue decision-making; the system is stable and reliable, supporting high-concurrency alarm receiving scenarios; and provides technical support for establishing an intelligent, standardized, and efficient fire alarm receiving system.

[0082] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of program modules is merely an example. In practical applications, the above functions can be assigned to different program modules as needed, that is, the internal structure of the device can be divided into different program units or modules to complete all or part of the functions described above. The program modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software program unit. Furthermore, the specific names of the program modules are only for easy differentiation and are not intended to limit the scope of protection of this application.

[0083] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0084] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0085] In the embodiments provided in this application, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings or direct couplings or communication connections may be through some interfaces, and the indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0086] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0087] Furthermore, the functional units in the various embodiments of this application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The integrated unit described above can be implemented in hardware or as a software functional unit.

[0088] It should be noted that the above embodiments can be freely combined as needed. The above description is only a preferred embodiment of the present invention. It should be pointed out that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent alarm handling guidance, characterized in that, include: Identify the case type and guidance type of the current alarm case, and obtain the locally cached problem tree and guidance template based on the case type and guidance type; Based on the question tree and the guidance template, obtain the answer information from the person who made the alarm; The system aggregates the responses from different types of alarm users to generate a summary of alarm handling guidelines, which is then pushed to both the alarm client and the alarm receiving client.

2. The intelligent alarm handling guidance method according to claim 1, characterized in that, The aggregation of answer information from the different types of alarm users to generate a summary of alarm handling guidance and response strategies includes: Based on the case type, the guidance type, and the type of the person reporting the incident, different prompt words are constructed; The large language model is invoked to perform semantic summarization on the answer information of the alarm user and the prompt words, and to generate a summary of alarm handling guidance. The key information in the answer information of the alarm caller is extracted through the RAG process. Based on the key information in the answer information of the alarm caller, the knowledge base is searched to obtain knowledge recommendation information to be added to the alarm handling guidance summary.

3. The intelligent alarm handling guidance method according to claim 2, characterized in that, The step of retrieving key information from the answer information of the person who made the alarm to obtain knowledge recommendation information from the knowledge base includes: Vector retrieval is used to match the key information to generate knowledge recommendation information, which includes laws and regulations, emergency response plans, historical cases, text and image materials, and video tutorials.

4. The intelligent alarm handling guidance method according to claim 1, characterized in that, The step of obtaining the answer information of the person making the alarm based on the question tree and the guidance template includes: Based on the response status of the person who made the alarm, the question nodes and corresponding question content of the question tree are adjusted in real time.

5. The intelligent alarm handling guidance method according to any one of claims 1 to 4, characterized in that, Also includes: The alarm handling client obtains the alarm handling instructions summary for the current alarm user and the alarm handling instructions summary for the current alarm case, and performs real-time analysis and comparison.

6. An intelligent alarm handling guidance system, characterized in that, include: The problem tree management module is used to identify the case type and guidance type of the current alarm case, and to obtain the locally cached problem tree and guidance template based on the case type and guidance type; The acquisition module is used to obtain the answer information of the alarm user based on the question tree and the guidance template; The guidance module is used to aggregate the answer information of different types of alarm users to generate a summary of alarm handling guidance and push it to the alarm client and the alarm receiving client.

7. The intelligent alarm handling guidance system according to claim 6, characterized in that, The guidance module is also used for: Based on the case type, the guidance type, and the type of the person reporting the incident, different prompt words are constructed; The large language model is invoked to perform semantic summarization on the answer information of the alarm user and the prompt words, and to generate a summary of alarm handling guidance. The key information in the answer information of the alarm caller is extracted through the RAG process. Based on the key information in the answer information of the alarm caller, the knowledge base is searched to obtain knowledge recommendation information to be added to the alarm handling guidance summary.

8. The intelligent alarm handling guidance system according to claim 7, characterized in that, The guidance module is also used for: Vector retrieval is used to match the key information to generate knowledge recommendation information, which includes laws and regulations, emergency response plans, historical cases, text and image materials, and video tutorials.

9. The intelligent alarm handling guidance system according to claim 6, characterized in that, The acquisition module is also used for: Based on the response status of the person who made the alarm, the question nodes and corresponding question content of the question tree are adjusted in real time.

10. The intelligent alarm handling guidance system according to any one of claims 6 to 9, characterized in that, Also includes: The comparison module is used for: The alarm handling client obtains the alarm handling instructions summary for the current alarm user and the alarm handling instructions summary for the current alarm case, and performs real-time analysis and comparison.

Images