Alarm processing method, device and electronic equipment applied to energy storage system

Abstract

The present application discloses an alarm processing method, device and electronic device applied to an energy storage system, and belongs to the field of computer technology. The method includes: obtaining a target alarm problem for a target energy storage system; calling a first target tool through an intelligent agent, and generating reply data for the target alarm problem based on the first data returned by the first target tool; the first target tool is used to process the second data corresponding to the target alarm problem and obtain the first data; the first target tool includes at least one tool in a pre-established tool library. The alarm processing method applied to an energy storage system disclosed in the present application is used to improve the intelligence level of alarm problem processing in an energy storage system.

Description

Technical Field

[0001] The present application belongs to the field of computer technology, and in particular, relates to an alarm processing method, device and electronic equipment applied to an energy storage system. Background Art

[0002] In energy storage systems, system alarm processing is an important part of ensuring system stability and security. Generally, users view system alarm issues on the client and manually select the alarm processing tool corresponding to the alarm issue. The client then applies deep learning, machine learning or other big data analysis technologies to process system alarm issues based on the user's selection or other operations.

[0003] However, as the complexity of the system increases, the level of intelligence in handling system alarm problems through manual selection by users and then applying deep learning, machine learning or other big data analysis technologies is very limited, making handling alarm problems more difficult and time-consuming. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes an alarm processing method, device and electronic device applied to an energy storage system to improve the intelligent level of alarm problem processing of the energy storage system.

[0005] In a first aspect, the present application provides an alarm processing method applied to an energy storage system, the method comprising:

[0006] Obtain target alarm issues for target energy storage systems;

[0007] The first target tool is called by an intelligent agent, and response data for a target alarm problem is generated based on the first data returned by the first target tool; the first target tool is used to process the second data corresponding to the target alarm problem and obtain the first data; the first target tool includes at least one tool in a pre-established tool library.

[0008] According to the alarm processing method applied to the energy storage system of the present application, a target alarm problem for a target energy storage system is obtained; a first target tool is called by an intelligent body, and response data for the target alarm problem is generated based on the first data returned by the first target tool. This allows the user to not need to manually select an alarm problem processing tool. The first target tool for processing the second data corresponding to the target alarm problem and obtaining the first data can be called by the intelligent body to generate response data for the target alarm problem, thereby significantly improving the intelligence level of alarm problem processing for the energy storage system.

[0009] According to an embodiment of the present application, the first target tool is called by an intelligent agent, and based on the first data returned by the first target tool, response data for the target alarm problem is generated, including:

[0010] The agent performs the following operations:

[0011] Determine a first target tool, and obtain input parameters based on the target alarm problem;

[0012] Calling a first target tool and inputting input parameters into the first target tool, so that the first target tool acquires first data based on the input parameters;

[0013] First data is received, and reply data is generated based on the first data.

[0014] According to one embodiment of the present application, generating reply data based on the first data includes:

[0015] Based on the first data, generate a target text;

[0016] Check whether the target text matches the target alarm question;

[0017] In the case where the target text matches the target alarm question, the target text is determined as the reply data.

[0018] According to one embodiment of the present application, after checking whether the target text matches the target alarm question, the method further includes:

[0019] When the target text does not match the target alarm problem, the second target tool is called by the intelligent agent to generate reply data based on the third data returned by the second target tool; the second target tool is used to process the fourth data corresponding to the target alarm problem to obtain the third data; the second target tool is different from the first target tool; the second target tool includes at least one tool in a pre-established tool library.

[0020] According to an embodiment of the present application, checking whether the target text matches the target alarm question includes:

[0021] Obtain target corpus corresponding to the target alarm problem in the target knowledge base;

[0022] Based on the target corpus, check whether the target text matches the target alarm question.

[0023] According to one embodiment of the present application, based on the target corpus, checking whether the target text matches the target alarm question includes:

[0024] The target corpus, target text and target warning question are input into the trained large language model included in the intelligent agent, and the test result output by the large language model is obtained; the test result is used to indicate whether the target text matches the target warning question.

[0025] According to one embodiment of the present application, determining a first target tool includes:

[0026] The target warning problem is processed based on the trained thinking chain model included in the intelligent agent, and the first target tool is determined based on the output of the thinking chain model.

[0027] According to one embodiment of the present application, determining a first target tool includes:

[0028] Performing divergent processing on the target alarm problem to obtain at least one divergent problem;

[0029] Each divergent problem is processed separately based on the trained thinking chain model included in the intelligent agent, and the first target tool is determined based on the output of the thinking chain model.

[0030] According to one embodiment of the present application, calling a first target tool includes:

[0031] A first target application programming interface is determined and accessed, and a first target tool is invoked.

[0032] In a second aspect, the present application provides an alarm processing device for an energy storage system, the device comprising:

[0033] An acquisition module, used to acquire target alarm problems for a target energy storage system;

[0034] The first generation module is used to call the first target tool through the intelligent agent, and generate response data for the target alarm problem based on the first data returned by the first target tool; the first target tool is used to process the second data corresponding to the target alarm problem to obtain the first data; the first target tool includes at least one tool in a pre-established tool library.

[0035] According to the alarm processing device applied to the energy storage system of the present application, a target alarm problem for a target energy storage system is obtained; a first target tool is called by an intelligent body, and response data for the target alarm problem is generated based on the first data returned by the first target tool. This allows the user to not need to manually select an alarm problem processing tool. The first target tool for processing the second data corresponding to the target alarm problem and obtaining the first data can be called by the intelligent body to generate response data for the target alarm problem, thereby significantly improving the intelligence level of alarm problem processing in the energy storage system.

[0036] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the computer program, the alarm processing method applied to the energy storage system as described in the first aspect above is implemented.

[0037] In a fourth aspect, the present application provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the alarm processing method applied to an energy storage system as described in the first aspect above is implemented.

[0038] In a fifth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the alarm processing method applied to an energy storage system as described in the first aspect above.

[0039] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0041] Figure 1 It is one of the flowcharts of the alarm processing method applied to the energy storage system provided in the embodiment of the present application;

[0042] Figure 2 This is a second flow chart of an alarm processing method applied to an energy storage system provided in an embodiment of the present application;

[0043] Figure 3 is a structural diagram of an alarm processing device applied to an energy storage system provided in an embodiment of the present application;

[0044] Figure 4 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.

[0046] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0047] In the related technology, users can view the alarm problems of the energy storage system on the client and manually select the alarm processing tool corresponding to the alarm problem. The client obtains all the operating data of the energy storage equipment according to the user's selection or other operations, and then applies deep learning, machine learning or other big data analysis technologies to handle the system alarm problem. However, with the increase in system complexity, the level of intelligence of this method of handling system alarm problems is very limited, making it more difficult and time-consuming to handle alarm problems. Moreover, this method of extracting all the operating data of the energy storage device and handing it over to the large model for processing cannot analyze and process all the operating data of the energy storage device in a more fine-grained manner, causing the processing process to occupy more resources. In addition, large models are prone to errors in the process of handling system alarm problems.

[0048] In conjunction with the accompanying drawings, the following describes in detail the alarm processing method for an energy storage system, the alarm processing device for an energy storage system, the electronic device, and the readable storage medium provided in the embodiments of the present application through specific embodiments and their application scenarios.

[0049] Among them, the alarm processing method applied to the energy storage system can be applied to the terminal, and can be specifically executed by hardware or software in the terminal.

[0050] The terminal includes, but is not limited to, a portable communication device such as a mobile phone or tablet computer with a touch-sensitive surface (e.g., a touch screen display and / or a touch pad). It should also be understood that in some embodiments, the terminal may not be a portable communication device, but a desktop computer with a touch-sensitive surface (e.g., a touch screen display and / or a touch pad).

[0051] In the following various embodiments, a terminal including a display and a touch-sensitive surface is described. However, it should be understood that the terminal may include one or more other physical user interface devices such as a physical keyboard, a mouse and a joystick.

[0052] The alarm processing method for an energy storage system provided in the embodiment of the present application may be executed by an electronic device or a functional module or functional entity in the electronic device that can implement the alarm processing method for an energy storage system. The electronic devices mentioned in the embodiment of the present application include but are not limited to mobile phones, tablet computers, computers, cameras, and wearable devices. The alarm processing method for an energy storage system provided in the embodiment of the present application is described below by taking the electronic device as an example of the execution subject.

[0053] like Figure 1 As shown, the alarm processing method applied to the energy storage system includes: step 110 and step 120.

[0054] Step 110: Obtain target alarm problems for the target energy storage system.

[0055] In actual implementation, the target alarm problem for the target energy storage system can be obtained based on the user terminal. The user terminal can be an electronic device such as a smart phone, a tablet, a computer, or any theoretically feasible electronic device, which is not specifically limited in this application.

[0056] In some embodiments, the target alarm problem for the target energy storage system can be obtained by obtaining the target input of the user at the user terminal. The target input can be any theoretically feasible input method such as voice input, text input, and picture input.

[0057] In actual implementation, the target alarm problem can be an alarm problem related to the operation of each device in the target energy storage system, such as arc-fault circuit-interrupter alarm, cycle life alarm, gateway alarm, alarm trend analysis and other issues.

[0058] Step 120: Call the first target tool through the intelligent agent, and generate response data for the target alarm problem based on the first data returned by the first target tool; the first target tool is used to process the second data corresponding to the target alarm problem and obtain the first data; the first target tool includes at least one tool in a pre-established tool library.

[0059] In actual execution, the agent may be an agent program that uses a large language model (such as GPT-4, etc.) as the core AI engine to support natural language understanding and generation. The first target tool may be a tool for processing a target alarm problem, and the first target tool may include at least one tool in a pre-established tool library.

[0060] In some embodiments, the intelligent agent may select and call a first target tool from a pre-established tool library based on the acquired target alarm question, so as to generate response data corresponding to the target alarm question based on the first target tool.

[0061] In actual execution, the intelligent agent may include a pre-trained thought chain model, and the intelligent agent may determine the first target tool from a pre-established tool library according to the target alarm problem through the thought chain model.

[0062] In some embodiments, the agent may call the first target tool based on an application programming interface corresponding to the first target tool.

[0063] In some embodiments, after obtaining a target alarm problem for a target energy storage system, a first target tool can be determined through an intelligent agent, and input parameters can be obtained based on the target alarm problem; the first target tool can be called, and the input parameters can be input into the first target tool so that the first target tool obtains first data based on the input parameters; the first data is received, and reply data is generated based on the first data.

[0064] In actual execution, the agent can pre-store prompt words. After obtaining the target alarm problem for the target energy storage system, the agent can process and analyze the target alarm problem, obtain input parameters, and determine and call the first target tool based on the input parameters. In some embodiments, the prompt words can be created based on the Agent architecture of MetaGPT and LangChain.

[0065] In some embodiments, the first target tool may obtain second data corresponding to the target alarm problem according to input parameters obtained based on the target alarm problem. For example, if the target alarm problem is an AFCI alarm problem, the first target tool may obtain second data corresponding to the AFCI alarm problem according to input parameters obtained based on the target alarm problem.

[0066] In some embodiments, the intelligent agent can use a large language model to organize the corpus based on the first data to generate a reply text for the target alarm question, and perform a discriminative evaluation on the reply text, and determine whether to determine the reply text as the reply data for the target alarm question based on the evaluation results.

[0067] According to the alarm processing method applied to the energy storage system of the embodiment of the present application, a target alarm problem for a target energy storage system is obtained; a first target tool is determined through an intelligent agent, and input parameters are obtained based on the target alarm problem; the first target tool is called, and the input parameters are input into the first target tool, so that the first target tool obtains first data based on the input parameters; the first data is received, and reply data is generated based on the first data, so that the first target tool obtains the first data corresponding to the target alarm problem in a refined manner based on the input parameters obtained through the target alarm problem, and in the process of generating reply data, less data is processed, thereby saving resources occupied by processing data.

[0068] In some embodiments, the agent may generate a target text based on the first data; check whether the target text matches the target alarm question; and determine the target text as the reply data if the target text matches the target alarm question.

[0069] In actual execution, the first data obtained based on the first target tool can be json format data, and the intelligent agent generates the target text based on the first data.

[0070] In actual execution, the intelligent agent can check whether the target text matches the target alarm problem based on the prompt words such as alarm-related knowledge and documents pre-stored in the large model.

[0071] In some embodiments, the large model may pre-store the following prompt words: 1. Problem (which may be the target alarm problem obtained); 2. Thinking (which may refer to determining the tool used to solve the target alarm problem, such as the first target tool); 3. Action (which may refer to calling the determined tool used to solve the target alarm problem, wherein the input parameters of the action may be the input parameters obtained by analyzing the target alarm problem, such as the power plant number, date, etc.); 4. Observation (which may refer to whether the output obtained after the action is executed is correct, and the observation may be repeated multiple times); 5. Reflection (which may refer to determining whether the output obtained after the action is executed matches the target alarm problem); 6. Final answer (which may refer to determining the output as the final answer to the target alarm problem when the output obtained after the action is executed matches the target alarm problem). The large model may obtain the response data of the target alarm problem based on the above prompt words.

[0072] In some embodiments, the intelligent agent can use the target alarm question and the target text as context, and combine with pre-stored alarm-related knowledge and documents to verify whether the target text matches the target alarm question and whether the target text is correct.

[0073] In some embodiments, when the target text matches the target alarm question, the target text is determined as the response data. On the contrary, when the target text does not match the target alarm question, the agent can reflect and call tools other than the first target tool for the target alarm question to obtain response data for the target alarm question.

[0074] According to the alarm processing method applied to the energy storage system of the embodiment of the present application, a target text is generated based on the first data; and it is checked whether the target text matches the target alarm problem; when the target text matches the target alarm problem, the target text is determined as the reply data, so that the intelligent agent checks the target text to avoid the situation where the reply data generated according to the target text does not match the target alarm problem, resulting in incorrect processing of the target alarm problem.

[0075] In some embodiments, when the inspection target text does not match the target alarm problem, the second target tool can be called by the intelligent agent to generate reply data based on the third data returned by the second target tool; the second target tool is used to process the fourth data corresponding to the target alarm problem to obtain the third data; the second target tool is different from the first target tool; the second target tool includes at least one tool in a pre-established tool library.

[0076] In actual execution, the third data obtained based on the second target tool can be json format data, and the intelligent agent generates the target text based on the third data.

[0077] In actual execution, the second target tool may be a tool in a pre-established tool library other than the first target tool. The second target tool may be a single tool or a combination of multiple tools.

[0078] In some embodiments, when the inspection target text does not match the target alarm question, the second target tool can be determined by the intelligent agent, and input parameters can be obtained based on the target alarm question; the second target tool can be called, and the input parameters can be input into the second target tool so that the second target tool can obtain third data based on the input parameters; the third data can be received, and reply data can be generated based on the third data.

[0079] In some embodiments, the second target tool may obtain fourth data corresponding to the target alarm problem according to the input parameters obtained based on the target alarm problem. For example, if the target alarm problem is a CLS alarm problem, the second target tool may obtain fourth data corresponding to the CLS alarm problem according to the input parameters obtained based on the target alarm problem.

[0080] In some embodiments, the intelligent agent can use a large language model to organize the corpus based on third data to generate a reply text for the target alarm question, and perform a discriminative evaluation on the reply text, and determine whether to determine the reply text as the reply data for the target alarm question based on the evaluation results.

[0081] According to the alarm processing method applied to the energy storage system of the embodiment of the present application, when the inspection target text does not match the target alarm problem, the second target tool can be called by the intelligent agent, and reply data can be generated based on the third data returned by the second target tool, so that the intelligent agent can inspect the target text to avoid the situation where the reply data generated according to the target text does not match the target alarm problem, resulting in incorrect processing of the target alarm problem.

[0082] In some embodiments, a target corpus corresponding to the target alarm question in the target knowledge base may be obtained; based on the target corpus, it is checked whether the target text matches the target alarm question.

[0083] In actual implementation, the target corpus corresponding to the target alarm problem in the target knowledge base may include alarm protocols, alarm suggestions, alarm processing rules and other knowledge related to alarm processing.

[0084] In actual execution, the target corpus corresponding to the target alarm problem in the target knowledge base can be obtained based on the Embedding vector retrieval, and based on the target corpus, it is checked whether the target text matches the target alarm problem.

[0085] In some embodiments, based on the large language model included in the agent, it is possible to check whether the target text matches the target alarm question according to the target corpus.

[0086] According to the alarm processing method applied to the energy storage system according to the embodiment of the present application, a target text is generated based on the first data; a target corpus corresponding to the target alarm problem in the target knowledge base is obtained; based on the target corpus, whether the target text matches the target alarm problem is checked, so that the intelligent agent checks the target text to avoid the situation where the response data generated according to the target text does not match the target alarm problem, resulting in incorrect processing of the target alarm problem.

[0087] In some embodiments, the target corpus, target text and target warning question can be input into a trained large language model included in the intelligent agent to obtain the verification result output by the large language model; the verification result is used to indicate whether the target text matches the target warning question.

[0088] In some embodiments, the target text and the target alarm problem can be linked as context, and the target text and the target alarm problem can be checked based on the target corpus through the large language model to see whether they match. For example, the target alarm problem is whether there is a CLS alarm problem, and if the target text is AFCI data anomaly, the test result output by the large language model indicates that the target text does not match the target alarm problem.

[0089] According to the alarm processing method applied to the energy storage system in the embodiment of the present application, by inputting the target corpus, the target text and the target alarm question into the trained large language model included in the intelligent agent, the inspection result indicating whether the target text matches the target alarm question output by the large language model is obtained, so that the intelligent agent verifies the target text to avoid the situation where the response data generated according to the target text does not match the target alarm question, resulting in incorrect processing of the target alarm question, thereby effectively improving the accuracy and reliability of the alarm processing results, so that the intelligent agent can autonomously optimize and improve its own work performance without relying on human intervention, and further improving the intelligence and reliability of the energy storage system.

[0090] In some embodiments, the target warning problem can be processed based on a trained thought chain model included in the intelligent agent, and the first target tool can be determined based on the output of the thought chain model.

[0091] In some embodiments, the target alarm problem can be processed based on the trained thinking chain model included in the intelligent agent to obtain a processing result. The processing result can include alarm problem information, such as alarm query time, query power station ID, etc. The first target tool can be determined based on the alarm problem information in the processing result output by the thinking chain model.

[0092] In some embodiments, if the target alarm problem is "Does power station 10023 have an AFCI alarm?", the target alarm problem can be processed based on the trained thinking chain model included in the intelligent agent to obtain a processing result. The processing result may include power station 10023 and AFCI, and based on the processing result, a tool for processing AFCI alarm is called.

[0093] According to the alarm processing method applied to the energy storage system of the embodiment of the present application, a target alarm problem for the target energy storage system is obtained; the target alarm problem is processed based on a trained thinking chain model included in the intelligent agent, a first target tool is determined based on the output of the thinking chain model, and response data for the target alarm problem is generated based on the first data returned by the first target tool. This allows the user to manually select an alarm problem processing tool, and the first target tool for processing the second data corresponding to the target alarm problem and obtaining the first data can be called through the intelligent agent to generate response data for the target alarm problem, thereby significantly improving the intelligence level of alarm problem processing for the energy storage system.

[0094] In some embodiments, the target alarm problem can be processed divergently to obtain at least one divergent problem; each divergent problem is processed separately based on the trained thinking chain model included in the intelligent agent, and the first target tool is determined based on the output of the thinking chain model.

[0095] In actual execution, the intelligent agent can perform divergent processing on the target warning problem based on the target knowledge base and the preset prompt words to obtain at least one divergent problem.

[0096] In some embodiments, the pre-set prompt words may be the following: "For the following alarm issues, please think divergently from multiple angles to help fully understand and find potential solutions", "Please generate 1 to 3 different thinking directions and divergent questions to help fully analyze the problem", "These issues should be explored from multiple angles such as causes, impacts, possible related factors, user impact, system design and preventive measures".

[0097] In actual execution, after obtaining at least one divergent problem, each divergent problem can be processed based on the trained thinking chain model included in the intelligent agent, and the first target tool can be determined based on the output of the thinking chain model. In some embodiments, a tool can be determined based on the processing result of each divergent problem processed by the thinking chain model. For example, the first target tool can include one or more sub-tools, and the first sub-tool can be determined based on the first divergent problem, and the second sub-tool can be determined based on the second divergent problem.

[0098] In some embodiments, a first target application programming interface may be determined and accessed, and a first target tool may be invoked.

[0099] In order to better understand the alarm processing method for an energy storage system provided in an embodiment of the present application, further explanation is given below. It should be understood that the following discussion is merely exemplary.

[0100] The present application provides an alarm processing method for an energy storage system. The specific process can be as follows: Figure 2 As shown:

[0101] Step 1: Obtain target alarm issues for the target energy storage system.

[0102] In actual implementation, the target alarm problem for the target energy storage system can be obtained based on the user terminal. The user terminal can be an electronic device such as a smart phone, a tablet, a computer, or any theoretically feasible electronic device, which is not specifically limited in this application.

[0103] In some embodiments, the target alarm problem for the target energy storage system can be obtained by obtaining the target input of the user at the user terminal. The target input can be any theoretically feasible input method such as voice input, text input, and picture input.

[0104] In actual implementation, the target alarm problem can be an alarm problem related to the operation of each device in the target energy storage system, such as arc fault interrupter (Arc-Fault Circuit-Interrupter) alarm, residual current operated protective device (RCD) alarm, cycle life (CycleLife) alarm, gateway (Gateway) alarm, alarm trend analysis and other issues.

[0105] Step 2: Process the target warning problem based on the trained thinking chain model included in the intelligent agent, and determine the target tool corresponding to the target warning problem based on the output of the thinking chain model.

[0106] In some embodiments, the target alarm problem can be processed based on the trained thinking chain model included in the intelligent agent to obtain a processing result. The processing result can include alarm problem information, such as alarm query time, query power station ID, etc. The first target tool can be determined based on the alarm problem information in the processing result output by the thinking chain model.

[0107] In some embodiments, if the target alarm problem is "Does power station 10023 have an AFCI alarm?", the target alarm problem can be processed based on the trained thinking chain model included in the intelligent agent to obtain a processing result. The processing result may include power station 10023 and AFCI, and based on the processing result, a tool for processing AFCI alarm is called.

[0108] In some embodiments, the target alarm problem can be processed divergently to obtain at least one divergent problem; each divergent problem is processed separately based on the trained thinking chain model included in the intelligent agent, and the first target tool is determined based on the output of the thinking chain model.

[0109] In actual execution, the intelligent agent can perform divergent processing on the target warning problem based on the target knowledge base and the preset prompt words to obtain at least one divergent problem.

[0110] In some embodiments, the pre-set prompt words may be the following: "For the following alarm issues, please think divergently from multiple angles to help fully understand and find potential solutions", "Please generate 1 to 3 different thinking directions and divergent questions to help fully analyze the problem", "These issues should be explored from multiple angles such as causes, impacts, possible related factors, user impact, system design and preventive measures".

[0111] In actual execution, after obtaining at least one divergent problem, each divergent problem can be processed based on the trained thinking chain model included in the intelligent agent, and the first target tool can be determined based on the output of the thinking chain model. In some embodiments, a tool can be determined based on the processing result of each divergent problem processed by the thinking chain model. For example, the first target tool can include one or more sub-tools, and the first sub-tool can be determined based on the first divergent problem, and the second sub-tool can be determined based on the second divergent problem.

[0112] Step 3: Identify and access the target application programming interface to call the target tool.

[0113] Step 4: The target tool processes the second data corresponding to the target alarm problem to obtain the first data.

[0114] In actual execution, the first data obtained based on the first target tool may be data in json format.

[0115] Step 5: The intelligent agent generates a target text based on the first data.

[0116] Step 6: The agent checks whether the target text matches the target warning question.

[0117] In actual execution, the intelligent agent can check whether the target text matches the target alarm problem based on the prompt words such as alarm-related knowledge and documents pre-stored in the large model.

[0118] In some embodiments, the intelligent agent can use the target alarm question and the target text as context, and combine with pre-stored alarm-related knowledge and documents to verify whether the target text matches the target alarm question and whether the target text is correct.

[0119] In some embodiments, based on the large language model included in the intelligent agent, the target corpus corresponding to the target alarm question in the target knowledge base can be used; based on the target corpus, it can be checked whether the target text matches the target alarm question.

[0120] In actual implementation, the target corpus corresponding to the target alarm problem in the target knowledge base may include alarm protocols, alarm suggestions, alarm processing rules and other knowledge related to alarm processing.

[0121] In actual execution, the target corpus corresponding to the target alarm problem in the target knowledge base can be obtained based on the Embedding vector retrieval, and based on the target corpus, it is checked whether the target text matches the target alarm problem.

[0122] Step 7: When the target text matches the target alarm question, the target text is determined as the response data.

[0123] In some embodiments, when the target text matches the target alarm question, the target text is determined as the response data. On the contrary, when the target text does not match the target alarm question, the agent can reflect and call tools other than the first target tool for the target alarm question to obtain response data for the target alarm question.

[0124] In some embodiments, when the inspection target text does not match the target alarm problem, the second target tool can be called by the intelligent agent to generate reply data based on the third data returned by the second target tool; the second target tool is used to process the fourth data corresponding to the target alarm problem to obtain the third data; the second target tool is different from the first target tool; the second target tool includes at least one tool in a pre-established tool library.

[0125] In actual execution, the third data obtained based on the second target tool can be json format data, and the intelligent agent generates the target text based on the third data.

[0126] In actual execution, the second target tool may be a tool in a pre-established tool library other than the first target tool. The second target tool may be a single tool or a combination of multiple tools.

[0127] In some embodiments, when the inspection target text does not match the target alarm question, the second target tool can be determined by the intelligent agent, and input parameters can be obtained based on the target alarm question; the second target tool can be called, and the input parameters can be input into the second target tool so that the second target tool can obtain third data based on the input parameters; the third data can be received, and reply data can be generated based on the third data.

[0128] In some embodiments, the second target tool may obtain fourth data corresponding to the target alarm problem according to the input parameters obtained based on the target alarm problem. For example, if the target alarm problem is a CLS alarm problem, the second target tool may obtain fourth data corresponding to the CLS alarm problem according to the input parameters obtained based on the target alarm problem.

[0129] In some embodiments, the intelligent agent can use a large language model to organize the corpus based on third data to generate a reply text for the target alarm question, and perform a discriminative evaluation on the reply text, and determine whether to determine the reply text as the reply data for the target alarm question based on the evaluation results.

[0130] The embodiment of the present application also provides an alarm processing device applied to an energy storage system.

[0131] like Figure 3 As shown, the alarm processing device 300 applied to the energy storage system includes: an acquisition module 310 and a first generation module 320 .

[0132] An acquisition module 310 is used to acquire a target alarm problem for a target energy storage system;

[0133] The first generation module 320 is used to call the first target tool through the intelligent agent, and generate response data for the target alarm problem based on the first data returned by the first target tool; the first target tool is used to process the second data corresponding to the target alarm problem and obtain the first data; the first target tool includes at least one tool in a pre-established tool library.

[0134] According to the alarm processing device applied to the energy storage system of the present application, a target alarm problem for a target energy storage system is obtained; a first target tool is called by an intelligent body, and response data for the target alarm problem is generated based on the first data returned by the first target tool. This allows the user to not need to manually select an alarm problem processing tool. The first target tool for processing the second data corresponding to the target alarm problem and obtaining the first data can be called by the intelligent body to generate response data for the target alarm problem, thereby significantly improving the intelligence level of alarm problem processing in the energy storage system.

[0135] In some embodiments, the first generation module 320 may include:

[0136] A determination unit, configured to determine a first target tool through an intelligent agent, and obtain input parameters based on a target alarm problem;

[0137] An acquisition unit, used for calling a first target tool and inputting input parameters into the first target tool, so that the first target tool acquires first data based on the input parameters;

[0138] The generating unit is used to receive the first data and generate the reply data based on the first data.

[0139] In some embodiments, the generating unit may include:

[0140] A generating subunit, used for generating a target text based on the first data;

[0141] A verification subunit is used to verify whether the target text matches the target alarm question;

[0142] The determination subunit is used to determine the target text as the response data when the target text matches the target alarm question.

[0143] In some embodiments, the alarm processing device 300 applied to the energy storage system also includes a second generation module for calling a second target tool through an intelligent agent when the target text does not match the target alarm problem, and generating reply data based on the third data returned by the second target tool; the second target tool is used to process the fourth data corresponding to the target alarm problem to obtain the third data; the second target tool is different from the first target tool; the second target tool includes at least one tool in a pre-established tool library.

[0144] In some embodiments, the verification subunit is used to obtain a target corpus corresponding to the target alarm question in the target knowledge base; based on the target corpus, verify whether the target text matches the target alarm question.

[0145] In some embodiments, the verification subunit is specifically used to input the target corpus, target text and target warning question into the trained large language model included in the intelligent agent, and obtain the verification result output by the large language model; the verification result is used to indicate whether the target text matches the target warning question.

[0146] In some embodiments, the determination unit is used to process the target warning problem based on a trained thinking chain model included in the intelligent agent, and determine the first target tool based on the output of the thinking chain model.

[0147] In some embodiments, the determination unit is used to perform divergent processing on the target alarm problem to obtain at least one divergent problem; each divergent problem is processed separately based on the trained thinking chain model included in the intelligent agent, and the first target tool is determined based on the output of the thinking chain model.

[0148] In some embodiments, the acquisition unit is used to determine and access a first target application programming interface and call a first target tool.

[0149] The alarm processing device applied to the energy storage system in the embodiment of the present application can be an electronic device, or a component in the electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal, or it can be other devices other than the terminal. Exemplarily, the electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, a vehicle-mounted electronic device, a mobile Internet device (Mobile Internet Device, MID), an augmented reality (augmented reality, AR) / virtual reality (virtual reality, VR) device, a robot, a wearable device, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a netbook or a personal digital assistant (personal digital assistant, PDA), etc., and can also be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television, TV), a teller machine or a self-service machine, etc., which is not specifically limited in the embodiment of the present application.

[0150] The alarm processing device applied to the energy storage system in the embodiment of the present application may be a device having an operating system. The operating system may be a Microsoft (Windows) operating system, an Android (Android) operating system, an IOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0151] The alarm processing device 300 applied to the energy storage system provided in the embodiment of the present application can achieve Figure 1 to Figure 2 To avoid repetition, the various processes implemented by the method embodiment are not described here.

[0152] In some embodiments, Figure 4 As shown, an embodiment of the present application further provides an electronic device 400, including a processor 401, a memory 402, and a computer program stored in the memory 402 and executable on the processor 401. When the program is executed by the processor 401, each process of the above-mentioned alarm processing method embodiment applied to the energy storage system is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0153] It should be noted that the electronic devices in the embodiments of the present application include the mobile electronic devices and non-mobile electronic devices mentioned above.

[0154] The embodiment of the present application also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, each process of the above-mentioned alarm processing method embodiment applied to the energy storage system is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0155] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0156] An embodiment of the present application further provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned alarm processing method applied to an energy storage system.

[0157] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0158] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned alarm processing method embodiment applied to the energy storage system, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0159] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0160] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0161] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0162] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

[0163] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0164] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present application, and that the scope of the present application is defined by the claims and their equivalents.

Claims

1. An alarm processing method applied to an energy storage system, characterized in that: include: Obtain target alarm issues for target energy storage systems; Invoking a first target tool through an intelligent agent, and generating response data for the target alarm problem based on first data returned by the first target tool; The first target tool is used to process the second data corresponding to the target alarm problem to obtain the first data; the first target tool includes at least one tool in a pre-established tool library.

2. The alarm processing method applied to the energy storage system according to claim 1, characterized in that: The step of calling the first target tool through the intelligent agent and generating response data for the target alarm problem based on the first data returned by the first target tool includes: The agent performs the following operations: Determine the first target tool, and obtain input parameters based on the target alarm problem; Calling the first target tool and inputting the input parameter into the first target tool, so that the first target tool acquires the first data based on the input parameter; The first data is received, and the reply data is generated based on the first data.

3. The alarm processing method applied to the energy storage system according to claim 2 is characterized in that: The generating the reply data based on the first data includes: Based on the first data, generate a target text; Checking whether the target text matches the target alarm question; In a case where the target text matches the target alarm question, the target text is determined as the reply data.

4. The alarm processing method applied to the energy storage system according to claim 3 is characterized in that: After checking whether the target text matches the target warning question, the method further includes: When the target text does not match the target alarm problem, the second target tool is called by the intelligent agent to generate the reply data based on the third data returned by the second target tool; the second target tool is used to process the fourth data corresponding to the target alarm problem to obtain the third data; the second target tool is different from the first target tool; the second target tool includes at least one tool in a pre-established tool library.

5. The alarm processing method applied to the energy storage system according to claim 3 is characterized in that: The checking whether the target text matches the target warning question includes: Obtaining a target corpus corresponding to the target alarm problem in a target knowledge base; Based on the target corpus, it is checked whether the target text matches the target warning question.

6. The alarm processing method applied to the energy storage system according to claim 5, characterized in that: The step of checking whether the target text matches the target warning question based on the target corpus includes: The target corpus, the target text and the target warning question are input into a trained large language model included in the agent, and a verification result output by the large language model is obtained; the verification result is used to indicate whether the target text matches the target warning question.

7. The alarm processing method applied to the energy storage system according to claim 2, characterized in that: The determining the first target tool comprises: The target warning problem is processed based on the trained thinking chain model included in the intelligent agent, and the first target tool is determined based on the output of the thinking chain model.

8. The alarm processing method applied to the energy storage system according to claim 2, characterized in that: The determining the first target tool comprises: Performing divergent processing on the target alarm problem to obtain at least one divergent problem; Each of the divergent problems is processed separately based on the trained thinking chain model included in the intelligent agent, and the first target tool is determined based on the output of the thinking chain model.

9. The alarm processing method applied to an energy storage system according to any one of claims 2 to 8, characterized in that: The calling of the first target tool comprises: A first target application programming interface is determined and accessed, and the first target tool is invoked.

10. An alarm processing device applied to an energy storage system, characterized in that: include: An acquisition module, used to acquire target alarm problems for a target energy storage system; A first generating module, configured to call a first target tool through an intelligent agent, and generate response data for the target alarm problem based on first data returned by the first target tool; The first target tool is used to process the second data corresponding to the target alarm problem to obtain the first data; the first target tool includes at least one tool in a pre-established tool library.

11. 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 processor executes the program, the alarm processing method applied to the energy storage system as described in any one of claims 1 to 9 is implemented.

12. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the alarm processing method applied to the energy storage system as described in any one of claims 1 to 9 is implemented.

13. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the alarm processing method applied to an energy storage system as described in any one of claims 1 to 9 is implemented.

Images