Energy storage problem reply method and device, equipment and storage medium

Abstract

The embodiment of the invention provides an energy storage problem reply method and device, equipment and a storage medium. The method comprises the steps of obtaining an energy storage problem proposed by a user; performing part-of-speech analysis on the energy storage problem to obtain a keyword of the energy storage problem; inputting the keyword of the energy storage problem into a pre-constructed knowledge graph for query to obtain a query result; wherein the knowledge graph is constructed according to energy storage knowledge data of at least one knowledge dimension, and the knowledge dimensions comprise a policy and regulation dimension, a domestic and overseas standard dimension, an operation and maintenance experience dimension, a technical characteristic dimension, a risk characteristic dimension, a fault characteristic dimension and a decommissioning characteristic dimension; and according to the query result, generating a reply for the energy storage problem.

Description

Technical Field

[0001] The present disclosure relates to the technical field of energy storage, and particularly to a method, device, equipment, and storage medium for replying to energy storage problems. Background Art

[0002] With the rapid development of energy storage technology in the power industry, the construction demand for energy storage power stations is increasing day by day. However, the construction of energy storage power stations is a complex project involving many professional knowledge fields: on the one hand, the relevant policies and regulations for energy storage are constantly updated, and there are significant differences in domestic and foreign standards, making it difficult for builders to comprehensively master them; on the other hand, there are a variety of energy storage technologies, each with its own safety characteristics and risk points. In addition, the installed capacity of new energy storage has shown an explosive growth trend, with the newly installed capacity in the current year exceeding the sum of all previous years for two consecutive years. As more and more lithium-ion battery energy storage power stations are put into operation, the health management of energy storage batteries has become a key link in the operation of power stations, and their health status directly affects the performance and lifespan of energy storage power stations.

[0003] Currently, energy storage knowledge is usually obtained by manually consulting materials and consulting experts, but these methods are inefficient and prone to information omission. Summary of the Invention

[0004] Embodiments of the present disclosure provide a method, device, equipment, and storage medium for replying to energy storage problems, so as to solve the problems of low efficiency and easy information omission caused by obtaining energy storage knowledge by manually consulting materials and consulting experts in the related art.

[0005] In a first aspect, embodiments of the present disclosure provide a method for replying to energy storage problems, the method including: Obtaining an energy storage problem proposed by a user; Performing part-of-speech analysis on the energy storage problem to obtain keywords of the energy storage problem; Inputting the keywords of the energy storage problem into a pre-constructed knowledge graph for query to obtain a query result; wherein, the knowledge graph is constructed based on energy storage knowledge data of at least one knowledge dimension, and the knowledge dimension includes: policy and regulation dimension, domestic and foreign standard dimension, operation and maintenance experience dimension, technical characteristic dimension, risk characteristic dimension, fault characteristic dimension, and decommissioning characteristic dimension; Generating a reply to the energy storage problem according to the query result.

[0006] In a second aspect, embodiments of the present disclosure provide a device for replying to energy storage problems, the device including: An obtaining module, configured to obtain an energy storage problem proposed by a user; An analysis module, configured to perform part-of-speech analysis on the energy storage problem to obtain keywords of the energy storage problem; A query module, configured to input keywords of the energy storage problem into a pre-constructed knowledge graph for querying to obtain a query result; wherein, the knowledge graph is constructed based on energy storage knowledge data of at least one knowledge dimension, and the knowledge dimension includes: policy and regulation dimension, domestic and foreign standard dimension, operation and maintenance experience dimension, technical characteristic dimension, risk characteristic dimension, fault characteristic dimension, and decommissioning characteristic dimension; A generation module, configured to generate a reply to the energy storage problem according to the query result.

[0007] In a third aspect, an embodiment of the present disclosure provides an energy storage problem reply device, including: a processor; and a memory configured to store computer-executable instructions, and the computer-executable instructions, when executed, cause the processor to implement the steps of the method described in the first aspect above.

[0008] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, which is used to store computer-executable instructions, and the computer-executable instructions, when executed by a processor, implement the steps of the method described in the first aspect above.

[0009] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, which includes a computer program, and the computer program, when executed by a processor, implements the steps of the method described in the first aspect above.

[0010] The above at least one technical solution provided by the embodiment of the present invention can achieve the following technical effects: In the embodiment of the present invention, the energy storage problem proposed by the user can be obtained first, and the part-of-speech analysis of the energy storage problem can be performed to obtain the keywords of the energy storage problem. Then, the keywords of the energy storage problem are input into a pre-constructed knowledge graph for querying to obtain a query result, wherein the knowledge graph is constructed based on energy storage knowledge data of at least one knowledge dimension, and the knowledge dimension specifically includes: policy and regulation dimension, domestic and foreign standard dimension, operation and maintenance experience dimension, technical characteristic dimension, risk characteristic dimension, fault characteristic dimension, and decommissioning characteristic dimension. Finally, a reply to the energy storage problem can be generated according to the query result.

[0011] In the embodiments of the present invention, keywords can be obtained by performing part-of-speech analysis on the energy storage problems raised by users, and then input into a knowledge graph constructed based on one or more knowledge dimensions, such as the policy and regulation dimension, the domestic and foreign standard dimension, etc., for query. In this way, relevant knowledge can be obtained quickly and comprehensively, effectively avoiding the problem of information omission that is prone to occur in the manual method. Thus, a reply to the energy storage problem can be generated accurately, providing users with more efficient, accurate, and comprehensive energy storage knowledge services, strongly promoting the intelligent process of knowledge acquisition and application in the energy storage field, and accurately answering questions about the construction, operation, and maintenance of lithium-ion battery energy storage power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in one or more embodiments of the present disclosure or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the present disclosure. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Figure 1 It is a schematic flowchart of a method for replying to energy storage problems provided by an embodiment of the present invention. Figure 2 It is a schematic diagram of the module composition of an energy storage problem reply device 200 provided by an embodiment of the present invention. Figure 3 It is a schematic diagram of the hardware structure of an energy storage problem reply device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0013] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of the present disclosure, in order to make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0014] The following will detail the technical solutions provided by each embodiment of the present invention in conjunction with the drawings.

[0015] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of a method for replying to energy storage problems provided by an embodiment of the present invention. As Figure 1 shown, the method includes the following steps: Step 102: Obtain the energy storage problem raised by the user.

[0016] Step 104: Perform part-of-speech analysis on the energy storage problem to obtain the keywords of the energy storage problem.

[0017] Step 106: Input the keywords of the energy storage problem into the pre-constructed knowledge graph for query to obtain the query result; wherein, the knowledge graph is constructed based on the energy storage knowledge data of at least one knowledge dimension, and the knowledge dimensions include: policy and regulation dimension, domestic and foreign standard dimension, operation and maintenance experience dimension, technical characteristic dimension, risk characteristic dimension, fault characteristic dimension, and decommissioning characteristic dimension.

[0018] Step 108: Generate a reply to the energy storage problem according to the query result.

[0019] In the embodiment of the present invention, the energy storage problem proposed by the user can be obtained.

[0020] In an embodiment of the present invention, in order to facilitate the user to obtain the energy storage knowledge they want to know, a question input box can be designed. The user can directly input the question they want to ask in the question input box. For example, questions such as what policy and regulation requirements need to be met for the construction of an energy storage power station and what are the reasons for the thermal runaway of lithium-ion batteries. After the user inputs the energy storage problem they want to ask, the energy storage problem input by the user can be obtained from the question input box.

[0021] After obtaining the energy storage problem proposed by the user, part-of-speech analysis can be performed on the energy storage problem to obtain the keywords of the energy storage problem.

[0022] In an embodiment of the present invention, after obtaining an energy storage problem, part-of-speech analysis can be performed on the energy storage problem. First, a sentence can be split into individual words using a word segmentation tool. For example, for the problem "What are the safety standards for a lithium-ion battery energy storage power station", it can be segmented into: ["lithium-ion battery", "energy storage power station", "of", "safety standards", "are", "what"]. Then, part-of-speech tagging is performed on the words obtained after word segmentation. For example, "lithium-ion battery", "energy storage power station", and "safety standards" can be tagged as nouns; "of" can be tagged as a particle; "are" can be tagged as a verb; and "what" can be tagged as a pronoun. After tagging the part of speech of each word, words with parts of speech such as conjunctions, particles, adverbs, prepositions, and pronouns can be removed, and only words with noun and verb parts of speech are retained. For example, the above particle "of" and pronoun "what" can be removed, retaining the nouns "lithium-ion battery", "energy storage power station", "safety standards", and the verb "are". Then, it can be determined whether the verb includes a general verb according to a predefined general verb library, and the general verb is removed. The general verb library is predefined and includes general and relatively meaningless verbs such as "are", "is", "in", etc. According to the general verb library, it can be recognized that the verb "are" in the above problem is a general verb, and at this time, this verb can be removed. After removing the general verb, the remaining nouns and non-general verbs can be determined as keywords.

[0023] In one example, the energy storage problem proposed by the user can be "How to maintain a lithium-ion battery energy storage power station". When performing part-of-speech analysis on this problem, the problem can be first segmented to obtain ["how", "maintain", "lithium-ion battery", "energy storage power station"], and then, part-of-speech tagging is performed on each word. "How" is tagged as an adverb, "maintain" is tagged as a verb, "lithium-ion battery" is tagged as a noun, and "energy storage power station" is tagged as a noun. Words other than nouns and non-general verbs among these words are removed, obtaining the nouns "lithium-ion battery", "energy storage power station", and the non-general verb "maintain", and these three words are determined as keywords. Therefore, the keywords for the energy storage problem "How to maintain a lithium-ion battery energy storage power station" can be "maintain", "lithium-ion battery", and "energy storage power station".

[0024] In an embodiment of the present invention, after obtaining the keywords of the energy storage problem, the keywords of the energy storage problem can be input into a pre-constructed knowledge graph for query to obtain a query result.

[0025] In one embodiment of the present invention, a knowledge graph can be pre-constructed. When constructing the knowledge graph, energy storage knowledge data of each knowledge dimension can be collected separately, where the knowledge dimension can include at least one of the following: policy and regulation dimension, domestic and foreign standards dimension, operation and maintenance experience dimension, technical characteristics dimension, risk characteristics dimension, fault characteristics dimension, and decommissioning characteristics dimension.

[0026] In one example, when collecting energy storage knowledge data of the policy and regulation dimension, it can be obtained from the official websites of government departments. For example, it can be obtained from the official websites of relevant ministries and commissions, etc., to obtain the policies and regulations of energy storage. These websites usually have a policy and regulation section, which publishes various energy storage-related policy documents, such as industrial development plans, subsidy policies, access standards, etc. For example, relevant policies and regulations can be obtained by searching for the keyword "energy storage" in the "Policy and Regulation" section of the official website of the government energy management department.

[0027] When collecting energy storage knowledge data of the domestic and foreign standards dimension, it can be retrieved and downloaded from the official websites of relevant standardization institutions. These institutions' websites usually have a standard query entry, and by inputting keywords such as "energy storage power station", "lithium-ion battery energy storage", "electrochemical energy storage system", etc., standard documents related to energy storage can be screened out, and energy storage knowledge data such as the design, installation, and acceptance standards involved in energy storage power stations can be collected.

[0028] When collecting energy storage knowledge data of the operation and maintenance experience dimension, it can be obtained from materials such as energy storage power station operation and maintenance manuals and maintenance records. These materials are usually recorded and saved by the operation and management department of the energy storage power station or relevant technical personnel. The operation and maintenance manual details information such as the operation process, maintenance key points, and inspection cycle of the equipment; the maintenance record contains the historical situation of equipment failures, maintenance measures, replaced parts, etc.

[0029] When collecting energy storage knowledge data of the technical characteristics dimension, it can be obtained from papers, books, and promotional materials published by research institutes, university laboratories, energy storage enterprises, etc. Research institutes and universities usually publish energy storage technology research papers in the academic resource library of their official websites or relevant academic databases, and can be retrieved and screened by setting keywords such as "lithium-ion battery energy storage technology" and "battery performance parameters"; while energy storage enterprises usually introduce in detail the energy storage technologies and technical features of their products on their official websites and product brochures, and the technical features of the enterprise, such as the main parameters of lithium-ion batteries, such as energy density, charge and discharge efficiency, cycle life, etc., can be obtained through their official websites and product brochures.

[0030] When collecting energy storage knowledge data in the dimension of risk characteristics, risk point data can be collected by obtaining analysis reports and accident investigation reports on past energy storage accidents. These reports can be obtained from channels such as professional industry research institutions, accident case databases of relevant government departments, and safety management materials of energy storage enterprises. For example, some large energy storage enterprises will conduct detailed investigations and analyses on typical accidents they have experienced or in the industry and form corresponding reports; government energy management departments will also issue official investigation reports after major accidents occur.

[0031] When collecting energy storage knowledge data in the dimension of fault characteristics, fault characteristic data can be collected by obtaining fault records in past energy storage accident events. These fault records can be sourced from the operation and maintenance databases of energy storage power stations, accident investigation reports, after-sales repair records of equipment manufacturers, etc. For example, energy storage power stations will record in detail the time of each fault occurrence, equipment information, fault phenomena, etc. during daily operation and maintenance, and these records constitute an important data source; when major accidents occur, relevant departments or enterprises will conduct in-depth investigations and form reports, which also contain rich fault details; equipment manufacturers will also accumulate repair cases when dealing with after-sales problems.

[0032] When collecting energy storage knowledge data in the dimension of decommissioning characteristics, it can be obtained from the decommissioning guidance documents for energy storage equipment issued by the government and the decommissioning process specifications announced by enterprises. Government departments will issue relevant policies, regulations and technical guidelines, which can be retrieved and downloaded through their official websites. On the enterprise side, large energy storage enterprises may announce their own energy storage equipment decommissioning processes and standards on their official websites.

[0033] As can be seen from the above, in the process of collecting knowledge in the energy storage field, the data in the policy and regulation dimension involves various energy storage-related policies, regulations and ordinances issued by the state and local governments, aiming to regulate and guide the development of the energy storage industry from a macro level; the domestic and international standards dimension focuses on the energy storage-related standards formulated by international organizations, countries and industries, such as technical standards, safety standards, etc., to ensure the standardized design and operation of energy storage systems; the operation and maintenance experience dimension covers various experiences accumulated in the daily operation and maintenance practice of energy storage power stations, including equipment inspection, maintenance cycle, fault handling, etc.; the technical characteristics dimension focuses on the knowledge of the principles, performance parameters, technical development trends, etc. of energy storage technologies; the risk characteristics dimension focuses on various risks that may be faced during the operation of energy storage systems, such as safety risks, market risks, etc.; the fault characteristics dimension mainly focuses on the phenomena, diagnostic methods and repair measures when energy storage systems fail; the retirement characteristics dimension focuses on the processing procedures, environmental protection requirements, etc. at the retirement stage of energy storage equipment. Since all links in the energy storage field are closely connected, the data obtained from these dimensions inevitably overlap and duplicate. For example, policies and regulations may put forward requirements for energy storage technical standards and operation and maintenance safety, thus overlapping with the domestic and international standards dimension; some content in technical characteristics may be related to the risk characteristics and fault characteristics dimensions, because technical principles and performance often determine potential risks and fault types; the fault handling experience in operation and maintenance experience may also be similar to the data in the fault characteristics dimension. Although duplicate data may be obtained, the data from each dimension can be viewed from different perspectives and complement each other, jointly ensuring the comprehensiveness of the knowledge coverage of the energy storage field, covering all aspects from policy guidance, standard specification to technical principles, operation and maintenance management, risk prevention and control, fault handling and retirement disposal, which helps to provide more efficient, accurate and comprehensive energy storage knowledge services in the future.

[0034] In one embodiment of the present invention, after collecting the energy storage knowledge data of each knowledge dimension, in order to facilitate the subsequent construction of a knowledge graph, the collected energy storage knowledge data can be subjected to structured processing. When performing structured processing, the energy storage knowledge data for different knowledge dimensions can be processed into different structured forms.

[0035] In one example, for the energy storage knowledge data in the dimension of policies and regulations, it can be structured in the form of "promulgation time - scope of application - key points". Among them, the key points can be determined by keyword extraction algorithms, such as the TF-IDF algorithm (Term Frequency-Inverse Document Frequency). For example, the structured energy storage knowledge data in the dimension of policies and regulations can be: x month x day, 20xx - nationwide - for energy storage power station construction projects, it is required that the proportion of energy storage supporting new energy power generation is not less than xx, and the energy storage duration is not less than x hours; for the energy storage knowledge data in the dimension of domestic and foreign standards, it can be structured in the form of "standard number - standard name - scope of application - key technical indicators". For example, the structured energy storage knowledge data in the dimension of domestic and foreign standards can be: IEC 62619:2017 - (Secondary batteries and battery packs containing alkaline or other non-acidic electrolytes - Safety requirements for secondary lithium batteries and battery packs for industrial applications) - Lithium-ion battery energy storage systems applied in the global industrial field - Strict regulations are made on safety performance indicators such as overcharge protection, over-discharge protection, short-circuit protection, and thermal runaway protection of the battery pack. For example, when the battery pack is overcharged to 130% of the rated capacity, it should be able to automatically cut off the charging circuit; for the energy storage knowledge data in the dimension of operation and maintenance experience, it can be structured in the form of "equipment name - inspection cycle - high-failure components - maintenance key points - composition of operation and maintenance costs - personnel skill requirements". For example, the structured energy storage knowledge data in the dimension of operation and maintenance experience can be: energy storage battery pack - once every 7 days - battery connection strips - Regularly check the connection tightness of the battery connection strips to prevent loosening from causing an increase in contact resistance and heating problems - Labor costs account for 40%, mainly used for tasks such as operation and maintenance personnel inspections, fault handling, and equipment maintenance - Material costs account for 30%, including battery replacement, parts repair, and consumable procurement; equipment depreciation costs account for 20%; other costs (such as electricity and water fees, site rental, etc.) account for 10% - Operation and maintenance personnel need to have knowledge of power electronics technology and electrochemistry, be familiar with the operation and maintenance of the battery management system, master electrical equipment maintenance skills, be able to proficiently use relevant detection instruments and meters, and have certain fault diagnosis and emergency handling capabilities; for the energy storage knowledge data in the dimension of technical characteristics, it can be structured in the form of "energy storage technology type - enterprise name - energy density (Wh / kg) - charge and discharge efficiency (%) - cycle life (times) - technical advantages - application scenarios". For example, the structured energy storage knowledge data in the dimension of technical characteristics can be: ternary lithium-ion battery technology - high-tech enterprise X - 200 - 93 - 4000 - High energy density, capable of providing high energy storage in a smaller volume and weight, with good power performance - Commonly used in occasions with high energy density requirements, such as power batteries for new energy vehicles, distributed energy storage systems, etc., to meet the long-range or high-power output requirements of equipment;For the energy storage knowledge data in the risk characteristic dimension, it can be structured in the form of "risk point name - risk level - cause of occurrence - possible consequences - preventive measures". For example, the structured energy storage knowledge data in the risk characteristic dimension can be: electrolyte leakage risk - medium - due to reasons such as poor battery packaging, external impact, and aging during long-term use - the electrolyte is corrosive, and leakage may damage the battery and surrounding equipment, affect battery performance, and may cause short circuits and fires in severe cases, increasing the failure probability of the energy storage system - strengthen the quality control of the battery packaging process and improve the strength of the battery housing; set up anti-leakage protection devices in the energy storage system design, such as electrolyte leakage detection sensors, to detect and control leakage in a timely manner; regularly conduct visual inspections and performance tests on the battery, and replace aging batteries in a timely manner; for the energy storage knowledge data in the fault characteristic dimension, it can be structured in the form of "fault phenomenon - diagnostic process - repair measures - classification of fault causes". For example, the structured energy storage knowledge data in the fault characteristic dimension can be: abnormal output voltage of the converter - 1. Measure the input voltage and output voltage waveforms of the converter with an oscilloscope, and compare with the normal waveforms to determine whether there is distortion; 2. Check whether the drive signals of the power module are normal, and whether there are missing or abnormal pulses; 3. Check the program of the control chip of the converter and read the parameters to see if there are incorrect settings or faults - 1. Replace damaged filter components such as capacitors and inductors according to the voltage waveform distortion; 2. Repair or replace the drive circuit of the power module; 3. Update the program of the control chip or replace the faulty chip, and reset the correct parameters - faults in the power module (such as component damage, abnormal drive circuit), problems in the control circuit (control chip failure, program error, improper parameter setting), external interference (such as excessive grid voltage fluctuations affecting the converter output); for the energy storage knowledge data in the retirement characteristic dimension, it can be structured in the form of "retirement identification standard - disassembly and recycling process - environmental protection key points - reuse methods". For example, the structured energy storage knowledge data in the retirement characteristic dimension can be: when the battery capacity decays to less than 80% of the rated capacity, or the battery internal resistance increases to more than 150% of the initial value, and it still cannot meet the performance requirements of the energy storage system after multiple repairs, the battery is determined to be retired - 1. First, discharge the retired battery to ensure its safety; 2. Use professional disassembly equipment to remove the battery pack housing and connecting components; 3. Classify and screen the battery cells, and detect their remaining capacity and performance; 4. Recombine or cascade use the reusable battery cells, and harmlessly treat the non-reusable battery cells - The disassembly process should be carried out in a well-ventilated workshop with environmental protection facilities to prevent the volatilization of electrolytes and dust pollution; properly recycle and treat the waste battery materials (such as heavy metals, electrolytes, etc.) generated during disassembly to avoid pollution to the soil and water sources;Adopt environment-friendly disassembly equipment and processes to reduce energy consumption and waste emissions. Screen and reorganize the battery monomers with better performance in retired batteries for applications with lower requirements for energy density and cycle life, such as low-speed electric vehicles and backup power sources for distributed energy storage. For the battery casing and some structural components, after cleaning and repair, they can be used to manufacture new battery packs or other industrial products. The valuable metal materials (such as lithium, cobalt, nickel, etc.) in the batteries can be extracted through professional recycling processes and then reused in battery production or other metal processing industries.

[0036] After collecting the energy storage knowledge data of each knowledge dimension and structuring the collected energy storage knowledge data, entity recognition can be performed on the collected energy storage knowledge data according to the named entity recognition technology to determine the entities corresponding to the collected energy storage knowledge data.

[0037] In the embodiments of the present invention, the existing named entity recognition technology can be directly adopted to perform entity recognition on the collected energy storage knowledge data to determine the corresponding entities, so as to realize the construction of the subsequent knowledge graph.

[0038] After determining the entities, the key entities can be determined from the determined entities according to the occurrence frequencies of the determined entities in the collected energy storage knowledge data. Then, according to the predefined relationship types, the determined key entities can be used as nodes to connect the nodes, and according to the predefined attribute types, attribute annotation can be performed on the determined key entities, and a knowledge graph can be constructed according to the existing knowledge graph construction tools, the relationships between the determined key entities, and the attributes of the determined key entities.

[0039] In the embodiments of the present invention, the relationship types between the entities of the energy storage knowledge data can be predefined in advance. In one example, the relationship types can be predefined as: causal relationship, application relationship, constraint relationship, composition relationship, dependence relationship, influence relationship.

[0040] For example, the relationship type between the entity "battery thermal runaway" and the entity "energy storage power station fire" can be a causal relationship, the relationship type between the entity "lithium-ion battery" and the entity "energy storage power station" can be an application relationship, the relationship type between the entity "energy storage power station construction standard" and the entity "energy storage power station construction" can be a constraint relationship, the relationship type between the entity "battery management system" and the entity "energy storage system" can be a composition relationship, the relationship type between the entity "energy storage project operation" and the entity "policy subsidy" can be a dependence relationship, and the relationship type between the entity "environmental temperature" and the entity "battery performance" can be an "influence relationship".

[0041] After determining the key entities, the text including the key entities can be analyzed to find out whether there is a statement structure that conforms to the predefined relationship rules. For example, when a statement like "According to the Safety Code for Energy Storage Power Stations, the construction of an energy storage power station should meet..." appears in the text, it can be determined as a constraint relationship. In addition, semantic analysis techniques in natural language processing can be used to understand the semantic associations between key entities in the text to assist in relationship annotation. For example, for the text data "In a high-temperature environment, the internal chemical reactions of the battery intensify, resulting in a decline in battery performance", it can be determined through semantic analysis that there is a causal relationship between the entity "high-temperature environment" and the entity "decline in battery performance".

[0042] When annotating the attributes of key entities, the key entities can be annotated according to the predefined attribute types. Among them, the predefined attribute types are the attributes that key entities usually have determined based on professional knowledge and industry standards in the energy storage field. For example, for the key entity "energy storage power station", according to the knowledge of the power engineering and energy storage industries, its attributes may include "installed capacity", "energy storage technology type", "geographical location", "commissioning time", "service life", etc. After defining the attribute types, the key entities can be directly annotated according to the predefined attribute types for each key entity.

[0043] After determining the key entities, and annotating the relationships between the key entities and the attributes of the key entities, a knowledge graph corresponding to the energy storage knowledge data can be constructed according to the existing knowledge graph construction tools.

[0044] In an example, an existing knowledge graph construction tool, such as Neo4j, can be used to construct a knowledge graph corresponding to the energy storage knowledge data. The entity "lithium-ion battery" can be used as a node, and its attributes can include energy density, charge-discharge efficiency, etc., and the application relationship can be used as an edge to connect the "lithium-ion battery" node and the "energy storage power station" node.

[0045] After pre-constructing the knowledge graph through the above process, the keywords of the energy storage questions proposed by the user can be input into the pre-constructed knowledge graph for query to obtain the query results.

[0046] In an example, the energy storage question proposed by the user can be: What are the safety risks and countermeasures of a lithium-ion battery energy storage power station in a high-temperature environment? After performing part-of-speech analysis on this energy storage question, the keywords of this energy storage question can be obtained as: "lithium-ion battery", "energy storage power station", "high-temperature environment", "safety risk", and "countermeasures". After inputting these keywords into the constructed knowledge graph for query, the following retrieval results can be obtained: From the technical characteristics dimension, performance change information such as the reduction of energy density, the decline of charge and discharge efficiency, and the change of the structural stability of battery materials in a high-temperature environment will be presented, as well as the impact of these changes on the overall operation performance of the energy storage power station, such as affecting the power storage capacity and charge and discharge speed of the energy storage power station; from the risk characteristics dimension, it will feedback an increased risk of thermal runaway, which may cause serious consequences such as fire and explosion, as well as safety risks such as shortened battery life and increased electrical fault risks. At the same time, it may provide the occurrence probability data of relevant risks in a high-temperature environment to quantify the risk level; from the operation and maintenance experience dimension, specific operation and maintenance countermeasures will be given, such as strengthening temperature monitoring, such as increasing the number of temperature sensors and monitoring frequency; optimizing the heat dissipation system, including regular maintenance of heat dissipation equipment or upgrading heat dissipation technology; adjusting the charge and discharge strategy, for example, reducing the charge and discharge rate during high-temperature periods, etc., to ensure the stable operation of the energy storage power station in a high-temperature environment; from the fault characteristics dimension, it will feedback the method of diagnosing faults caused by high temperature by monitoring parameters such as battery voltage, current, and internal resistance and using thermal imaging detection, and will also explain the repair measures for situations such as damage to battery monomers or modules and battery management system failures caused by high temperature, such as replacing damaged components and debugging and repairing the battery management system; from the policy and regulation dimension, it will provide that the policy requires the operator of the energy storage power station to formulate a high-temperature emergency plan, clarify the emergency response process and responsibility division, and stipulate the indicators and frequencies of safety monitoring in a high-temperature environment, etc., to ensure that the operation complies with the regulatory requirements; from the domestic and international standards dimension, it will show the regulations of international standards such as [xxxxxx] on the performance test methods and safety indicators of lithium-ion batteries at high temperature, as well as the detailed specifications of domestic standards such as [yyyyyy] on the design, construction, operation, etc. of energy storage power stations in a high-temperature environment, providing a standard basis for the construction and operation of the power station; from the retirement characteristics dimension, it will feedback that high temperature accelerates the aging of lithium-ion batteries, resulting in a reduced remaining available value at the time of retirement, and that special processes such as cooling pretreatment need to be adopted during the disassembly and recycling process to ensure safety and environmental protection.

[0047] After obtaining the query result, a response to the energy storage problem can be generated based on the query result. In the embodiment of the present invention, since the obtained query result may involve multiple knowledge dimensions, the query result can be integrated, and the integrated content can be used as the response to the user's energy storage problem.

[0048] For example, in the previous example, retrieval results involving 7 knowledge dimensions such as the dimension of policies and regulations, the dimension of domestic and foreign standards, the dimension of operation and maintenance experience, and the dimension of technical characteristics can be obtained. For the convenience of presenting to the user, the retrieval results can be integrated, and the integrated content can be used as a reply. The integrated content can be: In a high-temperature environment, for a lithium-ion battery energy storage power station, in terms of technical characteristics, its energy density will decrease, the charge and discharge efficiency will decline, and the stability of battery materials will be affected; this brings various safety risks, such as a significant increase in the risk of thermal runaway, which may cause fires and explosions, a shortened battery life, and an increased probability of electrical failures; in terms of operation and maintenance, temperature monitoring needs to be strengthened, the heat dissipation system needs to be optimized, and the charge and discharge strategy needs to be adjusted; for fault diagnosis, the battery parameters and thermal imaging detection can be monitored, and the damaged components can be replaced or the battery management system can be repaired; policy and regulation requirements state that a high-temperature emergency plan should be formulated, and the monitoring indicators and frequencies should be clarified; domestic and foreign standards also stipulate the performance testing, power station construction and operation, etc. under high temperature; in the decommissioning link, high temperature reduces the remaining value of the battery, and special processes are required for disassembly and recycling to ensure safety and environmental protection. This integrated content can be used as a reply and presented.

[0049] In an embodiment of the present invention, in addition to generating a reply to the energy storage problem, a follow-up question corresponding to the energy storage problem can also be generated.

[0050] In an embodiment of the present invention, the energy storage problem can be input into a pre-trained knowledge state level model to obtain the original knowledge state level corresponding to the energy storage problem; wherein, the original knowledge state level can be used to identify the understanding degree of the user who poses the energy storage problem about energy storage knowledge. Then, according to the original knowledge state level and the energy storage problem, a follow-up question corresponding to the energy storage problem can be generated to guide the user to understand more energy storage knowledge.

[0051] In an embodiment of the present invention, a knowledge state level model can be pre-trained. First, a model that performs well in natural language processing can be constructed, such as the BERT model. Then, sample data can be collected to train the model. The sample data is energy storage problems with annotated knowledge state levels. For example, the energy storage problem "What is energy storage technology? What are the common types?" annotated as the "completely novice stage", the energy storage problem "I know that thermal runaway is a risk of energy storage. Then how is it generally prevented?" annotated as the "preliminary understanding stage", the energy storage problem "When the thermal runaway risk occurs, what are the differences in the response time and fire extinguishing effect of different types of fire extinguishing systems?" annotated as the "in-depth mastery stage", etc. After the training is completed, after inputting a question into the trained knowledge state level model, the model can output the knowledge state level corresponding to the question, such as the "completely novice stage", the "in-depth mastery stage", etc.

[0052] In one embodiment of the present invention, after obtaining the original state level corresponding to the energy storage problem through the knowledge state level model, follow-up questions corresponding to the energy storage problem can be generated based on the original knowledge state level and the energy storage problem. Specifically, the target knowledge dimension corresponding to the energy storage problem can be determined first according to the keywords of the energy storage problem.

[0053] Among them, when determining the keywords of the energy storage problem, the nouns and non-general verbs in the energy storage problem can be determined as keywords as shown in the above embodiment. When determining the target knowledge dimension corresponding to the keywords, these keywords can be matched with the predefined keyword libraries of each knowledge dimension. That is, in the embodiments of the present invention, keyword libraries can be predefined for each knowledge dimension. For example, the keyword library predefined for the policy and regulation dimension may include: subsidy policy, access policy, support policy, supervision policy, tax policy, subsidy standard, access conditions, qualification requirements, approval process, project filing, etc. After successful matching, the corresponding knowledge dimension is determined as the target knowledge dimension corresponding to the energy storage problem.

[0054] After determining the target knowledge dimension, the target transfer influence factor corresponding to the target knowledge dimension can be determined according to the preset transfer influence factors corresponding to each knowledge dimension.

[0055] In the embodiments of the present invention, the preset transfer influence factors corresponding to different knowledge dimensions are different. For example, the preset transfer influence factors corresponding to the policy and regulation dimension can be historical transfer frequency, policy popularity, and regional policy attention; the preset transfer influence factors corresponding to the domestic and international standards dimension can be historical transfer frequency, standard update frequency, and standard application scope; the preset transfer influence factors corresponding to the operation and maintenance experience dimension can be historical transfer frequency, operation and maintenance cost change, and operation and maintenance personnel demand change; the preset transfer influence factors corresponding to the technical characteristics dimension are historical transfer frequency and technical popularity; the preset transfer influence factors corresponding to the risk characteristics dimension are historical transfer frequency and accident occurrence frequency; the preset transfer influence factors corresponding to the fault characteristics dimension are historical transfer frequency and fault occurrence frequency; the preset transfer influence factors corresponding to the decommissioning characteristics dimension are historical transfer frequency and the tightness of environmental protection policies.

[0056] After determining the target transfer influence factor corresponding to the target knowledge dimension, the target transfer probability of the energy storage problem from the original knowledge state level to the candidate knowledge state level can be determined according to the target transfer influence factor corresponding to the target knowledge dimension through a preset formula; among them, the candidate knowledge state level is higher than the original knowledge state level. For example, when the original knowledge state level is the complete novice stage, the candidate knowledge state level can be the preliminary understanding stage.

[0057] In the embodiments of the present invention, for different knowledge dimensions, different formulas can be set in combination with the transfer influencing factors corresponding to the knowledge dimensions.

[0058] In one example, for the policy and regulation dimension, from the original knowledge state level to the candidate knowledge state level the transfer probability can be , and the calculation formula can be:

[0059] Among them, is the transfer frequency from to in past Q&A. Assuming that in 100 Q&A analyzed, this transfer occurred 30 times, then ; is the policy heat, and the value range is 0 - 1. If there is an important policy introduced currently and the heat is high (which can be determined according to the corresponding search volume and view volume), = 0.5; is the transfer weight directly related to this policy, which can be set to 0.8, indicating that this policy has a greater promoting effect on the transfer; a is the regional policy attention. If the region has a high attention to the energy storage policy (which can be determined according to the mention times and mention frequency of the energy storage policy), = 0.4; is the weight of the transfer caused by the regional relevant policy, which can be set to 0.6.

[0060] For example, , = 0.5, , a = 0.4, , then

[0061] When is greater than a certain threshold, such as 1.5, it can be considered that the user is more likely to transfer to stage.

[0062] In another example, for the domestic and foreign standards dimension, from the original state level to the candidate state level the transfer probability is , and the calculation formula is as follows:

[0063] Among them, is the transfer frequency from to The transfer frequency. Suppose that in 40 Q&A sessions analyzed, such a transfer occurred 12 times, then ; is the standard update frequency, with a value range of 0 - 1. If there have been many updates to the energy storage-related standards recently, , is the influence coefficient of the standard update degree on the transfer, which can be set to 0.7; is the application scope of the industry standard. If a certain type of standard (such as battery safety standard) is widely applied in actual projects and in many types of projects in the current industry, , is the weight of the transfer caused by the application scope, which can be set to 0.8.

[0064] For example, , , , , , then

[0065] When is greater than a certain threshold, such as 2, it can be considered that the user is likely to transfer to status.

[0066] In another example, for the dimension of operation and maintenance experience, the transfer probability from the original knowledge state level to the candidate knowledge state level can be , and the calculation formula can be:

[0067] Where, is the transfer frequency from to in past Q&A sessions. Suppose that in 60 Q&A sessions analyzed, such a transfer occurred 20 times, then ; is the weight of the operation and maintenance cost change. If the operation and maintenance cost of an energy storage power station in a certain area has increased significantly due to equipment aging (which can be determined according to the factory time of the equipment), ; is the influence coefficient of the operation and maintenance cost change on the transfer, which can be set to 0.6; is the weight of the personnel skill demand heat. If the industry has a great demand for operation and maintenance personnel with the skill of battery system fault diagnosis (which can be determined by the change in the number of corresponding job recruitments), ; is the weight of the transfer caused by the skill demand heat, which can be set to 0.7.

[0068] For example, , , , , , then

[0069] When is greater than a certain threshold, such as 2.8, it can be considered that the user is more likely to transfer to phase.

[0070] In another example, for the technical feature dimension, the transition probability from the original knowledge state level to the candidate knowledge state level can be , and the calculation formula can be:

[0071] where is the transition frequency from to in past Q&A. Assuming that in 80 Q&A analyzed, this transition occurred 20 times, then ; is the technical popularity. If a certain technology is popular in the industry (which can be determined according to the number and frequency of times the technology is mentioned), ; is the influence coefficient of this popular technology on the transition, which can be set to 0.7.

[0072] For example, , , , then

[0073] When is greater than a certain threshold, such as 1.8, it can be considered that the user is more likely to transfer to phase.

[0074] In another example, for the risk feature dimension, the transition probability from the original knowledge state level to the candidate knowledge state level can be , and the calculation formula can be:

[0075] where is the transition frequency from to in past Q&A. Assuming that in 50 Q&A analyzed, this transition occurred 15 times, then ; is the historical accident occurrence frequency, with a value range of 0 - 1. If the thermal runaway risk frequently appears in recent energy storage accidents, ; is the impact coefficient corresponding to the occurrence frequency of this risk, which can be set to 0.7.

[0076] For example, 、 、 , then

[0077] When is greater than a certain threshold, such as 2.4, it can be considered that the user is likely to transfer to stage.

[0078] In another example, for the fault characteristic dimension, the transition probability from the original knowledge state level to the candidate knowledge state level can be , and the calculation formula can be:

[0079] Among them, is the transition frequency from to in previous questions and answers. Assuming that in 70 analyzed questions and answers, this transition occurred 25 times, then ; is the occurrence frequency of common faults. If battery short - circuit faults frequently occur in past cases, ; is the impact coefficient corresponding to the occurrence frequency of this fault, which can be set to 0.7.

[0080] For example, 、 、 , then

[0081] When is greater than a certain threshold, such as 2.9, it can be considered that the user is likely to transfer to stage.

[0082] In another example, for the decommissioning characteristic dimension, the transition probability from the original knowledge state level to the candidate knowledge state level can be , and the calculation formula can be:

[0083] Among them, is the transfer frequency from to . Assuming that in 40 Q&A sessions analyzed, this transfer occurred 10 times, then ; is the degree of tightness or looseness of environmental protection policies. If the environmental protection requirements for the retirement of energy storage equipment are significantly increased, ; is the influence coefficient of changes in environmental protection policies on the transfer, which can be set to 0.9.

[0084] For example, and , then

[0085] When is greater than a certain threshold, such as 2.5, it can be considered that the user is more likely to transfer to stage.

[0086] After determining the target transfer probability, the target knowledge state level can be determined from the original knowledge state level and the candidate knowledge state level according to the target transfer probability. Specifically, when the target transfer probability is higher than the preset threshold, it can be considered that the user is more likely to transfer to the next stage, and then the candidate knowledge state level can be determined as the target knowledge state level; when the target transfer probability is not higher than the preset threshold, it can be considered that the user is more likely to stay in the current stage, and then the original knowledge state level can be determined as the target knowledge state level.

[0087] After determining the target knowledge state level, at least one question can be selected from the preset question bank corresponding to the target knowledge state level as the follow-up question corresponding to the energy storage question proposed by the user, so as to guide the user to understand more in-depth energy storage knowledge. Of course, if there are multiple questions in the preset question bank corresponding to the target knowledge state level, questions related to the energy storage question proposed by the user, such as questions with the same keywords, can be selected as the follow-up questions.

[0088] As can be seen from the above, the embodiments of the present invention can accurately locate the target knowledge dimension based on the keywords of the questions proposed by the user, and combine the preset transfer influencing factors unique to each dimension, such as the historical transfer frequency, policy heat, and regional policy attention in the policy and regulation dimension, etc., to reasonably calculate the target transfer probability from the original knowledge state level to the higher candidate knowledge state level, and then determine the target knowledge state level, and select follow-up questions from the corresponding preset question bank. This method can deeply understand the user's knowledge level and demand direction, generate follow-up questions that fit the actual situation of the user, guide the user to gradually explore energy storage knowledge in depth, improve the pertinence and effectiveness of information acquisition, and help the user build a more complete energy storage knowledge system.

[0089] In an embodiment of the present invention, after generating a response to the energy storage problem proposed by the user, it is also possible to determine the confidence score corresponding to the response, and in the case of a low confidence score, that is, in the case where the response may be incorrect, generate a corresponding prompt to prompt the user that there is a risk of incorrectness in the response.

[0090] Among them, when determining the confidence score corresponding to the response, it can be determined according to the number of adjacent nodes of the keyword in the above-mentioned constructed knowledge graph and the number of citations.

[0091] In an example, the corresponding adjacent level value can be set according to the number of adjacent nodes of the keyword in the knowledge graph. Among them, the more the number of adjacent nodes, the closer the corresponding keyword is related to other nodes in the knowledge graph, the higher the credibility, and the higher the adjacent level value; correspondingly, the corresponding citation level value can be determined according to the number of times the keyword in the response is cited. Among them, the more the number of citations, the more important and recognized it is, and the higher the credibility. At this time, a higher citation level value can be set. After setting the adjacent level value and the citation level value, the sum of these two values can be determined as the confidence score corresponding to the response.

[0092] In the embodiment of the present invention, the confidence score corresponding to the response can be determined by the number of adjacent nodes of the keyword in the knowledge graph and the number of citations, and the user is prompted when the score is low. Since the number of adjacent nodes can reflect the degree of closeness of the knowledge involved in the response to other knowledge in the knowledge graph, the more the number, the higher the integration degree of the knowledge in the entire knowledge system, the more stable the foundation, and the higher the credibility; and the number of citations reflects the degree of recognition of the knowledge in the knowledge graph. The more frequently it is cited, the more important and reliable it is in the field of energy storage knowledge. Therefore, determining the confidence score based on these two factors can effectively evaluate the accuracy of the response, prompt the user in time when the confidence is low, avoid misleading the user due to receiving incorrect information, ensure the quality and reliability of the information obtained by the user, and improve the practicability and scientificity of the solution.

[0093] In another embodiment of the present invention, after generating a reply to the energy storage question, user feedback from the user who raised the energy storage question regarding the reply can also be obtained. When obtaining user feedback, multiple forms can be adopted, for example, directly asking the user in the form of direct questions, setting a feedback window, actively asking through customer service, etc., and the embodiment of the present invention does not limit this. After obtaining user feedback, the knowledge graph can be optimized according to the user feedback. For example, if the user feedback points out that the information of a certain entity is incorrect, such as "the energy density data of lithium-ion batteries is inaccurate", the energy density attribute of the entity "lithium-ion battery" in the knowledge graph is corrected; if the user feedback mentions that a certain entity lacks key attributes, such as "I hope to know the relevant information about the floor area of ​​the energy storage power station", then the attribute "floor area" is added to the "energy storage power station" entity in the knowledge graph; if a new concept is mentioned many times in the user feedback, such as "new solid electrolyte energy storage technology", and it is not included in the existing knowledge graph, it can be added to the knowledge graph as a new entity, and a relationship with other related entities is established.

[0094] In the embodiments of the present invention, the knowledge graph can be optimized through user feedback, errors in the knowledge graph can be corrected in a timely manner, missing information can be supplemented, responses can be made more accurate and reliable, user trust can be enhanced, users can feel a sense of participation, satisfaction and loyalty can be improved, the pace of development of the energy storage industry can be kept up to date, knowledge can be updated in a timely manner to adapt to changes, the practicality and advancement of the system can be ensured, the knowledge system can be continuously improved, and strong support can be provided for knowledge research and application in the field of energy storage.

[0095] In an embodiment of the present invention, the energy storage question raised by the user may be first obtained, and the part-of-speech analysis of the energy storage question may be performed to obtain the keywords of the energy storage question. Then, the keywords of the energy storage question may be input into a pre-constructed knowledge graph for query to obtain the query result, wherein the knowledge graph is constructed based on energy storage knowledge data of at least one knowledge dimension, and the knowledge dimension specifically includes: policy and regulation dimension, domestic and international standard dimension, operation and maintenance experience dimension, technical characteristic dimension, risk characteristic dimension, fault characteristic dimension and decommissioning characteristic dimension. Finally, a response to the energy storage question may be generated based on the query result.

[0096] In an embodiment of the present invention, keywords can be obtained by performing part-of-speech analysis on energy storage questions raised by users, and input into a knowledge graph constructed based on one or more knowledge dimensions, such as policy and regulation dimensions, domestic and international standard dimensions, etc. for query, so that relevant knowledge can be quickly and comprehensively acquired, and the problem of information omission that is easy to occur in manual methods can be effectively avoided, thereby accurately generating responses to energy storage questions, providing users with more efficient, accurate and comprehensive energy storage knowledge services, and effectively promoting the intelligent process of knowledge acquisition and application in the field of energy storage, and accurately responding to questions about the construction, operation and maintenance of lithium-ion battery energy storage power stations.

[0097] Corresponding to the above method for answering energy storage problems, an embodiment of the present invention further provides an energy storage problem answering device. Figure 2 It is a schematic diagram of the module composition of the energy storage problem answering device 200 provided by an embodiment of the present invention. As Figure 2 shown, the energy storage problem answering device 200 includes: An acquisition module 201, configured to acquire an energy storage problem proposed by a user; An analysis module 202, configured to perform part-of-speech analysis on the energy storage problem to obtain keywords of the energy storage problem; A query module 203, configured to input the keywords of the energy storage problem into a pre-constructed knowledge graph for querying to obtain a query result; wherein, the knowledge graph is constructed based on energy storage knowledge data of at least one knowledge dimension, and the knowledge dimension includes: policy and regulation dimension, domestic and foreign standard dimension, operation and maintenance experience dimension, technical characteristic dimension, risk characteristic dimension, fault characteristic dimension, and decommissioning characteristic dimension; A generation module 204, configured to generate a reply to the energy storage problem according to the query result.

[0098] Optionally, the method further includes ( Figure 2 not shown in ): A model module 205, configured to input the energy storage problem into a pre-trained knowledge state level model after acquiring the energy storage problem proposed by the user to obtain an original knowledge state level corresponding to the energy storage problem; wherein, the original knowledge state level is used to identify the understanding degree of the user who proposed the energy storage problem about energy storage knowledge;

[0099] Optionally, the questioning module 206 is configured to: Determine a target knowledge dimension corresponding to the energy storage problem according to the keywords of the energy storage problem; Determine the target transfer influencing factor corresponding to the target knowledge dimension according to the preset transfer influencing factors corresponding to each knowledge dimension; wherein, the preset transfer influencing factors corresponding to the policy and regulation dimension are historical transfer frequency, policy popularity, and regional policy attention; the preset transfer influencing factors corresponding to the domestic and foreign standards dimension are historical transfer frequency, standard update frequency, and standard application scope; the preset transfer influencing factors corresponding to the operation and maintenance experience dimension are historical transfer frequency, operation and maintenance cost change, and operation and maintenance personnel demand change; the preset transfer influencing factors corresponding to the technical characteristics dimension are historical transfer frequency and technical popularity; the preset transfer influencing factors corresponding to the risk characteristics dimension are historical transfer frequency and accident occurrence frequency; the preset transfer influencing factors corresponding to the fault characteristics dimension are historical transfer frequency and fault occurrence frequency; the preset transfer influencing factors corresponding to the decommissioning characteristics dimension are historical transfer frequency and the tightness of environmental protection policies; Determine the target transfer probability of the energy storage problem from the original knowledge state level to the candidate knowledge state level according to the target transfer influencing factor corresponding to the target knowledge dimension; wherein, the candidate knowledge state level is higher than the original knowledge state level; Determine the target knowledge state level from the original knowledge state level and the candidate knowledge state level according to the target transfer probability; Select at least one question from the preset question bank corresponding to the target knowledge state level as the follow-up question corresponding to the energy storage problem.

[0100] Optionally, the method further includes ( Figure 2 not shown in the figure): The acquisition module 207 is used to respectively acquire the energy storage knowledge data of each knowledge dimension before inputting the keywords of the energy storage problem into the pre-constructed knowledge graph for query; The recognition module 208 is used to perform entity recognition on the acquired energy storage knowledge data according to the named entity recognition technology to determine the entity corresponding to the acquired energy storage knowledge data; The first determination module 209 is used to determine the key entity from the determined entities according to the occurrence frequency of each determined entity in the acquired energy storage knowledge data; The processing module 210 is used to connect the nodes with the determined key entities as nodes according to the predefined relationship type, and perform attribute annotation on the determined key entities according to the predefined attribute type; The construction module 211 is used to construct a knowledge graph according to the knowledge graph construction tool, the determined key entities, the relationships between the determined key entities, and the attributes of the determined key entities.

[0101] Optionally, the method further includes (Figure 2 not shown in ( A second determination module 212, configured to determine a confidence score corresponding to the reply after generating the reply to the energy storage problem; the confidence score is determined according to the number of adjacent nodes of the keywords in the reply in the knowledge graph and the number of citations; A prompt module 213, configured to generate a prompt when the confidence score corresponding to the reply is lower than a preset score threshold; the prompt is used to remind that there is a risk of correctness in the reply.

[0102] Optionally, the method further includes ( Figure 2 not shown in ( A feedback module 214, configured to obtain user feedback on the reply from the user who proposed the energy storage problem after generating the reply to the energy storage problem; An optimization module 215, configured to optimize the knowledge graph according to the user feedback.

[0103] In an embodiment of the present invention, the energy storage problem proposed by the user can be obtained first, and part-of-speech analysis is performed on the energy storage problem to obtain the keywords of the energy storage problem. Then, the keywords of the energy storage problem are input into a pre-constructed knowledge graph for query, and a query result is obtained. The knowledge graph is constructed according to energy storage knowledge data of at least one knowledge dimension, and the knowledge dimension specifically includes: policy and regulation dimension, domestic and foreign standard dimension, operation and maintenance experience dimension, technical characteristic dimension, risk characteristic dimension, fault characteristic dimension, and decommissioning characteristic dimension. Finally, a reply to the energy storage problem can be generated according to the query result.

[0104] In an embodiment of the present invention, keywords can be obtained by performing part-of-speech analysis on the energy storage problem proposed by the user, and input into a knowledge graph constructed based on one or more knowledge dimensions, such as policy and regulation dimension, domestic and foreign standard dimension, etc., to quickly and comprehensively obtain relevant knowledge, effectively avoiding the problem of information omission that easily occurs in the manual method, so as to accurately generate a reply to the energy storage problem, provide a more efficient, accurate and comprehensive energy storage knowledge service for users, strongly promote the intelligent process of knowledge acquisition and application in the energy storage field, and accurately respond to questions about the construction and operation and maintenance of lithium-ion battery energy storage power stations.

[0105] Corresponding to the above energy storage problem reply method, an embodiment of the present invention further provides an energy storage problem reply device, Figure 3 which is a schematic hardware structure diagram of an energy storage problem reply device provided in an embodiment of the present invention.

[0106] The energy storage problem reply device may be a terminal device or a server provided in the above embodiment for replying to energy storage problems, etc.

[0107] The energy storage problem response device may vary greatly due to different configurations or performances, and may include one or more processors 301 and a memory 302. One or more storage application programs or data may be stored in the memory 302. Among them, the memory 302 may be short-term storage or persistent storage. The application programs stored in the memory 302 may include one or more modules (not shown in the figure), and each module may include a series of computer-executable instructions in the energy storage problem response device. Further, the processor 301 may be set to communicate with the memory 302 and execute a series of computer-executable instructions in the memory 302 on the energy storage problem response device. The energy storage problem response device may also include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more input / output interfaces 305, and one or more keyboards 306.

[0108] Specifically, in this embodiment, the energy storage problem response device includes a memory and one or more programs. One or more of the programs are stored in the memory, and one or more of the programs may include one or more modules. Each module may include a series of computer-executable instructions in the energy storage problem response device and is configured to be executed by one or more processors in the above embodiment.

[0109] In an embodiment of the present invention, the energy storage problem proposed by the user can be obtained first, and the part-of-speech analysis of the energy storage problem can be performed to obtain the keywords of the energy storage problem. Then, the keywords of the energy storage problem are input into a pre-constructed knowledge graph for querying to obtain a query result. Among them, the knowledge graph is constructed based on the energy storage knowledge data of at least one knowledge dimension, and the knowledge dimension specifically includes: policy and regulation dimension, domestic and foreign standard dimension, operation and maintenance experience dimension, technical characteristic dimension, risk characteristic dimension, fault characteristic dimension, and decommissioning characteristic dimension. Finally, a response to the energy storage problem can be generated according to the query result.

[0110] In an embodiment of the present invention, the keywords can be obtained by performing part-of-speech analysis on the energy storage problem proposed by the user and input into a knowledge graph constructed based on one or more knowledge dimensions, such as the policy and regulation dimension, domestic and foreign standard dimension, etc., to quickly and comprehensively obtain relevant knowledge, effectively avoiding the problem of information omission that is prone to occur in the manual method. Thus, a response to the energy storage problem can be accurately generated, providing users with more efficient, accurate, and comprehensive energy storage knowledge services, strongly promoting the intelligent process of knowledge acquisition and application in the energy storage field, and accurately answering questions about the construction and operation and maintenance of lithium-ion battery energy storage power stations.

[0111] Another embodiment of the present disclosure also provides a computer-readable storage medium for storing computer-executable instructions, which, when executed by a processor, implement the above process.

[0112] The storage medium in the embodiments of the present disclosure can implement each process of the above-mentioned energy storage problem response method embodiment and achieve the same effects and functions, which will not be repeated here.

[0113] Another embodiment of the present disclosure also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above process.

[0114] The computer program product in the embodiments of the present disclosure can implement each process of the above-mentioned energy storage problem response method embodiment and achieve the same effects and functions, which will not be repeated here.

[0115] In each embodiment of the present disclosure, the computer-readable storage medium includes a read-only memory (ROM), a random access memory (RAM), a magnetic disk, an optical disk, or the like.

[0116] In the 1990s, it was obvious to distinguish whether an improvement in a technology was an improvement in hardware (e.g., improvement in circuit structures such as diodes, transistors, switches, etc.) or an improvement in software (improvement in method processes). However, with the development of technology, many improvements in method processes today can be regarded as direct improvements in hardware circuit structures. Almost all designers obtain the corresponding hardware circuit structure by programming the improved method process into the hardware circuit. Therefore, it cannot be said that an improvement in a method process cannot be implemented with a hardware entity module. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logical function is determined by the user programming the device. The designer can program by himself to "integrate" a digital system on a piece of PLD, without having to ask the chip manufacturer to design and manufacture a dedicated integrated circuit chip. Moreover, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software, which is similar to the software compiler used in program development writing, and the original code before compilation also has to be written in a specific programming language, which is called Hardware Description Language (HDL), and there is not only one kind of HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. Currently, the most commonly used ones are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also be aware that as long as the method process is slightly logically programmed with the above-mentioned several hardware description languages and programmed into the integrated circuit, it is easy to obtain the hardware circuit that implements the logical method process.

[0117] The controller can be implemented in any suitable manner. For example, the controller can take the form of, for example, a microprocessor or a processor and a computer-readable medium storing computer-readable program code (such as software or firmware) executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of the controller include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in the form of pure computer-readable program code, it is entirely possible to logically program the method steps to enable the controller to be implemented in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. to achieve the same function. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be regarded as the structures within the hardware component. Or even, the devices for implementing various functions can be regarded as either software modules for implementing the method or structures within the hardware component.

[0118] The systems, devices, modules, or units illustrated in the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0119] For the convenience of description, the above devices are described by dividing them into various units according to their functions. Of course, when implementing the embodiments of the present disclosure, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0120] Those skilled in the art should understand that one or more embodiments of the present disclosure can be provided as a method, a system, or a computer program product. Therefore, one or more embodiments of the present disclosure can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, one or more embodiments of the present disclosure can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program code.

[0121] The present disclosure is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each flow and / or block in the flowchart and / or block diagram, and combinations of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing device produce a means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0122] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including an instruction means that implements the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0124] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover a non-exclusive inclusion, such that a process, method, commodity or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "including a..." does not exclude the presence of additional identical elements in the process, method, commodity or device including the element.

[0125] One or more embodiments of the present disclosure may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. One or more embodiments of the present disclosure may also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including storage devices.

[0126] Each embodiment in the present disclosure is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the key points of each embodiment are the differences from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the related content.

[0127] The above description is only for the embodiments of the present disclosure and is not intended to limit the present disclosure. For those skilled in the art, various changes and modifications can be made to the present disclosure. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present disclosure shall be included within the scope of the claims of the present disclosure.

Claims

1. A method for answering energy storage questions, characterized in that: The method comprises: Obtain energy storage questions raised by users; Performing part-of-speech analysis on the energy storage problem to obtain keywords of the energy storage problem; Input the keywords of the energy storage problem into a pre-built knowledge graph for querying to obtain query results; wherein the knowledge graph is constructed based on energy storage knowledge data of at least one knowledge dimension, and the knowledge dimension includes: policy and regulation dimension, domestic and international standard dimension, operation and maintenance experience dimension, technical characteristic dimension, risk characteristic dimension, fault characteristic dimension and decommissioning characteristic dimension; A response to the energy storage question is generated based on the query result.

2. The method according to claim 1, characterized in that After obtaining the energy storage question raised by the user, the method further includes: Inputting the energy storage problem into a pre-trained knowledge state level model to obtain an original knowledge state level corresponding to the energy storage problem; wherein the original knowledge state level is used to identify the degree of understanding of energy storage knowledge by the user who raised the energy storage problem; According to the original knowledge state level and the energy storage problem, a follow-up question corresponding to the energy storage problem is generated.

3. The method according to claim 2, characterized in that Generating follow-up questions corresponding to the energy storage problem according to the original knowledge state level and the energy storage problem includes: Determine the target knowledge dimension corresponding to the energy storage problem according to the keywords of the energy storage problem; According to the preset transfer influencing factors corresponding to each knowledge dimension, the target transfer influencing factors corresponding to the target knowledge dimension are determined; wherein, the preset transfer influencing factors corresponding to the policy and regulation dimension are the historical transfer frequency, policy popularity, and regional policy attention; the preset transfer influencing factors corresponding to the domestic and international standards dimension are the historical transfer frequency, standard update frequency, and standard application scope; the preset transfer influencing factors corresponding to the operation and maintenance experience dimension are the historical transfer frequency, operation and maintenance cost changes, and operation and maintenance personnel demand changes; the preset transfer influencing factors corresponding to the technical characteristics dimension are the historical transfer frequency and technology popularity; the preset transfer influencing factors corresponding to the risk characteristics dimension are the historical transfer frequency and accident frequency; the preset transfer influencing factors corresponding to the fault characteristics dimension are the historical transfer frequency and fault frequency; the preset transfer influencing factors corresponding to the retirement characteristics dimension are the historical transfer frequency and the degree of tightness of environmental protection policies; Determining the target transfer probability of the energy storage problem from the original knowledge state level to the candidate knowledge state level according to the target transfer influencing factor corresponding to the target knowledge dimension; wherein the candidate knowledge state level is higher than the original knowledge state level; Determining a target knowledge state level from the original knowledge state level and the candidate knowledge state levels according to the target transition probability; At least one question is selected from a preset question library corresponding to the target knowledge state level as a follow-up question corresponding to the energy storage problem.

4. The method according to claim 1, characterized in that: Before inputting the keywords of the energy storage problem into a pre-built knowledge graph for querying, the method further includes: Collecting energy storage knowledge data of each of the knowledge dimensions respectively; According to the named entity recognition technology, entity recognition is performed on the collected energy storage knowledge data to determine the entity corresponding to the collected energy storage knowledge data; Determine key entities from the determined entities according to the occurrence frequency of each determined entity in the collected energy storage knowledge data; According to the predefined relationship type, the determined key entities are used as nodes, the nodes are connected, and, according to the predefined attribute type, the determined key entities are attribute-labeled; A knowledge graph is constructed based on a knowledge graph construction tool, determined key entities, relationships between determined key entities, and attributes of determined key entities.

5. The method according to claim 1, characterized in that After generating a response to the energy storage question, the method further includes: Determine a confidence score corresponding to the reply; the confidence score is determined based on the number of adjacent nodes of the keywords in the reply in the knowledge graph and the number of times they are cited; When the confidence score corresponding to the reply is lower than a preset score threshold, a prompt is generated; the prompt is used to remind that there is a risk of correctness of the reply.

6. The method according to claim 1, characterized in that After generating a response to the energy storage question, the method further includes: Obtain user feedback on the reply from the user who raised the energy storage question; The knowledge graph is optimized according to the user feedback.

7. An energy storage problem answering device, characterized in that: The device comprises: An acquisition module is used to obtain energy storage questions raised by users; An analysis module, used for performing part-of-speech analysis on the energy storage problem to obtain keywords of the energy storage problem; A query module, used for inputting the keywords of the energy storage problem into a pre-built knowledge graph for querying and obtaining query results; wherein the knowledge graph is constructed based on energy storage knowledge data of at least one knowledge dimension, and the knowledge dimension includes: policy and regulation dimension, domestic and foreign standard dimension, operation and maintenance experience dimension, technical characteristic dimension, risk characteristic dimension, fault characteristic dimension and decommissioning characteristic dimension; A generation module is used to generate a response to the energy storage problem based on the query result.

8. An energy storage problem answering device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the method according to any one of claims 1 to 6 when executed by the processor.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store computer-executable instructions, and when the computer-executable instructions are executed by a processor, the steps of the method described in any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images