Power grid platform audio data stream link model based on data relation guidance

Abstract

The invention discloses a power grid platform audio data stream link model based on data relation guidance, and the data model constructs an initialized data structure based on basic data of risk early warning of a power grid 95598 platform. The data model initialization comprises the following data modules: power grid 95598 audio data, power grid operation state data and power grid risk event data; and other subsequent extensible data modules. The method is a datamation information technology development subitem in a power grid situation monitoring and risk early warning application general project, is constructed for sensitive word-based audio digital intelligent links and risk early warning under a power grid 95598 platform audio data stream framework, and has a brand new data structure; and the method is well coordinated with a digital operation process based on sensitive words under an audio data stream framework of a power grid 95598 platform.

Description

Technical Field

[0001] The present invention relates to the field of smart grid technology, and in particular to a data relationship-guided grid audio data stream link model and its application that cooperates well with a digital operation process based on sensitive words under the audio data stream framework of a grid 95598 platform. Background Art

[0002] With the continuous improvement of power supply service process control methods, State Grid Hebei Electric Power has established service quality control methods for online channels such as 95598, and provided a three-level control mechanism for early warning for cities, districts, counties, and power supply stations. However, with the continuous improvement of service quality management requirements, the accuracy of data processing and analysis dimensions of traditional technical conditions is increasingly unable to meet current technical needs.

[0003] The State Grid Hebei Service Situation Monitoring and Risk Warning Application Project was carried out under this background; the present invention is a technical achievement of one of the data information technology development sub-items.

[0004] From a technical perspective, the technical development project of the power user service situation monitoring and risk warning application project is also very urgent. On the one hand, with the continuous improvement of service quality management requirements, the data processing and analysis accuracy of the traditional technical conditions adopted by the original platform are increasingly unable to meet the current technical needs; for example, under the existing power user management and control platform technology, the construction of the data structure is mainly focused on the simple storage and manual processing of the power grid 95598 audio data, power grid operation status data and power grid risk event data, lacking in-depth mining and utilization of the complex relationships between data modules, and at the same time, there are also major deficiencies in the link processing of data relationships, especially the lack of effective guidance and processing strategies for multi-directional links between data module attributes.

[0005] On the other hand, the existing power user management and control platform is still relatively conservative in terms of data dimension, and has weak compatibility with the increasingly mature machine intelligence algorithms. For example, when processing the audio data stream and its associated risk event data under the power grid 95598 platform, there is a lack of a dedicated data structure to build intelligent data links and risk warning applications. There are also great limitations in processing the direct representation of data module attributes, multi-directional links of data relationships, and their application construction. These problems in this regard have also exacerbated the severity of the first problem mentioned above. Summary of the invention

[0006] The technical problem to be solved by the present invention is to overcome the various deficiencies of the prior art and provide a specialized new data structure, which is constructed for audio digital intelligent linking and risk warning based on sensitive words under the audio data stream framework of the power grid 95598 platform. It has a brand-new data structure and works well with the digital operation process based on sensitive words under the audio data stream framework of the power grid 95598 platform.

[0007] In order to solve the above technical problems, the technical solutions adopted by the present invention are as follows.

[0008] A power grid platform audio data stream link model guided by data relations is used for audio digital intelligent linking and risk warning based on sensitive words under the audio data stream framework of the power grid 95598 platform. The data model uses the basic data of the power grid 95598 platform risk warning to construct an initialization data structure. Specifically, the data model initialization includes the following data modules: power grid 95598 audio data, power grid operation status data, power grid risk event data; and other subsequent expandable data modules.

[0009] As a preferred technical solution of the present invention, the data model, in terms of data structure, uses the data form of attribute sets to characterize each of the above-mentioned initialization data modules; wherein: The attribute set of the power grid 95598 audio data in the data module is set to include: 95598 call audio storage, audio record ID, audio timestamp, audio file path, transcription text, sensitive word extraction tags associated with power grid risk events, sensitive words, other numerical data, and other unstructured data; The attribute set of the power grid operation status data of the data module includes: electrical data of the power grid operation in the concerned area, signal data of the power grid operation in the concerned area, power outage and fault data of the power grid in the concerned area, non-fault operation and maintenance data of the power grid in the concerned area; other power grid operation status data; the attribute set of this data module can be constructed by reference or import and data merging based on the existing database of the power system; The attribute set of the power grid risk event data in the data module is constructed as a two-dimensional array structure, that is, the data is first expanded along one dimension according to the type of power grid risk event, that is, different types of power grid risk events are classified and arranged, and on this basis, each data position on the first data dimension is expanded in the second dimension according to the risk level of the power grid risk event, thereby obtaining a two-dimensional data attribute set of the power grid risk event data module.

[0010] As a preferred technical solution of the present invention, in terms of data relations and data links, the data model essentially links different data according to the real data relations within each data module such as the power grid 95598 audio data, the power grid operation status data, the power grid risk event data, etc.; and in form, it constructs a data link structure that has both a center and is compatible with multi-directional links.

[0011] As a preferred technical solution of the present invention, specifically: the data link is centered on a certain selected initial data module, that is, one is selected as the central data module from each data module such as the power grid 95598 audio data, the power grid operation status data, the power grid risk event data, etc., and then based on the data relationship between the attribute sets of other data modules and the initial data module as the center, and combined with the pre-set or manually specified data standards for measuring the data relationship between different data module attribute sets, it serves as a data guide for other data modules to link the data relationship to the central data module; in data implementation, these data wizards are also constructed as a data set and or a side data parameter library, and the selected central data module is linked to other data modules according to the guidance of this data parameter library; in this way, the data in the parameter library has a one-to-one correspondence with the data link channels between different data modules, and the data relationship and data link of different data modules can be modified accordingly by modifying, copying, and deleting the parameter library.

[0012] As a preferred technical solution of the present invention, more specifically, in guiding the data link relationship, the following simplified and most practical data path is adopted: the data module power grid 95598 audio data is used as the initial central data module, and then a certain attribute value in the attribute set of the initial data module (that is, a single subset of the attribute set, rather than other types of subsets such as a subset of the attribute set containing more than two attribute values) is used as a data parameter reference value for guiding the data link relationship. Specifically, the data link relationship is guided by a single subset of the attribute set of the central data module power grid 95598 audio data, namely, the "sensitive word" attribute; wherein the data sources of sensitive words include: based on the power grid 95598 call audio and its transcribed text, and based on sensitive word extraction tags associated with power grid risk events.

[0013] As a preferred technical solution of the present invention, the sensitive words have two data dimension correlations: based on the power grid 95598 call audio and its transcribed text; based on sensitive word extraction tags associated with power grid risk events; such as identifying sensitive words from the power grid 95598 call audio and its transcribed text according to the sensitive word extraction tags associated with power grid risk events.

[0014] The application of the power grid platform audio data stream link model based on data relationship guidance includes: Application A: Based on the audio data stream of the 95598 platform of the power grid, early warning, statistics, auxiliary construction or other applications of single power outages, frequent power outage risks, power grid failures, power billing disputes, power user complaints, power complaint escalation risks, group power risk events, power risk event location positioning and other power grid risk events; Application B, based on the data output of the 95598 voice sensitive word system, assists in building a user portrait for power complaints; Application C, guided by the demands of power users, replaces sensitive words of power grid risk events with sensitive words of user demands, thereby importing the user demand data set as other subsequent expandable data modules into the bottom layer of the data structure, and using equivalent data processes and data algorithms to build an equivalent user demand data model for early warning, statistics, auxiliary construction or other applications of user demand events.

[0015] The beneficial effects of adopting the above technical solution are: This application is a technical achievement of a digital information technology development sub-item in the overall project of power grid situation monitoring and risk warning application development, mainly for 95598 audio data and risk keyword-based data model construction and risk warning application, to build a data model (including data structure and its internal data relationship) that works well with the digital operation process based on sensitive words under the audio data stream framework of the power grid 95598 platform. The development of this data system has pioneering technical innovation value and good practical technical advantages; First, a technical development main line of the present invention is to propose a new data structure, which is initialized to include power grid 95598 audio data, power grid operation status data, power grid risk event data, and other subsequent extended data modules; then, we characterize the data module itself through the attribute set of the data module, and at the same time directly associate the attributes between the data modules. The specific elements of the attribute set are based on the relevance of the power grid data itself, and have a multi-directional linked data relationship construction; At the same time, the second technical main line adopts a guidance strategy based on sensitive words in data relationship linking, and realizes the refined processing of data relationships through combination and further ordered processing and parameterized storage; as a whole, it provides a new data architecture idea and technical solution for the construction of audio digital intelligent link data model under the power grid 95598 platform.

[0016] The specific implementation modes below describe in detail the detailed beneficial effects of the technical solution of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a diagram of the data structure of a specific development case on the main line of the data module.

[0018] Figure 2 This is a schematic diagram of a specific data structure on the data link main line. DETAILED DESCRIPTION

[0019] The following examples describe the present invention in detail. In the description of the following examples, specific details such as specific system structures, technologies, etc. are proposed for illustration rather than limitation, so as to provide a thorough understanding of the present application embodiments. However, it should be clear to those skilled in the art that the present application can also be implemented in other embodiments without these specific details. In other cases, the detailed description of well-known systems, devices, circuits and methods is omitted to prevent unnecessary details from hindering the description of the present application.

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

[0021] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0022] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0023] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0024] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0025] Example 1: Power Grid Situation Monitoring and Risk Warning Application Project At present, the power grid system has preliminarily completed the construction of the information management and control platform for power users, and based on this platform, has preliminarily realized the real-time acquisition of call data of provincial 95598 power users, classification and automatic identification of power user demands, assessment of power user service risk index, monitoring of service management indicators, analysis of marketing event demands, application of service risk warning, monitoring of service management quality indicators and other dimensions of power user service quality management. Relying on this platform, through the access to the marketing business application system, business expansion registration system and 95598 call real-time data (less than 30 minutes), the real-time analysis of the calls and demands of power users of each local company has been preliminarily realized, and the analysis results have been promoted and applied to companies in cities, districts and counties across the province. At the same time, by constructing a risk theme model, the risk problems contained in the dispatched opinion work order have been preliminarily realized, and the three levels of red, orange and yellow have been set by constructing a three-level risk warning method.

[0026] The State Grid Hebei service situation monitoring and risk warning application project was carried out under this background.

[0027] The technical necessity of this project construction is as follows: 1). Carry out customer service analysis indicator monitoring, reduce service risks, and improve the level of service quality management. With the continuous improvement of service quality, the number of company complaints has dropped significantly in recent years, but there are still many potential risks in customer demands. It is necessary to control the checkpoints in advance, rely on the 95598 demand big data application and frequency curve, do a good job in monitoring and handling various demand changes, locate key areas, personnel, and problems of service hazards, and carry out layered supervision and tiered control. While doing a good job in demand analysis and monitoring, it is also necessary to expand the breadth and depth of business type analysis, realize the terminalization of indicator analysis business, the terminalization of problem distribution, and the terminalization of demand occurrence time, analyze and monitor various risk issues, and fully support the service management work of management units at all levels. 2). Improve the management of omni-channel customer demands from the "customer perspective" and optimize the means of problem tracing and control. According to the project reserve requirements of the State Grid Marketing Department in 2023, it is necessary to continue to optimize and improve the omni-channel customer demand analysis capabilities, realize the clustering and tracing analysis of various customer demands, support the quality of directional closed-loop problems at all levels, and continue to deepen the new model of service business risk control of "pre-verification, in-process warning, and post-event control". 3). Refine the risk monitoring and control efforts, and improve the service process control management level. The problem analysis dimensions in the marketing service management process need to be refined urgently. Through technical means, a multi-dimensional service problem, handling method, and service process monitoring mechanism should be established to conduct in-depth and detailed service problem analysis to avoid service problems from occurring less or not. 4). The service strategy library needs to be improved urgently to achieve proactive services with user grouping and problem normalization. The current service management process handles service problems in a rather arbitrary manner, and the typical cases and experiences of problem handling in various cities have not been effectively utilized. They are still handled one by one, lacking a standard grassroots handling template and recommended problem solutions, and cannot provide differentiated services.

[0028] The benefit analysis of this project construction is as follows: From the perspective of management benefits, as the company continues to strengthen the control of power supply service quality, the number of complaints has decreased year by year, but from the perspective of customer demands, potential service risks still exist, and typical problems and red line problems in various units still occur from time to time; by conducting customer demand analysis, service early warning and process quality control, we can timely grasp the current main demands of customers and possible service risks; through the construction of a customer attribute library and a service strategy library, combined with historical problem solutions, we can push accurate service handling methods to front-line business personnel, provide decision-making support for operators, effectively tilt resources, optimize work scheduling, improve work accuracy, and achieve early detection, early warning, and fast processing of service risks, and effectively improve the quality of power supply services. From the perspective of economic benefits, by monitoring the service status of each city company and building risk warning scenarios, operation managers can timely discover the current main demands of customers, main service risk points, and differences in quality and efficiency of personnel and things, and through corresponding guidance strategies, concentrate on efficient solutions, improve work efficiency, and achieve optimal resource allocation. At the same time, through the automatic screening of repeated demands and the modular automatic generation of daily analysis reports, the input of business analysts and the secondary risks caused by subjective judgment of personnel can be reduced, and the operating costs of enterprises can be reduced. From the perspective of social benefits, service risk prevention and control and quality management work should be carried out, and massive unstructured data resources should be fully utilized. Combined with risk warning and disposal models, possible service risk escalation events can be predicted, and effective public relations measures can be taken in advance to notify or appease customers, so as to achieve pre-emptive prevention, reduce the outbreak of large-scale group events, improve service levels, establish a good social image of the company, and continuously improve customer perception and customer experience. By analyzing power supply quality issues, the company can understand the operation of the distribution network from the headquarters side, and provide data reference for the provincial company to do a good job in power grid construction and transformation, strengthen operation and maintenance management, and improve power supply controllability.

[0029] Example 2, Technology Development Sub-item: 95598 Audio Data and Data Model Construction Based on Risk Keywords and Risk Warning Application This application is a technical achievement of one of the digital information technology development sub-items of the above-mentioned power grid situation monitoring and risk warning application general project. It is mainly aimed at 95598 audio data and data model construction and risk warning application based on risk keywords, and constructs a data model (including data structure and its internal data relationship) that works well with the digital operation process based on sensitive words under the audio data flow framework of the power grid 95598 platform.

[0030] The system data of this development sub-item is mainly based on 95598 voice translation text data, work orders of the power business application system, 95598 business support work orders, and power outage data of the power consumption information collection system. It is obtained and stored in the service situation monitoring and risk warning application system through RESTAPI, and provided to the analysis application for use. The results of the analysis are simultaneously output as statements (including reports) through the data communication protocol. Other basic power grid data can be subsequently imported, added, or linked as needed. This part is already very mature in the existing technology.

[0031] The core development item of the present invention is to build a data model for linking power grid risk events based on the combination of sensitive words in 95598 audio; by building a complete and in-depth data structure, it lays the foundation for the high-level and in-depth application of the data system in the back end. In the existing model of this technical subject, the corresponding sensitive words are generally provided for specific power grid risk event data by artificial setting. Such a data model means that different sensitive word combinations correspond to different power grid operation risk event data, and sometimes it is necessary to set specific sensitive word combinations for different types of users or user groups, which makes the system very bloated. In particular, this bloating and inefficiency of the data system are reflected in many aspects, including the sensitive word extraction data process, the power grid 95598 audio and power grid data linking link, and the risk warning data processing and presentation process, etc., resulting in the extensiveness and inefficiency of the data system.

[0032] Embodiment 3, main body data structure model Based on the technical background mentioned above and the development and construction of a data model example with a single subset of sensitive words as the underlying data object below, based on the above two considerations, the present invention constructs the following main technical architecture.

[0033] The main technical architecture includes two main lines: data module main line (I) and data link main line (II).

[0034] Example 4: Main Data Structure Model - Data Module Main Line (I) Attached Figure 1 The following is a data structure diagram of a specific development case of this data main line. For this first main line, at the bottom of the data structure, facing the audio data stream under the Grid 95598 platform and the grid risk event data associated with it, a data module set is constructed in the form of a data set. The data set initialization includes several independent data modules: Grid 95598 audio data, grid operation status data, grid risk event data; and other subsequent extensible data modules.

[0035] Furthermore, each data module is directly and consistently characterized by its attribute set, that is, in terms of data structure, the data module itself and the attributes of the data module are directly treated as equals.

[0036] For each attribute value in the attribute set, the underlying data set initialization based on the data model includes power grid 95598 audio data, power grid operation status data, power grid risk event data and allows subsequent expansion of other data module types. The attribute value is initialized to be compatible with the widest range of data object data types, including basic data types such as integers, strings, etc. or text data and other unstructured data.

[0037] In addition, for data modules such as the power grid 95598 audio data, power grid operation status data, and power grid risk event data, a set is constructed outside its attribute set to guide data references and links. This set can be constructed as a subset of the attribute set, or other data sets with unique identification characteristics such as the power grid 95598 audio data ID, the power grid operation data ID, or other identification data (the last part is the association between the first main line direction and the second main line direction).

[0038] Example 5: Main Data Structure Model - Data Link Main Line (II) For the second main line, i.e. the data link main line (II), the data set is based on the data modules such as the power grid 95598 audio data, power grid operation status data, and power grid risk event data, and further establishes its multi-directional data relationship link based on the data relationship between the attribute sets of each data module in principle. Figure 2 This is a schematic diagram of the data structure of a specific development case of this data main line.

[0039] Generally, the guidance of data link relationship takes a subset of the attribute set as the initial data processing object, that is, the data processing relationship and its data processing process are constructed based on the selected subset.

[0040] Specifically, it is common to use any subset of the attribute set as the initialized data processing object guided by the data link relationship; in extreme cases, the entire subset, that is, the complete set of data, is used as the initialized data processing object guided by the data link relationship; among them, the simplest and most practical data selection is: using a single subset of the attribute set as the initialized data processing object guided by the data link relationship.

[0041] In particular, with regard to the data structure of data relations, the data relation links within the data set are constructed as a core-guided type of multi-directional links, that is, the data link initially takes a selected data module as the core, based on the data relations between the attribute sets of each data module, and combined with the pre-set or manually specified data standards for measuring the data relations between the attribute sets, as the data guide for the data relation link, the selected core data module is guided to link with one or more other data modules.

[0042] Embodiment 6: Parameter library data structure (corresponding to multiple links of data modules) For Main Line 2, as well as the collection constructed outside the attribute collection in Main Line 1 to guide data references and links, this part of the technology can be constructed as a data set and / or a side data parameter library in terms of data implementation. The selected central data module can be linked to other data modules according to the guidance of this data parameter library; in this way, the data in the parameter library and the data link channels between different data modules have a one-to-one correspondence, and the data relationships and data links of different data modules can be modified accordingly by modifying, copying, and deleting the parameter library.

[0043] Embodiment 7, data model example The first set of simplified data model samples was constructed: The attribute set of the data module "Power Grid 95598 Audio Data" is set to include: 95598 call audio storage, audio record ID, audio timestamp, audio file path, transcription text, sensitive word extraction tags associated with power grid risk events, sensitive words, other numerical data, and other unstructured data; The attribute set of the power grid operation status data includes the power grid electrical and signal data, as well as the power grid operation and maintenance related data, especially the power outage and fault data. The attribute set of this data module can be constructed based on the existing database of the power system by reference or import and data merging; The attribute set of power grid risk event data is constructed as a two-dimensional array structure. First, the data is expanded along one dimension according to the type of power grid risk event. At the same time, each data column of the first data dimension is expanded along the second dimension according to the risk level of the power grid risk event, thereby obtaining a two-dimensional data attribute set of the power grid risk event data module; when the risk event type and the risk level are numerically equivalent, this two-dimensional array structure is correspondingly represented as a second-order tensor.

[0044] At this time, under the guidance of the data link relationship, a single subset of the attribute set, namely "sensitive words", is used as the initialization data processing object guided by the data link relationship; the data structure of the single subset is: {sensitive words (based on the power grid 95598 call audio and its transcribed text; based on sensitive word extraction tags associated with power grid risk events)}.

[0045] Next, we can consider the following: The link relationship between sensitive words and power grid risk event data is represented by: which sensitive words mean which type of power grid risk event will occur or is likely to occur. At this time, the data link relationship is improved as follows: The first step is to construct all possible sensitive words into a non-repetitive and unified data group, which is mathematically represented as a data set.

[0046] There are two possible implementations of step 2: A. For any selected or subsequently added target power grid risk monitoring event (whether for a specific individual power user or a group of combined power users), a sensitive word subset containing all possible ones is first generated by combinatorial traversal, and then one or several most effective sensitive word combinations are selected from all sensitive word subsets generated under the combinatorial traversal mode and all possible sensitive word combinations therein based on artificial standards or fitting proofreading of historical data as the best sensitive word combination for the selected target power grid risk monitoring event; at the same time, the best sensitive word combination obtained in this way allows dynamic adjustment according to subsequent application effectiveness or other factors; wherein, one or several most effective sensitive word combinations are selected from all sensitive word subsets generated under the combinatorial traversal mode and all possible sensitive word combinations therein based on fitting proofreading of artificial standards or historical data as the best sensitive word combination for the selected target power grid risk monitoring event; Generally, this data step is implemented by building a digitized and programmed data fitting and verification system, which performs comparison and proofreading based on manual analysis, historical data or expert knowledge base. This model also has an additional outstanding advantage, which is that it can easily introduce existing artificial intelligence algorithms for iterative fitting optimization, which can greatly improve the efficiency and accuracy of fitting and proofreading compared to the original data optimization based on application feedback.

[0047] B. Compared with the data path in A above, an equivalent data process is adopted, but all sensitive word subsets obtained in A are further ordered, that is, each sensitive word subset in the sensitive word subset group generates n! new subsets according to the order of sensitive words in the group (where n is the number of sensitive words in the specified sensitive word subset); in this way, a parameter path of the order of appearance of sensitive words is constructed on the data model, that is, a parameter associated with the order in which sensitive words appear in a specific power grid risk event or in the power grid 95598 audio data stream; wherein, each sensitive word subset in the sensitive word subset group generates n! new subsets according to the order of sensitive words in the group, on the one hand, when these ordered subsets are linked to the power grid risk event or the 95598 audio data stream for data point, their orderliness is associated with the time order in which the elements in these ordered subsets appear in the power grid risk event or the 95598 audio data stream, and the orderliness association is constructed in the time dimension; Therefore, compared with implementation path a, implementation path b is an obviously higher-order and more refined data model. Based on the accumulation of big data and the improvement of cloud server analysis computing power, the data correspondence structure between the order of appearance of sensitive words in the 95598 audio data stream of the power grid and the correspondence between risk events has strong certainty and distinctiveness. Therefore, the data model concretized under implementation path B has higher data sophistication and accuracy.

[0048] On the other hand, implementation path B can further introduce more variable parameters in the time correlation of the elements within its ordered subset. For example, in the description of the first aspect above, the default time dimension ordered correlation is based on the chronological order of the appearance of sensitive words. However, this chronological order may not be adopted in the data structure. Instead, it is allowed to set any shuffled order based on the effectiveness of data analysis or the use of data models, as long as the appearance order of the sensitive words finally selected and entered in the time dimension has a certain order.

[0049] The third step is to build the overall data system. A specific implementation path is to transform the data processes in steps 1 and 2 into data relationships and data storage methods in the data system. Specifically: In the final data model, various optimized sensitive word combinations are not actually stored for various power grid risk event data, etc., but only a full set of sensitive word combinations, i.e., the data set in step 1, and the combination parameters corresponding to the best sensitive word combination finally selected, recorded as A parameters, or combination parameters + ordered parameters, recorded as A + B parameters; At the same time, such A parameters or A+B parameters are constructed and stored in the form of instructions to optimize and simplify the data capacity of the data system; in subsequent risk warnings or other data analysis applications, the back-end server directly calls the instructions corresponding to the A parameters or A+B parameters. These instructions may appear in the form of text of just a few bytes and have good calling and transmission efficiency, while the restoration and display of the real sensitive word combination rely on the back-end server for calculation and generation, thereby avoiding the consumption of computing resources of the current data model and optimizing the energy efficiency of the constructed data system in terms of computing power model consumption.

[0050] For other data model instances, when constructing them, considering that the most simplified and practical data are used to select and construct the first set of simplified data model examples, if other types of initialized data processing objects guided by data link relationships using any subset of the attribute set are needed in the future, they can be directly constructed hierarchically and superimposed according to the simplified data model examples.

[0051] Specifically, based on the first set of simplified data model samples, when constructing with any subset of the attribute set as the data base, any element of any subset of the attribute set is treated equivalently to a single subset in the simplified data model sample; that's it. Example 8

[0052] By improving the above-mentioned main data structure model and adding more data modules, an extended data model for multiple applications is constructed. For example, the user demand data set is imported into the bottom layer of the data structure as a subsequent expandable data module, and the sensitive words of power grid risk events are replaced with user demand sensitive words based on the demands of power users. An equivalent user demand data model is constructed using equivalent data processes and data algorithms, which can be used for early warning, statistics, auxiliary construction or other applications of user demand events.

[0053] Other data modules that can be expanded or can be listed separately as data related data sets include: 95598 work order data, power user risk individual attribute data, power user risk group attribute data, power active service data, risk scenario data, · meteorological data (temperature, humidity, wind speed, precipitation, etc. related to power grid operation), · geographic information data (geographic location of power grid equipment, topography, etc.), · socio-economic data (economic development status and population density of power consumption areas, which may affect power demand and power grid risks), · user power consumption behavior data (users' power consumption habits, power consumption trends, etc.), · equipment maintenance record data (equipment maintenance time, maintenance content, maintenance personnel, etc.), · power market data (power price fluctuations, power trading conditions, etc., which may affect power grid operations and risks), etc. In this way, the application of this data model can also be greatly expanded.

[0054] From the above embodiments, it can be seen that the technical solution of the present invention develops a specialized data structure and a multi-directional data relationship link model and a detailed data processing strategy, which solves the deficiencies of the prior art in data processing on the 95598 platform of the power grid. The technical solution of the present invention is scalable, accurate and practical, and brings important technological progress and a basic data framework to the intelligent processing of the 95598 audio data of the power grid and its subsequent risk warning applications and other related applications, and is expected to show a variety of practical application values ​​in the further data application development of the 95598 platform of the power grid.

[0055] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0056] In various embodiments, the hardware implementation of the technology can directly adopt existing intelligent devices, including but not limited to industrial computers, PCs, smart phones, handheld stand-alone computers, floor-standing stand-alone computers, etc. Its input device preferably adopts a screen keyboard, its data storage and calculation module adopts existing memory, calculator, controller, its internal communication module adopts existing communication ports and protocols, and its remote communication adopts existing GPRS network, World Wide Internet, etc.

[0057] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0058] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0059] Each functional unit in each embodiment of the present invention may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of a software functional unit. If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and may also instruct the relevant hardware to complete it through a computer program. The computer program may be stored in a computer-readable storage medium, and the computer program may be executed by the processor to implement the steps of the above-mentioned method embodiments. Among them, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice within the jurisdiction.

[0060] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A data relationship-guided power grid platform audio data stream link model, used for audio digital intelligent linking and risk warning based on sensitive words under the power grid 95598 platform audio data stream framework, characterized by: The data model is constructed with the basic data of the risk warning of the power grid 95598 platform to initialize the data structure. Specifically, the data model initialization includes the following data modules: power grid 95598 audio data, power grid operation status data, power grid risk event data; and other subsequent expandable data modules.

2. According to the data relationship-guided power grid platform audio data stream link model according to claim 1, it is characterized by: In terms of data structure, the data model uses the data form of attribute sets to represent each of the above initialization data modules; among them: The attribute set of the power grid 95598 audio data module is set to include: 95598 call audio storage, audio record ID, audio timestamp, audio file path, transcription text, sensitive word extraction tags associated with power grid risk events, sensitive words, other numerical data, and other unstructured data; The attribute set of the power grid operation status data module includes: electrical data of the power grid operation in the concerned area, signal data of the power grid operation in the concerned area, power outage and fault data of the power grid in the concerned area, non-fault operation and maintenance data of the power grid in the concerned area; other power grid operation status data; the attribute set of this data module can be constructed by reference or import and data merging based on the existing database of the power system; The attribute set of the power grid risk event data module is constructed as a two-dimensional array structure, that is, first, the data is expanded along one dimension according to the type of power grid risk event, that is, different types of power grid risk events are classified and arranged, and on this basis, each data position on the first data dimension is expanded in the second dimension according to the risk level of the power grid risk event, thereby obtaining a two-dimensional data attribute set of the power grid risk event data module.

3. According to the data relationship-guided power grid platform audio data stream link model of claim 1, it is characterized by: In terms of data relationships and data links, the data model essentially links different data according to the real data relationships within each data module, such as the power grid 95598 audio data, power grid operation status data, and power grid risk event data; In terms of form, it is to build a data link structure that has both a center and is compatible with multi-directional links.

4. According to claim 3, a data relationship-guided power grid platform audio data stream link model is characterized by: Specifically: the data link is centered on a certain selected initial data module, that is, one of the data modules such as the power grid 95598 audio data, the power grid operation status data, the power grid risk event data, etc. is selected as the central data module, and then based on the data relationship between the attribute sets of other data modules and the initial data module as the center, and combined with the pre-set or manually specified data standards for measuring the data relationship between the attribute sets of different data modules, it serves as a data guide for other data modules to link data relationships to the central data module; in terms of data implementation, these data guides are also constructed as a data set and or a side data parameter library, and the selected central data module is linked to other data modules according to the guidance of this data parameter library; In this way, the data in the parameter library and the data link channels between different data modules have a one-to-one correspondence. The data relationships and data links of different data modules can be modified accordingly by modifying, copying, and deleting the parameter library.

5. According to claim 4, a data relationship-guided power grid platform audio data stream link model is characterized in that: More specifically, in guiding the data link relationship, the following simplified and most practical data path is adopted: the data module power grid 95598 audio data is used as the initial central data module, and then a certain attribute value in the attribute set of the initial data module (that is, a single subset of the attribute set, rather than other types of subsets such as a subset of the attribute set containing more than two attribute values) is used as the data parameter reference value for guiding the data link relationship. Specifically, the data link relationship is guided by a single subset of the attribute set of the central data module power grid 95598 audio data, namely, the "sensitive word" attribute; wherein, the data sources of sensitive words include: based on the power grid 95598 call audio and its transcribed text, and based on sensitive word extraction tags associated with power grid risk events.

6. According to claim 5, a data relationship-guided power grid platform audio data stream link model is characterized by: The sensitive words have two data dimension correlations: based on the 95598 power grid call audio and its transcribed text; Based on the sensitive word extraction tags associated with power grid risk events; for example, sensitive words are identified from the power grid 95598 call audio and its transcribed text according to the sensitive word extraction tags associated with power grid risk events.

7. The application of an audio data stream link model of a power grid platform guided by data relationships as described in any one of claims 1-6, including application A: based on the audio data stream of the power grid 95598 platform, early warning, statistics, auxiliary construction or other applications of single power outages, frequent power outage risks, power grid failures, electricity billing disputes, power user complaints, power complaint escalation risks, group power risk events, power risk event location positioning and other power grid risk events.

8. The application of a data relationship-guided power grid platform audio data stream linking model as described in any one of claims 1-6, including application B: auxiliary construction of power complaint user portraits based on the data output of the 95598 voice sensitive word system.

9. An application of an audio data stream linking model of a power grid platform guided by data relationships as described in any one of claims 1-6, including application C: replacing sensitive words of power grid risk events with sensitive words of user demands based on the demands of power users, thereby importing the user demand data set as other subsequent expandable data modules into the bottom layer of the data structure, and using equivalent data processes and data algorithms to build an equivalent user demand data model for early warning, statistics, auxiliary construction or other applications of user demand events.

Images