Project data fusion method and device based on power grid business operation management
By obtaining and converging text and image feature vectors of project data from multiple systems and platforms in power grid enterprises, the problems of data inconsistency and system silos are solved, and the efficiency and consistency of project management are improved.
Patent Information
- Application Number
- CN202410947801.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-16
- Publication Date
- 2024-11-12
AI Technical Summary
Power grid companies face data inconsistencies and system silos, resulting in project management complexity and processing delays, affecting business operation efficiency and project success rate.
By obtaining project data from multiple systems and platforms, keywords in text and images are extracted according to project identifiers, converted into feature vectors and fused, and unified data management and processing are achieved.
It reduces the complexity of project management, improves the consistency and processing efficiency of project data, and overcomes the problems of data inconsistency and system silos.
Smart Images

Figure CN118940207A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of power grid business data processing, and in particular to a project data fusion method and device based on power grid business operation management. Background Art
[0002] With the transformation of the global energy structure and the advancement of technology, power grid companies are facing unprecedented challenges and opportunities. Especially in the context of the rapid development of information technology, the digital transformation of power grids has become the key to improving corporate competitiveness and responding to market changes. Power grid companies not only need to deal with complex power grid operation issues, but also must optimize project management processes to adapt to rapidly changing technology and market demands. This environment requires power grid companies to continuously innovate and improve their business operations and project management models.
[0003] The core of digital transformation is to use advanced information technology to integrate and optimize internal resources and processes. The introduction of digital project management systems can effectively solve problems such as cumbersome multi-system operations, difficulty in centralized data management, and inefficient project tracking, and achieve digital management of the entire process from project planning, implementation to monitoring and evaluation. How to overcome problems such as inconsistent data and system islands within an enterprise has become a research focus. Summary of the invention
[0004] In view of this, the embodiment of the present application provides a project data fusion method and device based on power grid business operation management, which can overcome the problems of data inconsistency and system islands within the enterprise, reduce the complexity of project management, improve the consistency of project data, and reduce processing delays.
[0005] In order to achieve the above objectives, this application adopts the following technical solutions:
[0006] In a first aspect, an embodiment of the present application provides a project data fusion method based on power grid business operation management, including:
[0007] Acquire project data for power grid business operation management, the project data including multiple texts and multiple images, the project data originating from multiple systems and platforms, the multiple texts and the multiple images both including project identification information;
[0008] Determining a target text and a target image corresponding to a target item in the project data based on the item identification information, the target item being any item in the project data;
[0009] Extracting a first keyword from the target text, and converting the first keyword into a first feature vector;
[0010] Extracting a second keyword from the target image, and converting the second keyword into a second feature vector;
[0011] The first feature vector and the second feature vector are fused to obtain a fused feature vector of the target project, and the project data is managed based on each of the fused feature vectors.
[0012] Based on the first aspect, in some embodiments, obtaining project data for power grid business operation management includes:
[0013] Sending a project data acquisition request to various systems and platforms associated with the power grid business, wherein the project data acquisition request includes a project identifier;
[0014] Receiving project data sent by various systems and platforms associated with power grid services, wherein the project data is obtained by various systems and platforms responding to the project data acquisition request to retrieve project data corresponding to the project identifier, and the project data includes the corresponding project identifier;
[0015] Wherein, the project identifier is one or more. In the case where there are multiple project identifiers, after acquiring the project data, the project identifier in the project data is extracted and each project data containing the same project identifier is combined into a project data set.
[0016] Based on the first aspect, in some embodiments, obtaining project data for power grid business operation management includes:
[0017] Sending a project data acquisition request to various systems and platforms associated with the power grid business, wherein the project data acquisition request includes a time period;
[0018] Receiving project data sent by various systems and platforms associated with power grid services, wherein the project data is obtained by various systems and platforms responding to the project data acquisition request to retrieve project data within the time period;
[0019] Extracting project identification keywords from each project data, and determining the project identification of each project data according to the project identification keywords;
[0020] In the case where there are multiple project identifiers, each project data containing the same project identifier is combined into a project data set to obtain multiple project data sets.
[0021] Based on the first aspect, in some embodiments, extracting the first keyword from the target text includes:
[0022] Segmenting a target text identified by a first project to determine a first frequency and position of each word in the target text, wherein the first project identification is any project identification;
[0023] Calculate the span of each word in the target text identified by the first item based on the ratio of the distance between the first position where each word first appears and the second position where it last appears in the target text identified by the first item to the total number of words in the target text;
[0024] Recording M words with the largest product of span and first frequency as candidate keywords of the target text identified by the first item, where M is an integer greater than or equal to 3;
[0025] Determine a second frequency of occurrence of each candidate keyword in the target text identified by other items;
[0026] The difference between the first frequency and the second frequency is calculated, and the N candidate keywords with the largest difference are recorded as the first keywords of the target text, where N is an integer greater than or equal to 2, and N is less than M.
[0027] Based on the first aspect, in some embodiments, extracting the second keyword from the target image includes:
[0028] Extracting at least one text image area from a preset area in the target image;
[0029] Preprocess the text image area to obtain the text image to be recognized;
[0030] Connected regions are extracted from the text image to be recognized to obtain multiple connected regions;
[0031] Determine a text connected region among the plurality of connected regions;
[0032] Detecting the outline of each character in the text connected area, segmenting to obtain multiple characters based on the outline and recording the positional relationship of each character in the text connected area;
[0033] Identify the plurality of characters, and generate text to be extracted according to the positional relationship of each character in the text connected area;
[0034] A second keyword is extracted from the text to be extracted.
[0035] Based on the first aspect, in some embodiments, preprocessing the text image region to obtain the text image to be recognized includes:
[0036] Converting the text image region into a grayscale image region;
[0037] Performing binarization processing on the grayscale image region to obtain a binarized image region;
[0038] Determine the boundary of the grayscale image area according to the binary image area;
[0039] Generate a quadrilateral according to the boundary fitting of the grayscale image area;
[0040] The positions of the corner points of the quadrilateral are adjusted so that the quadrilateral becomes a rectangle, thereby obtaining the text image to be recognized.
[0041] Based on the first aspect, in some embodiments, the connected region extraction is performed on the text image to be recognized to obtain multiple connected regions, including:
[0042] Determine a pixel cluster in each row of pixels of the to-be-recognized text image, wherein each pixel in the pixel cluster is connected and each of the grayscale values is a preset grayscale value;
[0043] Pixel clusters connected to each other in different rows are determined according to the positions of each pixel cluster to obtain a connected region; wherein, in the same connected region, there are pixel points located in the same column between pixel clusters in adjacent rows.
[0044] Based on the first aspect, in some embodiments, converting the first keyword into a first feature vector includes: there are multiple first keywords, each element in the first feature vector corresponds to a first keyword, and according to the frequency of occurrence of each first keyword in the target text multiplied by a preset weight, a value of the element corresponding to each first keyword is calculated to obtain the first feature vector;
[0045] The second keyword is converted into a second feature vector, including: there are multiple second keywords, each element in the second feature vector corresponds to a second keyword, and the value of the element corresponding to each second keyword is calculated according to the frequency of occurrence of each second keyword in the target image multiplied by a preset weight to obtain the second feature vector.
[0046] Based on the first aspect, in some embodiments, fusing the first feature vector with the second feature vector to obtain a fused feature vector of the target item includes:
[0047] According to the values of the elements at the corresponding positions of the first feature vector and the second feature vector, the values of the elements of the fused feature vector are calculated based on a preset algorithm; wherein, if an element of the first feature vector does not exist, the value of the element of the fused feature vector is determined according to the value of the corresponding element of the second feature vector; if an element of the second feature vector does not exist, the value of the element of the fused feature vector is determined according to the value of the corresponding element of the first feature vector;
[0048] Generate a vector quantity element and a fusion algorithm element as two elements of the fusion feature vector; wherein the vector quantity element represents the quantity of the fusion feature vector, and the fusion algorithm element represents the preset algorithm;
[0049] Generate a fused feature vector according to the values of the elements of the fused feature vector calculated by a preset algorithm, as well as the vector quantity element and the fusion algorithm element;
[0050] The first feature vector, the second feature vector and the fused feature vector are associated with the project identifier and stored in the same location.
[0051] In a second aspect, an embodiment of the present application provides a project data fusion device based on power grid business operation management, including:
[0052] A data acquisition module, used to acquire project data for power grid business operation management, wherein the project data includes multiple texts and multiple images, the project data is derived from multiple systems and platforms, and the multiple texts and the multiple images both contain project identification information;
[0053] a target determination module, configured to determine a target text and a target image corresponding to a target project in the project data based on the project identification information, wherein the target project is any project in the project data;
[0054] A first keyword module, used to extract a first keyword from the target text and convert the first keyword into a first feature vector;
[0055] A second keyword module, used for extracting a second keyword from the target image and converting the second keyword into a second feature vector;
[0056] A fusion module is used to fuse the first feature vector and the second feature vector to obtain a fused feature vector of the target project, and manage the project data based on each of the fused feature vectors.
[0057] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:
[0058] In the embodiment of the present application, the project data of the power grid business operation management is obtained, and the project data comes from multiple systems and platforms. The text and image of a certain project are screened out from the project data according to the project identifier of the project data, and keywords are extracted from the screened images and texts and converted into feature vectors. The feature vectors of the images and texts are merged, thereby merging the project data of multiple systems and platforms related to the power grid business. When it is necessary to process the power grid business of a certain project, the corresponding project data can be retrieved for processing, thereby reducing the complexity of project management, improving the consistency of project data, and reducing processing delays, and overcoming problems such as inconsistent data and system islands within the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0060] Figure 1 It is a flowchart of a project data fusion method based on power grid business operation management provided by an embodiment of the present application;
[0061] Figure 2 It is a structural diagram of a project data fusion device based on power grid business operation management provided in an embodiment of the present application;
[0062] Figure 3 It is a schematic diagram of a terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0063] The present application is described more clearly below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the effects of the present application, but are not intended to limit the present application in any form. It should be noted that, for those of ordinary skill in the art, several variations and improvements may be made without departing from the concept of the present application. These all fall within the scope of protection of the present application.
[0064] In order to make the purpose, technical solutions and advantages of the present application clearer, specific embodiments will be described below in conjunction with the accompanying drawings.
[0065] In the context of digital transformation, power grid companies are facing many problems and challenges in project management, which seriously affect the efficiency of business operations and the success rate of projects. Traditional project management often involves multiple systems and platforms, resulting in frequent information islands. This not only increases the complexity of project management, but also leads to inconsistent data and processing delays, thus affecting the timeliness and accuracy of decision-making.
[0066] The operation steps in business processing are numerous and complex, especially in a multi-system environment, where each business operation often requires logging into multiple systems to complete. For example, the entire process from project initiation to data archiving may require going through multiple different systems and tools, such as ERP (Enterprise Resource Planning), financial management systems, project management software, etc. This decentralized system operation greatly prolongs business processing time, increases operational complexity, and increases the risk of errors.
[0067] Based on this, in the embodiment of the present application, the project data of multiple systems and platforms can be obtained in advance through the terminal device, the text and image of a certain project can be screened out from these project data according to the project identification, the keywords can be extracted from the screened images and texts and converted into feature vectors, and the feature vectors of the images and texts can be merged, thereby merging the project data of multiple systems and platforms related to the power grid business in one place. When it is necessary to process the power grid business of a certain project, the corresponding project data can be retrieved for processing, thereby reducing the complexity of project management, improving the consistency of project data, and reducing processing delays.
[0068] See also Figure 1 The project data fusion method based on power grid business operation management provided in the embodiment of the present application may include the following steps:
[0069] Step 101, obtaining project data for power grid business operation management.
[0070] The project data includes multiple texts and multiple images, the project data comes from multiple systems and platforms, and the multiple texts and multiple images all contain project identification information. Exemplarily, the above systems and platforms can be ERP systems, i6000 systems, financial management systems, project management platforms, etc.
[0071] In some embodiments, step 101 may specifically include: sending a project data acquisition request to each system and platform associated with the power grid business, wherein the project data acquisition request includes a project identifier; receiving project data sent by each system and platform associated with the power grid business, wherein the project data is obtained by each system and platform responding to the project data acquisition request to retrieve project data corresponding to the project identifier, wherein the project data includes the corresponding project identifier. The project identifier is one or more, and in the case where there are multiple project identifiers, after the project data is acquired, the project identifier in the project data is extracted and each project data containing the same project identifier is synthesized into a project data set.
[0072] In some other embodiments, step 101 may specifically include: sending a project data acquisition request to each system and platform associated with the power grid business, wherein the project data acquisition request includes a time period; receiving project data sent by each system and platform associated with the power grid business, wherein the project data is obtained by each system and platform in response to the project data acquisition request to retrieve project data within the time period; extracting project identification keywords from each project data, and determining the project identification of each project data based on the project identification keywords; in the case of multiple project identifications, combining each project data containing the same project identification into a project data set to obtain multiple project data sets. Exemplarily, the above process of extracting project identification keywords from each project data may be: extracting project identification keywords based on the name of the project data.
[0073] In the embodiment of the present application, a project data acquisition request can be sent to each system and platform through a terminal device, and the project data acquisition request includes a project identifier, and the project data in each system and platform includes the project identifier. At this time, each system and platform can send the project data corresponding to the project identifier to the terminal device. Alternatively, a project data acquisition request can be sent to each system and platform through a terminal device, and the project data acquisition request includes a time period. The project data in each system and platform does not directly include a project identifier. The project identifier keywords in each project data can be extracted first, and the project identifier of each project data can be determined based on the project identifier keywords. Each system and platform then sends the project data within the time period to the terminal device.
[0074] For example, the embodiment of the present application can be pre-executed to send the project data of certain projects to the terminal device, which processes the project data to achieve the fusion of the project data. When the project data of a certain project needs to be processed, the fused project data can be retrieved for processing.
[0075] Step 102: determining a target text and a target image corresponding to a target project in the project data based on the project identification information.
[0076] In this step, the target item is any item in the item data.
[0077] In some embodiments, project identification information may be extracted from all texts and images of the project data, and then the target project identification of the target project may be matched with the project identification information extracted from each text and image to determine the target text and target image of the target project.
[0078] Specifically, the above operation is performed on each project, that is, all texts and images in the acquired project data can be classified, and the project corresponding to each text and image can be determined to facilitate subsequent processing.
[0079] Step 103: extract a first keyword from the target text, and convert the first keyword into a first feature vector.
[0080] In some embodiments, the process of extracting the first keyword from the target text may specifically include: segmenting the target text identified by the first project, determining the first frequency and position of each word in the target text, where the first project identifier is any project identifier; calculating the span of each word in the target text identified by the first project based on the ratio of the distance between the first position where each word first appears and the second position where it last appears in the target text identified by the first project, and the total number of words in the target text; recording the M words with the largest product of the span and the first frequency as candidate keywords for the target text identified by the first project, where M is an integer greater than or equal to 3; determining the second frequency of each candidate keyword in the target text identified by other projects; calculating the difference between the first frequency and the second frequency, and recording the N candidate keywords with the largest difference as the first keyword of the target text, where N is an integer greater than or equal to 2, and N is less than M.
[0081] In this step, N candidate keywords with large spans and large frequency differences are used as the first keywords of the target text, which can better reflect the characteristics of the target text that are different from other texts and improve the accuracy of keyword extraction.
[0082] In some other embodiments, the process of extracting the first keyword from the target text may specifically include: inputting the target text into a deep learning model, and outputting the first keyword of the target text. The deep learning model may include multiple first network models and a second network model, the target text is first input into the multiple first network models, the multiple first network models respectively output a group of candidate keywords, the multiple groups of candidate keywords are then input into the second network model, and the second network model then outputs the final first keyword.
[0083] In some embodiments, the implementation process of "converting the first keyword into a first feature vector" in step 103 may include: there are multiple first keywords, each element in the first feature vector corresponds to a first keyword, and the value of the element corresponding to each first keyword is calculated according to the frequency of occurrence of each first keyword in the target text multiplied by a preset weight to obtain the first feature vector.
[0084] Specifically, the number of elements in the first feature vector and the keyword corresponding to each element can be pre-specified, and the keyword corresponding to each element in the first feature vector and the value of the keyword (ie, the value of the element) can be known.
[0085] The value of the keyword can be calculated by multiplying the frequency of the first keyword in the target text by the preset weight. Exemplarily, the frequency of the first keyword in the target text is the result of dividing the number of times the first keyword appears in the target text by the total number of words in the target text, and the preset weight is pre-set based on the importance of each first keyword.
[0086] Step 104: extract a second keyword from the target image, and convert the second keyword into a second feature vector.
[0087] In some embodiments, the process of extracting the second keyword from the target image may specifically include: extracting at least one text image area from a preset area in the target image; preprocessing the text image area to obtain a text image to be recognized; performing connected area extraction on the text image to be recognized to obtain multiple connected areas; determining a text connected area among the multiple connected areas; detecting the outline of each character in the text connected area, segmenting to obtain multiple characters based on the outline and recording the positional relationship of each character in the text connected area; recognizing the multiple characters, and generating a text to be extracted based on the positional relationship of each character in the text connected area; and extracting the second keyword from the text to be extracted.
[0088] In this step, the target image can be an image of some standard documents. For the target image of the standard document, the text image area where the content to be extracted is located can be extracted from the target image first, and then keyword extraction is performed in the extracted text image area. First, multiple connected areas are determined, and then a text connected area is determined among the multiple connected areas. Then, characters are segmented from the text connected area and the positional relationship of each character in the text connected area is recorded. Next, each segmented character is identified, and then the positional relationship of each character in the text connected area is used to generate the text to be extracted. Finally, the second keyword is extracted from the text to be extracted.
[0089] Exemplarily, the standard file usually includes the file name and the specific content of the file, so in this step, the second keyword can be extracted from the text corresponding to the file name, or the second keyword can be extracted from the text corresponding to the file name and the specific content of the file. The process of extracting the second keyword from the text to be extracted can refer to the relevant content in step 103, which will not be repeated here.
[0090] In some scenarios, the target image may be distorted and deformed during the generation process due to reasons such as improper placement of standard files. In this case, the target image needs to be corrected before character recognition and keyword extraction to improve recognition accuracy. Based on this, the specific implementation process of the above-mentioned "preprocessing the text image area to obtain the text image to be recognized" may include: converting the text image area into a grayscale image area; binarizing the grayscale image area to obtain a binary image area; determining the boundary of the grayscale image area based on the binary image area; generating a quadrilateral based on the boundary fitting of the grayscale image area; adjusting the corner point positions of the quadrilateral so that the quadrilateral is a rectangle to obtain the text image to be recognized.
[0091] Exemplarily, the specific implementation process of the above-mentioned "extracting connected regions of the text image to be recognized to obtain multiple connected regions" may include: determining pixel clusters in each row of pixels of the text image to be recognized, wherein each pixel in the pixel cluster is connected and the grayscale values are all preset grayscale values; determining pixel clusters that are interconnected between different rows according to the positions of each pixel cluster to obtain connected regions; wherein, in the same connected region, there are pixel points located in the same column between pixel clusters in adjacent rows.
[0092] Specifically, in the text image to be recognized, the grayscale values of the pixels corresponding to the text characters are the same, that is, the preset grayscale value. Based on this, the grayscale values of the pixels in each row of the text image to be recognized can be compared with the preset grayscale value, and the pixels with the same grayscale value as the preset grayscale value can be found, that is, the pixels of the text characters can be found. If these pixels are connected, they form a pixel cluster. Thus, the pixel clusters in each row of pixels are found, and then it is determined that the connected pixel clusters form multiple connected areas, that is, the areas where the text characters are located.
[0093] In some embodiments, the implementation process of "converting the second keyword into a second feature vector" in step 104 may include: there are multiple second keywords, each element in the second feature vector corresponds to a second keyword, and the value of the element corresponding to each second keyword is calculated according to the frequency of occurrence of each second keyword in the target image multiplied by a preset weight to obtain the second feature vector.
[0094] Specifically, the number of elements in the second feature vector and the keyword corresponding to each element can be pre-specified, and the keyword corresponding to each element in the second feature vector and the value of the keyword (ie, the value of the element) can be known.
[0095] The value of the keyword can be calculated by multiplying the frequency of the second keyword appearing in the target image by a preset weight. Exemplarily, the frequency of the second keyword appearing in the target text is the result obtained by dividing the number of times the second keyword appears in the target text by the total number of words in the target image, and the preset weight is pre-set based on the importance of each second keyword.
[0096] Step 105: fuse the first feature vector and the second feature vector to obtain a fused feature vector of the target project, and manage the project data based on each of the fused feature vectors.
[0097] In some embodiments, the specific implementation process of step 105 may include: calculating the value of the element of the fused feature vector based on a preset algorithm according to the values of the elements at corresponding positions of the first feature vector and the second feature vector; wherein, if an element of the first feature vector does not exist, determining the value of the element of the fused feature vector according to the value of the corresponding element of the second feature vector; if an element of the second feature vector does not exist, determining the value of the element of the fused feature vector according to the value of the corresponding element of the first feature vector; generating a vector quantity element and a fusion algorithm element as two elements of the fused feature vector; wherein, the vector quantity element represents the number of fused feature vectors, and the fusion algorithm element represents the preset algorithm; generating a fused feature vector according to the values of the elements of the fused feature vector calculated by the preset algorithm, as well as the vector quantity element and the fusion algorithm element; associating the first feature vector, the second feature vector and the fused feature vector with the project identifier and storing them in the same location.
[0098] For example, the preset algorithm may be to add the values of the elements at the corresponding positions of the first eigenvector and the second eigenvector. Alternatively, the preset algorithm may be to synthesize the values of the elements at the corresponding positions of the first eigenvector and the second eigenvector into a new value, for example, if the value of an element of the first eigenvector is 11 and the value of the element at the corresponding position of the second eigenvector is 23, then a new value 1123 is synthesized.
[0099] If an element of the first eigenvector does not exist, the value of the element of the fused eigenvector is determined according to the value of the corresponding element of the second eigenvector. For example, if the value of an element of the second eigenvector is 23 and the value of the element at the corresponding position of the first eigenvector does not exist, a new value 0023 is synthesized.
[0100] If an element of the second eigenvector does not exist, the value of the element of the fused eigenvector is determined according to the value of the corresponding element of the first eigenvector. For example, if the value of an element of the first eigenvector is 11 and the value of the element at the corresponding position of the second eigenvector does not exist, a new value of 1100 is synthesized.
[0101] The above-mentioned project data fusion method based on power grid business operation management can obtain project data of multiple systems and platforms through terminal devices in advance, filter out text and images of a certain project from these project data according to project identification, extract keywords from the filtered images and texts and convert them into feature vectors, fuse the feature vectors of images and texts, and thus fuse project data of multiple systems and platforms related to power grid business in one place. When it is necessary to process the power grid business of a certain project, the corresponding project data can be retrieved for processing, thereby reducing the complexity of project management, improving the consistency of project data, and reducing processing delays.
[0102] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0103] Corresponding to the project data fusion method based on power grid business operation management described in the above embodiment, Figure 2 A structural block diagram of a project data fusion device based on power grid business operation management provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown.
[0104] See also Figure 2 The embodiment of the present application provides a project data fusion device based on power grid business operation management, including a data acquisition module 201, a target determination module 202, a first keyword module 203, a second keyword module 204 and a fusion module 205.
[0105] The data acquisition module 201 is used to acquire project data for power grid business operation management, wherein the project data includes multiple texts and multiple images, and the project data originates from multiple systems and platforms, and the multiple texts and multiple images both contain project identification information.
[0106] The target determination module 202 is used to determine a target text and a target image corresponding to a target item in the project data based on the item identification information, and the target item is any item in the project data.
[0107] The first keyword module 203 is used to extract a first keyword from the target text and convert the first keyword into a first feature vector.
[0108] The second keyword module 204 is used to extract a second keyword from the target image and convert the second keyword into a second feature vector.
[0109] The fusion module 205 is used to fuse the first feature vector and the second feature vector to obtain a fused feature vector of the target project, and manage the project data based on each of the fused feature vectors.
[0110] Optionally, the data acquisition module 201 can be specifically used to: send a project data acquisition request to each system and platform associated with the power grid business, the project data acquisition request including a project identifier; receive project data sent by each system and platform associated with the power grid business, the project data is obtained by each system and platform responding to the project data acquisition request to retrieve project data corresponding to the project identifier, and the project data includes the corresponding project identifier; wherein the project identifier is one or more, and in the case where there are multiple project identifiers, after acquiring the project data, the project identifier in the project data is extracted and each project data containing the same project identifier is combined into a project data set.
[0111] Optionally, the data acquisition module 201 can be specifically used to: send a project data acquisition request to each system and platform associated with the power grid business, the project data acquisition request including a time period; receive project data sent by each system and platform associated with the power grid business, the project data is obtained by each system and platform responding to the project data acquisition request to retrieve project data within the time period; extract project identification keywords in each project data, and determine the project identification of each project data based on the project identification keywords; in the case of multiple project identifications, combine each project data containing the same project identification into a project data set to obtain multiple project data sets.
[0112] Optionally, the first keyword module 203 can be specifically used to: segment the target text of the first project identifier to determine the first frequency and position of each word in the target text, where the first project identifier is any project identifier; calculate the span of each word in the target text of the first project identifier based on the ratio of the distance between the first position where each word first appears and the second position where it last appears in the target text of the first project identifier to the total number of words in the target text; record the M words with the largest product of the span and the first frequency as candidate keywords of the target text of the first project identifier, where M is an integer greater than or equal to 3; determine the second frequency of each candidate keyword in the target text of other project identifiers; calculate the difference between the first frequency and the second frequency, and record the N candidate keywords with the largest difference as the first keyword of the target text, where N is an integer greater than or equal to 2, and N is less than M.
[0113] Optionally, the second keyword module 204 can be specifically used to: extract at least one text image area from a preset area in the target image; preprocess the text image area to obtain a text image to be recognized; perform connected area extraction on the text image to be recognized to obtain multiple connected areas; determine a text connected area among the multiple connected areas; detect the outline of each character in the text connected area, obtain multiple characters based on the outline segmentation and record the positional relationship of each character in the text connected area; recognize the multiple characters, and generate a text to be extracted based on the positional relationship of each character in the text connected area; extract the second keyword from the text to be extracted.
[0114] Optionally, the preprocessing of the text image area to obtain the text image to be recognized may specifically include: converting the text image area into a grayscale image area; binarizing the grayscale image area to obtain a binary image area; determining the boundary of the grayscale image area based on the binary image area; generating a quadrilateral based on the boundary fitting of the grayscale image area; adjusting the corner point positions of the quadrilateral so that the quadrilateral is a rectangle, thereby obtaining the text image to be recognized.
[0115] Optionally, the connected region extraction of the text image to be recognized to obtain multiple connected regions may specifically include: determining pixel clusters in each row of pixel points of the text image to be recognized, wherein each pixel point in the pixel cluster is connected and the grayscale values are preset grayscale values; determining pixel clusters that are interconnected between different rows according to the positions of each pixel cluster to obtain connected regions; wherein, there are pixel points located in the same column between pixel clusters in adjacent rows in the same connected region.
[0116] Optionally, the first keyword module 203 may be specifically used for: when there are multiple first keywords, each element in the first feature vector corresponds to a first keyword, and according to the frequency of occurrence of each first keyword in the target text multiplied by a preset weight, the value of the element corresponding to each first keyword is calculated to obtain the first feature vector;
[0117] The second keyword module 204 can be specifically used for: when there are multiple second keywords, each element in the second feature vector corresponds to a second keyword, and the value of the element corresponding to each second keyword is calculated by multiplying the frequency of occurrence of each second keyword in the target image by a preset weight to obtain the second feature vector.
[0118] Optionally, the fusion module 205 can be specifically used to: calculate the value of the element of the fused feature vector based on a preset algorithm according to the value of the element at the corresponding position of the first feature vector and the second feature vector; wherein, if an element of the first feature vector does not exist, determine the value of the element of the fused feature vector according to the value of the corresponding element of the second feature vector; if an element of the second feature vector does not exist, determine the value of the element of the fused feature vector according to the value of the corresponding element of the first feature vector; generate a vector quantity element and a fusion algorithm element as two elements of the fused feature vector; wherein, the vector quantity element represents the number of fused feature vectors, and the fusion algorithm element represents the preset algorithm; generate a fused feature vector according to the value of the element of the fused feature vector calculated by the preset algorithm, as well as the vector quantity element and the fusion algorithm element; associate the first feature vector, the second feature vector and the fused feature vector with the project identifier and store them in the same location.
[0119] Figure 3 Schematic diagram of a terminal device provided by an embodiment of the present invention. Figure 3 As shown, the terminal device 300 of this embodiment includes a processor 310 and a memory 320. The memory 320 stores a computer program that can be run on the processor 310, such as a project data fusion program based on power grid business operation management. When the processor 310 executes the computer program, the steps in the above-mentioned project data fusion method embodiment based on power grid business operation management are implemented, such as Figure 1 Alternatively, when the processor 310 executes the computer program, the functions of each module / unit in the above-mentioned device embodiments are realized, for example, Figure 2 Functions of modules 201 to 205 are shown.
[0120] The terminal device 300 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal device may include, but is not limited to, a processor 310 and a memory 320. Those skilled in the art will appreciate that Figure 3 It is only an example of the terminal device 300 and does not constitute a limitation of the terminal device 300. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include input and output devices, network access devices, buses, etc.
[0121] The processor 320 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0122] The memory 320 may be an internal storage unit of the terminal device 300, such as a hard disk or memory of the terminal device 300. The memory 320 may also be an external storage device of the terminal device 300, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device 300. Further, the memory 320 may also include both an internal storage unit of the terminal device 300 and an external storage device. The memory 320 is used to store the computer program and other programs and data required by the terminal device. The memory 320 may also be used to temporarily store data that has been output or is to be output.
[0123] 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.
[0124] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0125] 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 project data fusion method based on power grid business operation management, characterized in that: include: Acquire project data for power grid business operation management, the project data including multiple texts and multiple images, the project data originating from multiple systems and platforms, the multiple texts and the multiple images both including project identification information; Determining a target text and a target image corresponding to a target item in the project data based on the item identification information, the target item being any item in the project data; Extracting a first keyword from the target text, and converting the first keyword into a first feature vector; Extracting a second keyword from the target image, and converting the second keyword into a second feature vector; The first feature vector and the second feature vector are fused to obtain a fused feature vector of the target project, and the project data is managed based on each of the fused feature vectors.
2. The project data fusion method based on power grid business operation management according to claim 1 is characterized in that: The obtaining of project data for power grid business operation management includes: Sending a project data acquisition request to various systems and platforms associated with the power grid business, wherein the project data acquisition request includes a project identifier; Receiving project data sent by various systems and platforms associated with power grid services, wherein the project data is obtained by various systems and platforms responding to the project data acquisition request to retrieve project data corresponding to the project identifier, and the project data includes the corresponding project identifier; Wherein, the project identifier is one or more. In the case where there are multiple project identifiers, after acquiring the project data, the project identifier in the project data is extracted and each project data containing the same project identifier is combined into a project data set.
3. The project data fusion method based on power grid business operation management according to claim 1 is characterized in that: The obtaining of project data for power grid business operation management includes: Sending a project data acquisition request to various systems and platforms associated with the power grid business, wherein the project data acquisition request includes a time period; Receiving project data sent by various systems and platforms associated with power grid services, wherein the project data is obtained by various systems and platforms responding to the project data acquisition request to retrieve project data within the time period; Extracting project identification keywords from each project data, and determining the project identification of each project data according to the project identification keywords; In the case where there are multiple project identifiers, each project data containing the same project identifier is combined into a project data set to obtain multiple project data sets.
4. The project data fusion method based on power grid business operation management according to claim 2 or 3, characterized in that: The extracting the first keyword from the target text includes: Segmenting a target text identified by a first project to determine a first frequency and position of each word in the target text, wherein the first project identification is any project identification; Calculate the span of each word in the target text identified by the first item based on the ratio of the distance between the first position where each word first appears and the second position where it last appears in the target text identified by the first item to the total number of words in the target text; Recording M words with the largest product of span and first frequency as candidate keywords of the target text identified by the first item, where M is an integer greater than or equal to 3; Determine a second frequency of occurrence of each candidate keyword in the target text identified by other items; The difference between the first frequency and the second frequency is calculated, and the N candidate keywords with the largest difference are recorded as the first keywords of the target text, where N is an integer greater than or equal to 2, and N is less than M.
5. The project data fusion method based on power grid business operation management according to claim 1 is characterized in that: The extracting the second keyword from the target image comprises: Extracting at least one text image area from a preset area in the target image; Preprocess the text image area to obtain the text image to be recognized; Connected regions are extracted from the text image to be recognized to obtain multiple connected regions; Determine a text connected region among the plurality of connected regions; Detecting the outline of each character in the text connected area, segmenting to obtain multiple characters based on the outline and recording the positional relationship of each character in the text connected area; Identify the multiple characters, and generate text to be extracted according to the positional relationship of each character in the text connected area; A second keyword is extracted from the text to be extracted.
6. The project data fusion method based on power grid business operation management according to claim 5 is characterized in that: The preprocessing of the text image region to obtain the text image to be recognized includes: Converting the text image region into a grayscale image region; Performing binarization processing on the grayscale image region to obtain a binarized image region; Determine the boundary of the grayscale image area according to the binary image area; Generate a quadrilateral according to the boundary fitting of the grayscale image area; The positions of the corner points of the quadrilateral are adjusted so that the quadrilateral becomes a rectangle, thereby obtaining the text image to be recognized.
7. The project data fusion method based on power grid business operation management according to claim 5 is characterized in that: The connected regions are extracted from the text image to be recognized to obtain multiple connected regions, including: Determine a pixel cluster in each row of pixels of the to-be-recognized text image, wherein each pixel in the pixel cluster is connected and each of the grayscale values is a preset grayscale value; Pixel clusters connected to each other in different rows are determined according to the positions of each pixel cluster to obtain a connected region; wherein, in the same connected region, there are pixel points located in the same column between pixel clusters in adjacent rows.
8. The project data fusion method based on power grid business operation management according to claim 1 is characterized in that: The converting the first keyword into a first feature vector includes: there are multiple first keywords, each element in the first feature vector corresponds to a first keyword, and according to the frequency of occurrence of each first keyword in the target text multiplied by a preset weight, the value of the element corresponding to each first keyword is calculated to obtain the first feature vector; The converting of the second keyword into a second feature vector includes: there are multiple second keywords, each element in the second feature vector corresponds to a second keyword, and the value of the element corresponding to each second keyword is calculated according to the frequency of occurrence of each second keyword in the target image multiplied by a preset weight to obtain the second feature vector.
9. The project data fusion method based on power grid business operation management according to claim 8 is characterized in that: The fusing the first feature vector and the second feature vector to obtain a fused feature vector of the target item includes: According to the values of the elements at the corresponding positions of the first feature vector and the second feature vector, the values of the elements of the fused feature vector are calculated based on a preset algorithm; wherein, if an element of the first feature vector does not exist, the value of the element of the fused feature vector is determined according to the value of the corresponding element of the second feature vector; if an element of the second feature vector does not exist, the value of the element of the fused feature vector is determined according to the value of the corresponding element of the first feature vector; Generate a vector quantity element and a fusion algorithm element as two elements of the fusion feature vector; wherein the vector quantity element represents the quantity of the fusion feature vector, and the fusion algorithm element represents the preset algorithm; Generate a fused feature vector according to the values of the elements of the fused feature vector calculated by a preset algorithm, as well as the vector quantity element and the fusion algorithm element; The first feature vector, the second feature vector and the fused feature vector are associated with the project identifier and stored in the same location.
10. A project data fusion device based on power grid business operation management, characterized in that: include: A data acquisition module, used to acquire project data for power grid business operation management, wherein the project data includes multiple texts and multiple images, the project data is derived from multiple systems and platforms, and the multiple texts and the multiple images both contain project identification information; a target determination module, configured to determine a target text and a target image corresponding to a target project in the project data based on the project identification information, wherein the target project is any project in the project data; A first keyword module, used to extract a first keyword from the target text and convert the first keyword into a first feature vector; A second keyword module, used to extract a second keyword from the target image and convert the second keyword into a second feature vector; A fusion module is used to fuse the first feature vector and the second feature vector to obtain a fused feature vector of the target project, and manage the project data based on each of the fused feature vectors.
