Multi-dimensional data analysis system and multi-dimensional data analysis method

Abstract

The invention relates to the field of enterprise data analysis, and provides a multi-dimensional data analysis system and method, and the system comprises a data integration module which is used for setting a data port, and is connected with a data source in an enterprise; the multi-dimensional modeling module is used for acquiring enterprise business data according to the data source and constructing a multi-dimensional data model; the visual display module is used for carrying out visual display on the customer business according to the multi-dimensional data model; and the intelligent decision module is used for detecting the visually displayed enterprise business data through a preset operation expert model, and sending out an early warning signal when the enterprise business data deviates from a preset index. According to the invention, internal data sources can be effectively managed and integrated, and the data processing capability is improved; automatic data identification and access reduce manual intervention and reduce the possibility of operation errors. The business data can be deeply analyzed from multiple dimensions, and business processing is carried out by constructing a multi-dimensional data model suitable for enterprise requirements.

Description

Technical Field

[0001] The present invention relates to the field of enterprise data analysis, and in particular to a multidimensional data analysis system and a multidimensional data analysis method. Background Art

[0002] At present, the existing group information management system usually involves the docking of various business ports, the collection of multi-source data and data sharing operations. The multi-source data involved in each business port will be distributed in different business systems and different types of databases, and data processing will be performed after collection. Since the data volume of most enterprises does not reach the level of big data, it will be a great waste to spend a lot of money to establish a big data system. Information data involves the operational data of the enterprise's business, and may also be the control sensing data of the production line or transportation system, and the data volume is large.

[0003] As the amount of enterprise data continues to grow and business complexity increases, existing technologies have introduced big data analysis and AI data processing to process multidimensional data. However, both big data analysis and AI data processing rely on high-quality data. If the data is erroneous, missing, or inconsistent, the analysis results may be seriously affected. AI data processing also requires a large amount of accurate data to train the model, otherwise the model may learn the wrong pattern. As data is collected and analyzed, privacy and security issues become more prominent. Enterprises need to ensure compliance with data protection regulations, such as GDPR, and take appropriate security measures to protect sensitive data. The processes of AI models and big data analysis are often opaque and difficult to explain. This may make it difficult for decision makers to understand the decision basis of the model, especially when it comes to high-risk decisions.

[0004] Therefore, how to use a new technology to achieve transparent analysis of enterprise data and visually warn of abnormalities is a technical direction that needs to be explored. Summary of the invention

[0005] The present invention proposes a multidimensional data analysis system and a multidimensional data analysis method, which are used to solve the problems in the above-mentioned background technology.

[0006] In a first aspect, the present invention provides a multidimensional data analysis system, comprising:

[0007] Data integration module: used to set up data ports and connect to data sources within the enterprise;

[0008] Multidimensional modeling module: used to obtain enterprise business data and build multidimensional data models based on data sources;

[0009] Visualization display module: used to visualize customer business based on multidimensional data models;

[0010] Intelligent decision-making module: It is used to detect the visualized enterprise business data through the preset operation expert model, and issue a warning signal when the enterprise business data deviates from the preset indicators.

[0011] In combination with the first aspect, the data port includes a data extraction port, a data conversion port and a data loading port;

[0012] The data extraction port is used to determine the corresponding data extraction task and the data extraction scheme when receiving the first data task;

[0013] The conversion port is used to determine the corresponding data conversion task and the data conversion scheme when receiving the first data task;

[0014] The data loading port is used to determine the corresponding data loading area and the data loading scheme when receiving the first data task.

[0015] In combination with the first aspect, determining the data extraction scheme includes:

[0016] Determine the data scenario and data characteristics of the target data according to the data extraction task;

[0017] According to the data scenario, determine the extraction data source of the target data and the data link where the target data is located;

[0018] According to the data characteristics, determine the data type of the target data, and determine the data volume of the target data in the data source;

[0019] According to the data volume and data links, set data extraction rules to obtain the target data from the data source.

[0020] In combination with the first aspect, determining the data conversion scheme includes:

[0021] According to the data conversion task, determine the conversion format and conversion rules of the target data;

[0022] A multi-dimensional detection mechanism is set up to detect the target data through data detection instructions, determine the conversion rules of the target data, and determine the data addresses of the target data corresponding to different conversion rules;

[0023] According to the data address, the conversion order of the target data and the first data key value and the target key value corresponding to different target data are set;

[0024] According to the first data key value and the data conversion sequence, a format switching mechanism is set to perform format conversion on the target data.

[0025] In combination with the first aspect, determining the data loading scheme includes:

[0026] Determine the loadable data items according to the data loading area;

[0027] According to the data extraction port, determine the target data volume in the data source;

[0028] Determine data loading rules based on the loadable data items and the target data volume in the data source; wherein the data loading rules include: data compression rules, data deletion rules and data necessity screening rules;

[0029] According to the data loading rules, set the forward thread group and the reverse thread group;

[0030] Using the positive-sequence thread group, compress and delete the target data in the data source to obtain the first loaded data;

[0031] The first loaded data is screened for necessity using the reverse thread, and the screened data is used as the target data.

[0032] In combination with the first aspect, constructing the multidimensional data model includes:

[0033] Divide the enterprise business data into multiple dimensions to generate first business data, second business data and third business data; wherein the first business data is used to characterize the enterprise business data describing the business content, the second business data is used to characterize the enterprise business data describing the business process, and the third business data is used to characterize the enterprise business data describing the business type;

[0034] Processing the first business data, the second business data, and the third business data to determine multi-dimensional characterization information, where the multi-dimensional characterization information is used to characterize business content characteristics, business process characteristics, and business type characteristics;

[0035] Determine the business nodes in the enterprise business framework based on multi-dimensional representation information;

[0036] By integrating the business nodes with the preset data classification and recognition network, a multidimensional data model is generated.

[0037] In combination with the first aspect, the visual display model includes:

[0038] Multidimensional data classification unit: used to divide data dimensions and classify enterprise business data according to the multidimensional data model;

[0039] Multidimensional data visualization configuration unit: used to obtain classification results and match corresponding visualization display components; wherein the visualization component is used to call the corresponding visualization display tool;

[0040] Business display sorting unit: used to determine the priority sequence of visualization display components and the visualization display order of enterprise business data when receiving a visualization display request;

[0041] Visual display unit: used to set a first display interface and a second display interface according to a visual display sequence; wherein the first display interface is used to perform sequential display according to the visual display sequence, and the second display interface is used to determine the corresponding enterprise business data in the visual display sequence when receiving a display instruction for any enterprise business data, and display it through the second display interface.

[0042] In combination with the first aspect, the intelligent decision-making model includes:

[0043] Detection unit: performs real-time detection on the visualized enterprise business data by running an expert model to determine whether there are data anomalies in the enterprise business data;

[0044] Deviation calculation unit: used to extract business parameters of enterprise business data when data anomalies exist, and perform deviation calculation based on business parameters and target business parameters to determine deviation values;

[0045] Early warning unit: used to match the preset early warning prompt channel according to the deviation value and upload it to the early warning terminal.

[0046] In combination with the first aspect, the operation expert model includes the following construction steps:

[0047] Obtain enterprise business information and set business indicators for different enterprise business information;

[0048] According to the business indicators, build corresponding visualized business numbers and visualized business attributes;

[0049] A first operation expert is set according to the visualized business number; wherein the first operation expert is used to check the business process and business progress according to the visualized business number;

[0050] According to the visualized business attributes, a second operation expert is set; wherein the second operation expert is used to visualize the business attributes and make business anomaly determinations;

[0051] The first operation expert and the second operation expert are integrated to generate an operation expert model.

[0052] In a second aspect, the present invention provides a multidimensional data analysis method, comprising:

[0053] Set up data ports to connect to data sources within the enterprise;

[0054] According to the data source, obtain enterprise business data and build a multidimensional data model;

[0055] Visualize customer business based on multidimensional data models;

[0056] The visually displayed enterprise business data is tested through the preset operation expert model, and an early warning signal is issued when the enterprise business data deviates from the preset indicators.

[0057] The beneficial effects of the present invention are:

[0058] The present invention can enable enterprises to effectively manage and integrate internal data sources through the data integration module, thereby improving the efficiency and quality of data processing; automated data identification and access reduce manual intervention and reduce the possibility of operational errors. The multidimensional modeling module enables enterprises to conduct in-depth analysis of business data from multiple dimensions, and by building a multidimensional data model that suits the needs of the enterprise, business content and patterns can be more accurately identified. The visualization module displays complex data in an intuitive way, and the intelligent decision-making module detects and warns data through expert models to discover anomalies in the business.

[0059] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0060] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0062] Figure 1 A system composition diagram of a multidimensional data analysis system in an embodiment of the present invention;

[0063] Figure 2 A method flow chart of a multidimensional data analysis method in an embodiment of the present invention;

[0064] Figure 3 It is a diagram for implementing data port settings in an embodiment of the present invention;

[0065] Figure 4 This is a data display diagram of a service node in an embodiment of the present invention;

[0066] Figure 5 A business inspection diagram in an embodiment of the present invention;

[0067] Figure 6This is a visualization diagram of business data in an embodiment of the present invention. DETAILED DESCRIPTION

[0068] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0069] like Figure 1 As shown, this embodiment provides a multidimensional data analysis system, including:

[0070] Data integration module: used to set up data ports and connect to data sources within the enterprise;

[0071] Multidimensional modeling module: used to obtain enterprise business data and build multidimensional data models based on data sources;

[0072] Visualization display module: used to visualize customer business based on multidimensional data models;

[0073] Intelligent decision-making module: It is used to detect the visualized enterprise business data through the preset operation expert model, and issue a warning signal when the enterprise business data deviates from the preset indicators.

[0074] The principle of the above technical solution is:

[0075] In the actual implementation process, the present invention includes a data integration module, a multidimensional modeling module, a visualization display module and an intelligent decision-making model.

[0076] The data integration module is mainly used to integrate and manage data sources within the enterprise. The data integration module includes but is not limited to: data port settings, that is, the system administrator sets the appropriate data port to connect various data sources within the enterprise (such as ERP system, CRM system, etc.) to the system to ensure that all data can be smoothly imported and integrated. Identification and access of data sources, including data uploaded by employees and data from various devices: After completing the data port settings, available data sources will be automatically identified and accessed. After the access is completed, all data will be synchronized to the data integration module in real time. As shown in the attached Figure 3As shown in the figure, in the subway scenario, the system administrator (or the operation and maintenance personnel) can perceive the electrical operation status of the primary equipment of the subway substation through the primary main wiring diagram (main data channel). The primary main wiring diagram displays the circuit breakers, disconnectors, automatic device status, and 1500V DC circuit breaker trolley position of each voltage level; it displays the main telemetry of each voltage level (busbar voltage, current of each incoming and outgoing line, transformer group load rate, temperature, etc.). After selecting a device on the primary wiring diagram, you can view the detailed information and inspection results of the device (there are access ports for different devices on the primary wiring diagram), and quickly view the video screen of the device through video linkage. The interface displays the primary main wiring diagram of the target substation. The equipment in the drawing is colored according to the voltage level and equipment status, that is, different devices have different colors when transmitting through different data ports, and the measurement values ​​are colored according to the quality code.

[0077] The multidimensional modeling module accesses the data source and constructs a multidimensional data model based on the data analysis and decision-making needs at different levels of the enterprise. In the specific implementation process, it can extract and clean business data: extract the business data that needs to be analyzed from the data integration module, and clean the data to remove invalid data and duplicate records. After completing the data cleaning, the multidimensional data model uses data mining, ETL and other methods to build a multidimensional data model suitable for the enterprise based on business characteristics and analysis purposes. Observe and understand business data from different dimensions (such as time, region, user, etc.) to achieve multidimensional data analysis.

[0078] The visualization module is used to convert complex data into intuitive visualization forms according to the multidimensional data model. In actual implementation, it includes: generating data visualization, displaying the data generated according to the multidimensional data model in the form of charts, dashboards, etc.; Figure 4 As shown, click on the data information details to view the detailed information of the current device, including real-time data, historical trends, and historical alarms. In addition, click on real-time data to view the measurement values ​​associated with the current device, including telemetry, telesignaling, power, and protection. Data interaction and slicing: In the process of data visualization, there are a variety of data interaction and slicing tools, which can be used to refine the data by clicking, dragging, etc.

[0079] The intelligent decision-making module detects and warns the visualized enterprise business data through the preset operational expert model. In the specific implementation process, it includes: training and learning of expert models, dividing the historical data set, one part is used for training expert models, and the other part is used for model learning. In this way, the model is continuously optimized and improved to make it more suitable for the ever-changing business environment. Detection and warning of data anomalies. When the enterprise business data is abnormal, the intelligent decision-making module will immediately issue a warning signal to remind decision makers to take corresponding measures. At the same time, it automatically tracks and monitors data changes to make the warning more real-time.

[0080] The beneficial effects of the above technical solution are:

[0081] Through the data integration module, the present invention enables enterprises to effectively manage and integrate internal data sources, thereby improving the efficiency and quality of data processing; automated data identification and access reduce manual intervention and the possibility of operational errors. The multidimensional modeling module enables enterprises to conduct in-depth analysis of business data from multiple dimensions, and by building a multidimensional data model that suits the needs of the enterprise, business content and patterns can be more accurately identified. The visualization module displays complex data in an intuitive way, and the intelligent decision-making module detects and warns data through expert models to discover anomalies in the business.

[0082] As an embodiment of the present invention, the data port includes a data extraction port, a data conversion port and a data loading port;

[0083] The data extraction port is used to determine the corresponding data extraction task and the data extraction scheme when receiving the first data task;

[0084] The conversion port is used to determine the corresponding data conversion task and the data conversion scheme when receiving the first data task;

[0085] The data loading port is used to determine the corresponding data loading area and the data loading scheme when receiving the first data task.

[0086] The principle of the above technical solution is:

[0087] In specific implementation, the function of the data extraction port of the present invention is to first analyze the task when receiving the first data task, determine the content and scope of the data to be extracted, and then select a suitable data extraction scheme according to the specific requirements of the task, and transmit the extracted data to the next port.

[0088] The function of the data conversion port is to analyze the data in the task when receiving the first data task, determine the data type and format that needs to be converted, and then select the appropriate data conversion solution according to the requirements of the task, and return the converted data to the user or store it.

[0089] The function of the data loading port is to perform final analysis and processing on the data in the task when the first data task is received, determine the data module and location that need to be loaded, and then select the appropriate data loading solution according to the requirements of the task to load the processed data module into the data system.

[0090] The beneficial effects of the above technical solution are:

[0091] The present invention sets three data ports. When receiving a data task, a data extraction plan, a data conversion plan, and a data loading plan can be formulated according to the data sequence. By processing and loading data in the above sequence, it can be ensured that the original data is not affected when the protective gear is collected, and the efficiency, accuracy and types of data conversion are richer. When loading data, there is also an exclusive data loading area, and data loading failure or abnormality will not occur.

[0092] As an embodiment of the present invention, the determining of the data extraction scheme includes:

[0093] Determine the data scenario and data characteristics of the target data according to the data extraction task;

[0094] According to the data scenario, determine the extraction data source of the target data and the data link where the target data is located;

[0095] According to the data characteristics, determine the data type of the target data, and determine the data volume of the target data in the data source;

[0096] According to the data volume and data links, set data extraction rules to obtain the target data from the data source.

[0097] The principle of the above technical solution is:

[0098] In the specific implementation process, the data scenario and data features of the target data are determined according to the data extraction task. This is to select the appropriate extraction scheme and data source. For example, according to the extraction task, the subway scenario is determined, and the data features also correspond to the data features of the subway. The identification of data scenarios and data features needs to be combined with business logic, domain knowledge, and technical constraints. Figure 5As shown in the figure, in actual operation, according to business tasks, such as patrol tasks, patrol around the substation layout, assist operation and maintenance personnel to understand the operating status of subway power supply equipment, check the equipment layout in the station, and judge whether the equipment is abnormal by combining alarm records, protection status, and image recognition status. The patrol results can be viewed through the layout diagram. The camera can be used to retrieve and play video information, and support the control of camera angle, focus, and lighting.

[0099] According to the data scenario, determine the extraction data source of the target data and the data link where the target data is located. In the specific implementation process, it is necessary to determine the actual application scenario of the data task, such as data source, business process, system architecture, etc., and determine the target data to be extracted and the related data links based on this information. The selection of data sources and data links is determined according to business processes and business nodes.

[0100] According to the data characteristics, determine the data type of the target data, and determine the data volume of the target data in the data source. In the specific implementation process, it is necessary to conduct in-depth analysis of the attributes, format, structure, etc. of the target data, and determine the appropriate data type and data volume in combination with the requirements of the data extraction task. The determination of the data type is usually based on a combination of factors such as data characteristics, data quality, and data efficiency.

[0101] According to the data volume and data links, set data extraction rules to obtain the target data from the data source. In the specific implementation process, it is necessary to formulate appropriate data extraction rules and strategies according to the data source and data links to call the target data from the data source. Data extraction rules are determined based on business needs, data scale, data frequency, data nodes and other factors.

[0102] The beneficial effects of the above technical solution are:

[0103] In the process of data extraction, the present invention can locate data according to data scene and data characteristics, determine target data and data volume of the area, and extract target data, so that the extracted data is more accurate and will not cause data redundancy and data loss.

[0104] As an embodiment of the present invention, determining the data conversion scheme includes:

[0105] According to the data conversion task, determine the conversion format and conversion rules of the target data;

[0106] A multi-dimensional detection mechanism is set up to detect the target data through data detection instructions, determine the conversion rules of the target data, and determine the data addresses of the target data corresponding to different conversion rules;

[0107] According to the data address, the conversion order of the target data and the first data key value and the target key value corresponding to different target data are set;

[0108] According to the first data key value and the data conversion sequence, a format switching mechanism is set to perform format conversion on the target data.

[0109] The principle of the above technical solution is:

[0110] In the specific implementation, according to the data conversion task, determine the conversion format and conversion rules of the target data. By clarifying the goals and requirements of data conversion, such as which data format needs to be converted, or which specific conversion rules to follow. Conversion formats include but are not limited to table formats, image formats, and text formats. Conversion rules include but are not limited to text format conversion rules, image format conversion rules, audio and video format conversion rules, data structure conversion rules (for example: conversion between relational databases and non-relational databases; data serialization and deserialization), data type conversion (numeric type conversion or character encoding conversion), data integration conversion rules, and data cleaning conversion rules.

[0111] A multi-dimensional detection mechanism is set up to detect the target data through data detection instructions, determine the conversion rules of the target data, and determine the data addresses of the target data corresponding to different conversion rules. In the specific implementation process, a multi-dimensional detection mechanism will be used to check and verify the data to ensure the integrity and consistency of the data. The multi-dimensional detection mechanism includes a business anomaly detection mechanism, a business access item detection mechanism, a business encryption and decryption detection mechanism, and a power failure detection and response mechanism. At the same time, the corresponding data address of the target data will be set according to different conversion rules.

[0112] According to the data address, the conversion order of the target data and the first data key value and the target key value corresponding to different target data are set. In the specific implementation process, by establishing the association between the data, in the process of data conversion, the correct conversion can be performed according to the data address and other metadata information.

[0113] According to the first data key value and the data conversion sequence, a format switching mechanism is set to perform format conversion on the target data. In the specific implementation process, the target data is converted from the original format to the target required format by setting a format switching mechanism, which is automated and called according to the data conversion task.

[0114] The beneficial effects of the above technical solution are:

[0115] Regarding data format conversion, the present invention can set data conversion formats and rules according to specific conversion tasks. During the data conversion process, data key values ​​are used to detect data and determine the order of data conversion, and a rapid data switching mechanism is set to improve the efficiency of data format conversion.

[0116] As an embodiment of the present invention, determining the data loading scheme includes:

[0117] Determine the loadable data items according to the data loading area;

[0118] According to the data extraction port, determine the target data volume in the data source;

[0119] Determine data loading rules based on the loadable data items and the target data volume in the data source; wherein the data loading rules include: data compression rules, data deletion rules and data necessity screening rules;

[0120] According to the data loading rules, set the forward thread group and the reverse thread group;

[0121] Using the positive-sequence thread group, compress and delete the target data in the data source to obtain the first loaded data;

[0122] The first loaded data is screened for necessity using the reverse thread, and the screened data is used as the target data.

[0123] The principle of the above technical solution is:

[0124] In the specific implementation, the loadable data items are determined according to the data loading area. In the specific implementation process, the data items that should be loaded in this area are determined according to the scope and rules of the data loading area. In the actual implementation, it is necessary to combine the results of data preprocessing and cleaning, as well as the definition of the data model.

[0125] According to the data extraction port, determine the target data volume in the data source. In the specific implementation process, the number and size of data items to be loaded are determined according to the target data volume in the data source provided by the data extraction port. During implementation, factors such as the timeliness and capacity of the data need to be considered.

[0126] Determine the data loading rules based on the loadable data items and the target data volume in the data source. In the specific implementation process, determine the appropriate data loading rules based on the number of loadable data items and the target data volume in the data source, as well as the business logic and model definition of the data. The specific implementation process includes data compression, deletion, deduplication, filtering and other operations.

[0127] According to the data loading rules, set the forward thread group and the reverse thread group. In the specific implementation process, according to the data loading rules, set the forward thread group and the reverse thread group to be executed. The forward thread group may be used to load data, while the reverse thread group may be used to filter data.

[0128] The target data in the data source is compressed and deleted by using the forward thread group to obtain the first loaded data; the first loaded data is screened for necessity by using the reverse thread, and the screened data is used as the target data. In the specific implementation process, the accuracy of the target data acquisition can be ensured by performing the necessity screening on the first loaded data by using the reverse thread.

[0129] The beneficial effects of the above technical solution are:

[0130] During the data loading process, the present invention pre-sets data loading rules, processes the loaded data, compresses and deletes the data based on the forward thread group, and screens the data for necessity based on the reverse order to determine the target data.

[0131] As an embodiment of the present invention, the building of a multidimensional data model includes:

[0132] Divide the enterprise business data into multiple dimensions to generate first business data, second business data and third business data; wherein the first business data is used to characterize the enterprise business data describing the business content, the second business data is used to characterize the enterprise business data describing the business process, and the third business data is used to characterize the enterprise business data describing the business type;

[0133] Processing the first business data, the second business data, and the third business data to determine multi-dimensional characterization information, where the multi-dimensional characterization information is used to characterize business content characteristics, business process characteristics, and business type characteristics;

[0134] Determine the business nodes in the enterprise business framework based on multi-dimensional representation information;

[0135] By integrating the business nodes with the preset data classification and recognition network, a multidimensional data model is generated.

[0136] The principle of the above technical solution is:

[0137] In the specific implementation process, the enterprise business data is divided into multiple dimensions to generate the first business data, the second business data and the third business data, which is to split and classify the business data of the enterprise according to certain dimensions to form business data of different dimensions. Among them, the first business data is used to describe the content of the business, the second business data is used to describe the process of the business, and the third business data is used to describe the form of the business. The first business data, the second business data and the third business data are processed to determine the multidimensional characterization information, which is to process the generated business data, determine the multidimensional characteristics of the business data, and perform fine-grained division of the business data. The multidimensional characterization information is a comprehensive presentation of the business content characteristics, business process characteristics and business type characteristics. According to the multidimensional characterization information, the business nodes in the enterprise business framework are determined, which is to model the enterprise business according to the multidimensional characterization information, divide it into independent business nodes, and collect and identify business data through the business nodes. The business nodes are integrated with the preset data classification and recognition network to generate a multidimensional data model, which is to integrate the generated business nodes and the data classification and recognition network to form a complete multidimensional data model, thereby performing multidimensional transformation of business data.

[0138] The beneficial effects of the above technical solution are:

[0139] The present invention processes data in multiple dimensions and can describe business content, business processes and business types through multi-dimensional division of data. The triple division makes data analysis and identification more accurate.

[0140] As an embodiment of the present invention, the visual display model includes:

[0141] Multidimensional data classification unit: used to divide data dimensions and classify enterprise business data according to the multidimensional data model;

[0142] Multidimensional data visualization configuration unit: used to obtain classification results and match corresponding visualization display components; wherein the visualization component is used to call the corresponding visualization display tool;

[0143] Business display sorting unit: used to determine the priority sequence of visualization display components and the visualization display order of enterprise business data when receiving a visualization display request;

[0144] Visual display unit: used to set a first display interface and a second display interface according to a visual display sequence; wherein the first display interface is used to perform sequential display according to the visual display sequence, and the second display interface is used to determine the corresponding enterprise business data in the visual display sequence when receiving a display instruction for any enterprise business data, and display it through the second display interface.

[0145] The principle of the above technical solution is:

[0146] In the specific implementation process, the multidimensional data classification unit will divide and classify the data according to the multidimensional data model, and will also classify the enterprise business data. The multidimensional data visualization configuration unit is responsible for obtaining the results of the multidimensional data classification and matching the corresponding visualization display components. These visualization components can usually call the corresponding visualization tools. When the business display sorting unit receives the visualization display request, it will determine the priority of the visualization display component and determine the visualization display order of the enterprise business data. The visualization display unit will set the first display interface and the second display interface according to the priority sequence provided by the business display sorting unit. Among them, the first display interface will be displayed in sequence according to the priority sequence, and the second display interface will find the corresponding enterprise business data in the visualization display sequence according to the received instructions, and display it through the second display interface.

[0147] The beneficial effects of the above technical solution are:

[0148] In the process of data display, the present invention can set corresponding visualization display components according to data types through multi-dimensional classification of data, set professional visualization display sequence, and perform visualization display. In terms of display process and display effect, it is easier to read.

[0149] As an embodiment of the present invention, the intelligent decision model includes:

[0150] Detection unit: performs real-time detection on the visualized enterprise business data by running an expert model to determine whether there are data anomalies in the enterprise business data;

[0151] Deviation calculation unit: used to extract business parameters of enterprise business data when data anomalies exist, and perform deviation calculation based on business parameters and target business parameters to determine deviation values;

[0152] Early warning unit: used to match the preset early warning prompt channel according to the deviation value and upload it to the early warning terminal.

[0153] The principle of the above technical solution is:

[0154] In the specific implementation process, the detection unit will analyze the enterprise business data obtained from the visual display and detect it in real time by running the expert model to determine whether there is any abnormal data. If data anomalies are detected, appropriate measures can be taken, such as issuing alarms or taking corrective measures. After discovering data anomalies, the deviation calculation unit will extract relevant business parameters and calculate the deviation value based on these parameters and the target business parameters. This deviation value can be any measure that can reflect data deviation, such as mean, standard deviation, etc. The early warning unit will match the preset early warning prompt channel according to the calculated deviation value and upload it to the early warning terminal. In this way, when data deviation occurs, an early warning can be automatically triggered to remind system operation and maintenance or administrators to pay attention and take action.

[0155] The beneficial effects of the above technical solution are:

[0156] The enterprise business of the present invention can perform real-time detection during operation to determine whether there is an abnormality. If there is an abnormality, the deviation value can be calculated to achieve accurate calculation of business deviation to prevent errors. When an error occurs, real-time early warning can also be achieved.

[0157] As an embodiment of the present invention, the operation expert model includes the following construction steps:

[0158] Obtain enterprise business information and set business indicators for different enterprise business information;

[0159] According to the business indicators, build corresponding visualized business numbers and visualized business attributes;

[0160] A first operation expert is set according to the visualized business number; wherein the first operation expert is used to check the business process and business progress according to the visualized business number;

[0161] According to the visualized business attributes, a second operation expert is set; wherein the second operation expert is used to visualize the business attributes and make business anomaly determinations;

[0162] The first operation expert and the second operation expert are integrated to generate an operation expert model.

[0163] The principle of the above technical solution is:

[0164] In the specific implementation process, the data port includes the data extraction port, the data conversion port and the data loading port. The function of the data port is to input various data into the system and process them. The data extraction port is responsible for extracting data from various data sources; the data conversion port is responsible for converting the data into the required format; the data loading port is responsible for loading the processed data into the system for data analysis.

[0165] The intelligent decision-making model consists of a detection unit, a deviation calculation unit, and an early warning unit. The three units are responsible for real-time monitoring and analysis of the data input into the system, timely detection of problems and responses. The operation expert model is composed of multiple operation experts. Each operation expert will model based on specific business information to conduct business process and progress inspections. The first operation expert will conduct business process and progress inspections based on the visualized business number; the second operation expert will determine business anomalies based on the visualized business attributes. Finally, the two operation experts are integrated to form a complete operation expert model.

[0166] The beneficial effects of the above technical solution are:

[0167] The intelligent local detection model of the present invention performs real-time calculations on the data, and the expert operation model performs abnormal audit and judgment on the business data through corresponding expert operation rules, thereby achieving complete processing.

[0168] In a second aspect, the present invention provides a multidimensional data analysis method, comprising:

[0169] Set up data ports to connect to data sources within the enterprise;

[0170] According to the data source, obtain enterprise business data and build a multidimensional data model;

[0171] Visualize customer business based on multidimensional data models;

[0172] The visually displayed enterprise business data is tested through the preset operation expert model, and an early warning signal is issued when the enterprise business data deviates from the preset indicators.

[0173] The principle of the above technical solution is:

[0174] like Figure 2 As shown, in the actual implementation process, the present invention includes the following steps

[0175] Integrate and manage the data sources within the enterprise. The data integration module includes but is not limited to: data port settings, that is, the system administrator sets the appropriate data port to connect various data sources within the enterprise (such as ERP system, CRM system, etc.) to the system to ensure that all data can be smoothly imported and integrated. Identification and access of data sources, including data uploaded by employees and data from various devices: After completing the data port settings, the available data sources will be automatically identified and accessed. After the access is completed, all data will be synchronized to the data integration module in real time. As shown in the attached Figure 3As shown in the figure, in the subway scenario, the system administrator (or the operation and maintenance personnel) can perceive the electrical operation status of the primary equipment of the subway substation through the primary main wiring diagram (main data channel). The primary main wiring diagram displays the circuit breakers, disconnectors, automatic device status, and 1500V DC circuit breaker trolley position of each voltage level; it displays the main telemetry of each voltage level (busbar voltage, current of each incoming and outgoing line, transformer group load rate, temperature, etc.). After selecting a device on the primary wiring diagram, you can view the detailed information and inspection results of the device (there are access ports for different devices on the primary wiring diagram), and quickly view the video screen of the device through video linkage. The interface displays the primary main wiring diagram of the target substation. The equipment in the drawing is colored according to the voltage level and equipment status, that is, different devices have different colors when transmitting through different data ports, and the measurement values ​​are colored according to the quality code.

[0176] Access the data source and build a multidimensional data model according to the data analysis and decision-making needs at different levels of the enterprise. In the specific implementation process, business data can be extracted and cleaned: business data that needs to be analyzed is extracted from the data integration module, and these data are cleaned to remove invalid data and duplicate records. After the multidimensional data model completes the data cleaning, it uses data mining, ETL and other methods to build a multidimensional data model suitable for the enterprise according to business characteristics and analysis purposes. Observe and understand business data from different dimensions (such as time, region, user, etc.) to achieve multidimensional data analysis.

[0177] According to the multidimensional data model, complex data is converted into an intuitive visual expression. In actual implementation, it includes: generating data visualization, displaying the data generated by the multidimensional data model in the form of charts, dashboards, etc.; as shown in the attached Figure 4 As shown, click on the data information details to view the detailed information of the current device, including real-time data, historical trends, and historical alarms. In addition, click on real-time data to view the measurement values ​​associated with the current device, including telemetry, telesignaling, power, and protection. Data interaction and slicing: In the process of data visualization, there are a variety of data interaction and slicing tools, which can be used to refine the data by clicking, dragging, etc.

[0178] The intelligent decision-making module detects and warns the visualized enterprise business data through the preset operational expert model. In the specific implementation process, it includes: training and learning of expert models, dividing the historical data set, one part is used for training expert models, and the other part is used for model learning. In this way, the model is continuously optimized and improved to make it more suitable for the ever-changing business environment. Detection and warning of data anomalies. When the enterprise business data is abnormal, the intelligent decision-making module will immediately issue a warning signal to remind decision makers to take corresponding measures. At the same time, it automatically tracks and monitors data changes to make the warning more real-time.

[0179] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A multidimensional data analysis system, characterized in that: include: Data integration module: used to set up data ports and connect to data sources within the enterprise; Multidimensional modeling module: used to obtain enterprise business data and build multidimensional data models based on data sources; Visualization display module: used to visualize customer business based on multidimensional data models; Intelligent decision-making module: It is used to detect the visualized enterprise business data through the preset operation expert model, and issue a warning signal when the enterprise business data deviates from the preset indicators.

2. A multidimensional data analysis system as claimed in claim 1, characterized in that: The data ports include a data extraction port, a data conversion port and a data loading port; The data extraction port is used to determine the corresponding data extraction task and the data extraction scheme when receiving the first data task; The conversion port is used to determine the corresponding data conversion task and the data conversion scheme when receiving the first data task; The data loading port is used to determine the corresponding data loading area and the data loading scheme when receiving the first data task.

3. A multidimensional data analysis system as claimed in claim 1, characterized in that: The data extraction scheme is determined to include: Determine the data scenario and data characteristics of the target data according to the data extraction task; According to the data scenario, determine the extraction data source of the target data and the data link where the target data is located; According to the data characteristics, determine the data type of the target data, and determine the data volume of the target data in the data source; According to the data volume and data links, set data extraction rules to obtain the target data from the data source.

4. A multidimensional data analysis system as claimed in claim 1, characterized in that: Determining the data conversion scheme includes: According to the data conversion task, determine the conversion format and conversion rules of the target data; A multi-dimensional detection mechanism is set up to detect the target data through data detection instructions, determine the conversion rules of the target data, and determine the data addresses of the target data corresponding to different conversion rules; According to the data address, the conversion order of the target data and the first data key value and the target key value corresponding to different target data are set; According to the first data key value and the data conversion sequence, a format switching mechanism is set to perform format conversion on the target data.

5. A multidimensional data analysis system as claimed in claim 1, characterized in that: Determining the data loading scheme includes: Determine the loadable data items according to the data loading area; According to the data extraction port, determine the target data volume in the data source; Determine data loading rules based on the loadable data items and the target data volume in the data source; wherein the data loading rules include: data compression rules, data deletion rules and data necessity screening rules; According to the data loading rules, set the forward thread group and the reverse thread group; Using the positive-sequence thread group, compress and delete the target data in the data source to obtain the first loaded data; The first loaded data is screened for necessity using the reverse thread, and the screened data is used as the target data.

6. A multidimensional data analysis system as claimed in claim 1, characterized in that: The constructing of the multidimensional data model comprises: Divide the enterprise business data into multiple dimensions to generate first business data, second business data and third business data; wherein the first business data is used to characterize the enterprise business data describing the business content, the second business data is used to characterize the enterprise business data describing the business process, and the third business data is used to characterize the enterprise business data describing the business type; Processing the first business data, the second business data, and the third business data to determine multi-dimensional characterization information, where the multi-dimensional characterization information is used to characterize business content characteristics, business process characteristics, and business type characteristics; Determine the business nodes in the enterprise business framework based on multi-dimensional representation information; By integrating the business nodes with the preset data classification and recognition network, a multidimensional data model is generated.

7. A multidimensional data analysis system as claimed in claim 1, characterized in that: The visual display model includes: Multidimensional data classification unit: used to divide data dimensions and classify enterprise business data according to the multidimensional data model; Multidimensional data visualization configuration unit: used to obtain classification results and match corresponding visualization display components; wherein the visualization component is used to call the corresponding visualization display tool; Business display sorting unit: used to determine the priority sequence of visualization display components and the visualization display order of enterprise business data when receiving a visualization display request; Visual display unit: used to set a first display interface and a second display interface according to a visual display sequence; wherein the first display interface is used to perform sequential display according to the visual display sequence, and the second display interface is used to determine the corresponding enterprise business data in the visual display sequence when receiving a display instruction for any enterprise business data, and display it through the second display interface.

8. A multidimensional data analysis system as claimed in claim 1, characterized in that: The intelligent decision-making model includes: Detection unit: performs real-time detection on the visualized enterprise business data by running an expert model to determine whether there are data anomalies in the enterprise business data; Deviation calculation unit: used to extract business parameters of enterprise business data when data anomalies exist, and perform deviation calculation based on business parameters and target business parameters to determine deviation values; Early warning unit: used to match the preset early warning prompt channel according to the deviation value and upload it to the early warning terminal.

9. A multidimensional data analysis system as claimed in claim 1, characterized in that: The operation expert model includes the following construction steps: Obtain enterprise business information and set business indicators for different enterprise business information; According to the business indicators, build corresponding visualized business numbers and visualized business attributes; A first operation expert is set according to the visualized business number; wherein the first operation expert is used to check the business process and business progress according to the visualized business number; According to the visualized business attributes, a second operation expert is set; wherein the second operation expert is used to visualize the business attributes and make business anomaly determinations; The first operation expert and the second operation expert are integrated to generate an operation expert model.

10. A multidimensional data analysis method, characterized in that: include: Set up data ports to connect to data sources within the enterprise; According to the data source, obtain enterprise business data and build a multidimensional data model; Visualize customer business based on multidimensional data models; The visually displayed enterprise business data is tested through the preset operation expert model, and an early warning signal is issued when the enterprise business data deviates from the preset indicators.

Images