Enterprise data visual interaction management system fusing digital twinning and BI board

Abstract

The invention belongs to the technical field of enterprise data management, and particularly relates to an enterprise data visual interaction management system fusing digital twinning and a BI board. Comprising an enterprise multi-source heterogeneous data adaptation access module, a data preprocessing output module, an enterprise entity twin dynamic construction module, a BI visual dimension data aggregation module, an enterprise operation anomaly multi-feature fusion recognition module and a fusion visual interaction presentation module. Influences of data chaos and deviation on analysis results are avoided from the source through an enterprise multi-source heterogeneous data adaptation access module, real-time synchronization of enterprise operation states and early warning of equipment faults are realized based on dynamic digital twinners, and a BI visual dimension data aggregation module meets requirements of users of different levels and reduces docking cost. And by distinguishing positive and negative features and calculating a directional anomaly deviation rate and a comprehensive anomaly degree, misjudgment caused by traditional single features or non-distinguishing directions is avoided, and enterprises are assisted to respond to risks in time and carry out targeted rectification.

Description

Technical Field

[0001] This invention relates to the field of enterprise data management technology, specifically to an enterprise data visualization and interactive management system that integrates digital twins and BI dashboards. Background Technology

[0002] In the field of enterprise digital management, the application of digital twin technology and BI dashboard technology has become an important direction for improving operational management efficiency. The two have significant advantages in the contextual mapping of physical entities and the quantitative analysis of operational data, respectively. At present, most enterprise management solutions based on digital twins focus on single-dimensional status display or anomaly monitoring, and fail to deeply integrate with the multi-dimensional quantitative analysis of BI dashboards.

[0003] For example, Chinese invention patent with publication number CN115730114A discloses a device data visualization method based on digital twins. The invention's technical solution is to display device attribute data and operating status in real time through a digital twin model, thereby enabling the rapid presentation of abnormal content.

[0004] However, the above-mentioned invention only focuses on the status feedback of the equipment itself and does not integrate the multi-dimensional aggregation and analysis capabilities of BI technology for enterprise operation data. When the equipment malfunctions and related equipment alarms simultaneously, it is difficult for staff to judge the impact of the malfunction on the overall operation from the complex status data and operation indicators. Furthermore, it is impossible to accurately judge the enterprise's operating status by combining indicators from multiple fields, which is not conducive to enterprise data management and improving enterprise management efficiency. Summary of the Invention

[0005] The purpose of this invention is to provide an enterprise data visualization and interactive management system that integrates digital twins and BI dashboards to address the technical deficiencies mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an enterprise data visualization interactive management system integrating digital twin and BI dashboard, including an enterprise multi-source heterogeneous data adaptation and access module, a data preprocessing output module, an enterprise entity twin dynamic construction module, a BI visualization dimension data aggregation module, an enterprise operational anomaly multi-feature fusion identification module, and a fusion visualization interactive presentation module;

[0007] The enterprise multi-source heterogeneous data adaptation and access module is used for the secure adaptation and classification access of multi-source heterogeneous data inside and outside the enterprise, and the data preprocessing output module cleans, transforms and integrates the accessed raw data.

[0008] The Enterprise Entity Twin Dynamic Construction Module constructs a dynamic digital twin that maps to the enterprise's physical entity based on pre-processed standardized data, synchronizing the enterprise's operational status in real time. The BI Visualization Dimensional Data Aggregation Module aggregates data according to the analytical dimensions of the BI dashboard based on pre-processed standardized data, generating a structured BI visualization data source to support multi-dimensional display of the BI dashboard.

[0009] The enterprise operation anomaly multi-feature fusion identification module extracts operation features from the enterprise dynamic twin dataset, calculates the comprehensive anomaly degree through a multi-feature fusion algorithm, identifies enterprise operation anomalies, and sends the operation anomaly information to the fusion visualization interactive presentation module; the fusion visualization interactive presentation module integrates digital twin and BI dashboard data to build an integrated visualization interactive interface.

[0010] Furthermore, the enterprise multi-source heterogeneous data adaptation and access module connects to the enterprise's internal ERP, CRM, SCM and MES business systems, as well as external industry databases, market research platforms and supply chain partner data systems through preset multi-type data source interfaces.

[0011] Furthermore, it automatically identifies the data format of each data source, converts unstructured data into semi-structured data based on preset format mapping rules, and converts semi-structured data into standardized structured data. At the same time, it verifies data source access permissions and ensures data transmission security through a triple mechanism of "token verification + IP whitelist + encrypted data transmission". It also marks the adapted data into six categories: sales data, inventory data, financial data, production data, customer data, and equipment data, generates a data access classification list, and transmits the classified raw data to the data preprocessing output module in real time.

[0012] The data preprocessing output module receives the classified raw data and sequentially performs data cleaning, data transformation, and data integration operations. The integrated standardized data is then divided into two outputs: one output filters key data according to the twin modeling requirements and transmits it to the enterprise entity twin dynamic construction module; the other output extracts and summarizes indicators according to the BI dimension requirements and transmits them to the BI visualization dimension data aggregation module.

[0013] Furthermore, the enterprise entity twin dynamic construction module imports the basic model of the enterprise physical entity and associates and binds standardized data with the entity attributes in the model;

[0014] Through a mechanism of "real-time data-driven + rule engine update", the status of the twin is dynamically updated: when the production equipment data changes, the 3D model parameters of the corresponding equipment in the twin are updated synchronously; when the sales order volume changes, the "sales-inventory" linkage relationship in the twin is adjusted in real time.

[0015] At the same time, a two-way data feedback channel between the "twin and the physical entity" is constructed: if the twin detects that a certain device parameter is close to the threshold, it sends an early warning signal to the device's IoT control system to trigger a device maintenance reminder;

[0016] Generate a dynamic twin dataset for the enterprise, which includes real-time status data, historical change trajectory data, and relationship data of each entity in the twin. Transmit this dataset to the multi-feature fusion identification module for abnormal enterprise operations.

[0017] Furthermore, the BI visualization dimension data aggregation module determines the core analysis dimension system of the BI dashboard, including time dimension, business dimension, organizational dimension, and regional dimension; it aggregates data according to the "dimension-indicator" mapping rule, and encapsulates the aggregated data in a structured manner to generate a BI dimension data cube that conforms to the import format of BI tools. The cube contains four elements: "dimension label - indicator value - timestamp - data source"; and the data cube is transmitted to the integrated visualization interactive presentation module.

[0018] Furthermore, the analysis process of the enterprise operation anomaly multi-feature fusion identification module is as follows:

[0019] Extract feature indicators that are strongly correlated with operational anomalies from the enterprise dynamic twin dataset, and clarify the advantages and disadvantages of each type of feature. Among them, positive features include monthly actual sales revenue in sales features and monthly actual net cash flow in cash flow features, while negative features include monthly inventory backlog rate in inventory features, actual equipment failure rate in production features, and monthly actual customer churn rate in customer features.

[0020] Calculate the abnormal deviation rate Ri of a single feature. If there is a feature whose abnormal deviation rate Ri exceeds the corresponding preset abnormal deviation rate threshold, then the enterprise is determined to have operational abnormalities.

[0021] If there are no features whose abnormal deviation rate Ri exceeds the corresponding preset abnormal deviation rate threshold, then the importance weight wi of each feature is obtained; the comprehensive abnormality degree F is calculated, and the abnormality judgment threshold T is determined based on the statistical data of the company's operations over the past three years. If F > T, then the company is judged to have operational abnormalities; and the operational abnormality information, including the abnormality judgment result, abnormal deviation rate Ri and comprehensive abnormality degree F, is transmitted to the integrated visualization interactive presentation module.

[0022] Furthermore, the analysis and calculation process of the abnormal deviation rate Ri is as follows:

[0023] Determine the reasonable benchmark value Xio for each type of feature: Obtain the average value of the indicator for the same period of the companies in the neighboring 3 years and the average value of the indicator for the same period of the companies in the same industry. Take the arithmetic mean of the average value of the indicator for the same period of the companies in the neighboring 3 years and the average value of the indicator for the same period of the companies in the same industry as the reasonable benchmark value Xio for that type of feature.

[0024] The abnormal deviation rate is calculated based on the direction of positive and negative features: For positive features, the actual value Xi of the feature is collected, and the abnormal deviation rate Ri is calculated using the formula Ri=(Xio-Xi) / Xio×100%. When Xi≥Xio, the abnormal deviation rate Ri=0. For negative features, the abnormal deviation rate Ri is calculated using the formula Ri=(Xi-Xio) / Xio×100%. When Xi≤Xio, the abnormal deviation rate Ri=0.

[0025] Furthermore, the integrated visualization interactive presentation module communicates with the multi-level performance comprehensive evaluation module. The multi-level performance comprehensive evaluation module evaluates and analyzes the performance status of the integrated visualization interactive presentation module, and determines whether to generate a performance degradation alarm signal through analysis, and sends the performance degradation alarm signal to the smart terminal of the corresponding operation and maintenance personnel.

[0026] Furthermore, the specific analysis process of the multi-level performance evaluation module is as follows:

[0027] The monitoring period is set to duration L1. The number of times the integrated visualization interactive presentation module fails to display during the monitoring period is collected and marked as the presentation frequency value. The total duration during which the integrated visualization interactive presentation module fails to display during the monitoring period is marked as the presentation time value. If the presentation frequency value or the presentation time value exceeds the corresponding preset threshold, a performance degradation alarm signal is generated.

[0028] Furthermore, if the frequency and time values ​​do not exceed the corresponding preset thresholds, when the integrated visualization interactive presentation module displays, it collects the area ratio of the blurred area and the brightness deviation value and marks them as the blurred area ratio and brightness deviation detection value, respectively. If the blurred area ratio or brightness deviation detection value exceeds the corresponding preset threshold, it is determined that the integrated visualization interactive presentation module is in a state that needs optimization.

[0029] The total duration during which the integrated visualization interactive presentation module is in a presentation optimization state during the monitoring period is obtained and marked as the presentation optimization coefficient. If the presentation optimization coefficient exceeds the preset presentation optimization coefficient threshold, a performance degradation warning signal is generated. If the presentation optimization coefficient does not exceed the preset presentation optimization coefficient threshold, the interactive presentation decision coefficient is obtained through multi-dimensional comprehensive evaluation and analysis. If the interactive presentation decision coefficient exceeds the preset interactive presentation decision coefficient threshold, a performance degradation warning signal is generated.

[0030] Furthermore, the specific analysis process for the multi-dimensional comprehensive evaluation and analysis is as follows:

[0031] The number of times users click on the integrated visualization interactive presentation module during the monitoring period is obtained and marked as the interaction frequency. The response delay time of the corresponding click operation is marked as the click response value. The interaction feature coefficient is obtained by calculating the ratio of the number of times the click response value exceeds the preset click response threshold to the interaction frequency. The interaction presentation decision coefficient is obtained by weighted summation of the presentation frequency value, presentation time value, presentation optimization coefficient and interaction feature coefficient.

[0032] Compared with the prior art, the beneficial effects of the present invention are:

[0033] 1. In this invention, the multi-source heterogeneous data adaptation and access module avoids the impact of data chaos and bias on the analysis results from the source. Based on the dynamic digital twin, it realizes real-time synchronization of enterprise operating status and early warning of equipment failure. The BI visualization dimension data aggregation module provides a structured data source for visualization presentation. By distinguishing positive and negative features, calculating the directional anomaly deviation rate and the comprehensive anomaly degree, it helps enterprises respond to risks in a timely manner and make targeted rectifications. Furthermore, the integrated interface improves the efficiency of users' understanding of operating status, supports personalized interactive operations, and is conducive to optimizing enterprise management processes.

[0034] 2. In this invention, the multi-level performance comprehensive evaluation module monitors the operating status of the integrated visualization interactive presentation module from multiple dimensions, accurately generates alarm signals, detects potential performance problems in advance, avoids display failures or interactive lag, ensures the continuous and stable operation of the integrated visualization interactive presentation module, maintains the smoothness of users' visualization and interaction of business data, ensures the overall application effect of the system, provides enterprises with efficient, accurate and secure business data management and decision support, and helps improve operational efficiency and risk control capabilities. Attached Figure Description

[0035] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0036] Figure 1 This is a system block diagram of Embodiment 1 of the present invention;

[0037] Figure 2 This is a system block diagram of Embodiment 2 of the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Example 1: As Figure 1 As shown, the enterprise data visualization interactive management system that integrates digital twins and BI dashboards proposed in this invention includes an enterprise multi-source heterogeneous data adaptation and access module, a data preprocessing output module, an enterprise entity twin dynamic construction module, a BI visualization dimension data aggregation module, an enterprise operational anomaly multi-feature fusion identification module, and a fusion visualization interactive presentation module.

[0040] The enterprise multi-source heterogeneous data adaptation and access module enables secure adaptation and classification access of multi-source heterogeneous data within and outside the enterprise, solving the data silo problem caused by inconsistent data formats and scattered data sources, providing complete original data support for subsequent processing, and ensuring the comprehensiveness, security and orderliness of data access from the source;

[0041] Specifically, the enterprise multi-source heterogeneous data adaptation and access module first connects to internal business systems such as ERP, CRM, SCM and MES, as well as external data systems such as industry databases, market research platforms and supply chain partners through preset multi-type data source interfaces (including API interface, database direct connection interface, file import interface and IoT device data receiving interface).

[0042] Next, it automatically identifies the data format of each data source (such as JSON, XML, CSV, relational database tables, unstructured text / images), and converts unstructured data into semi-structured data based on preset format mapping rules (such as extracting key business indicators from text through NLP), and converts semi-structured data into standardized structured data. At the same time, it verifies data source access permissions and ensures data transmission security through a triple mechanism of "token verification + IP whitelist + encrypted data transmission". The triple security mechanism reduces the risk of data leakage and ensures the compliance of data collection.

[0043] Finally, the adapted data is tagged according to six categories: sales data, inventory data, financial data, production data, customer data, and equipment data, generating a data access classification list. The categorized raw data is then transmitted to the data preprocessing output module in real time. The classification and tagging provide a basis for subsequent targeted processing and avoid data confusion.

[0044] The data preprocessing output module cleans the incoming raw data (identifies and removes extreme outliers in sales and financial data using the 3σ principle, supplements missing items through "interpolation filling + business rule filling", such as filling missing values ​​in inventory data based on historical inventory change trends, and deleting duplicate data through hash comparison to ensure data accuracy), transforms (eliminates the impact of dimensional differences on subsequent modeling), and integrates (based on enterprise business entity association rules, such as associating sales data and inventory data through "product ID", associating human resources data and financial cost data through "department ID", integrating scattered classified data into an "enterprise business entity data mart", which includes three major association data chains: "product-sales-inventory", "department-human resources-cost", and "equipment-production-energy consumption"), and outputs standardized and integrated high-quality data.

[0045] Specifically, the data preprocessing output module receives the classified raw data, performs data cleaning, data transformation, and data integration operations in sequence, and finally outputs the integrated standardized data in two paths. One path filters key data (such as production equipment parameters, real-time inventory levels, and sales order volume) according to the "twin modeling requirements" and transmits it to the enterprise entity twin dynamic construction module. The other path extracts summary indicators (such as monthly sales and quarterly profit margin) according to the "BI dimension requirements" and transmits them to the BI visualization dimension data aggregation module.

[0046] The Enterprise Entity Twin Dynamic Construction Module constructs a dynamic digital twin based on pre-processed standardized data, which is mapped (1:1 mapping) to the enterprise's physical entity (including business processes, equipment, and supply chain). It synchronizes the enterprise's operating status in real time, providing the enterprise with an intuitive "digital mirror" and helping users quickly grasp the overall business situation, avoiding the limitations of anomaly identification under a single data dimension.

[0047] Specifically, the enterprise entity twin dynamic construction module first imports the basic model of the enterprise's physical entity (including organizational structure model, production process model, supply chain network model, and equipment 3D model), and then associates and binds standardized data with the entity attributes in the model (such as the "failure rate" and "capacity" of equipment, and the "order response time" of business processes).

[0048] Then, through the "real-time data-driven + rule engine update" mechanism, the status of the twin is dynamically updated: for example, when the production equipment data (such as temperature and speed) changes, the parameters of the corresponding three-dimensional model of the equipment in the twin are updated synchronously; when the sales order volume changes, the "sales-inventory" linkage relationship in the twin is adjusted in real time (such as when the order volume increases, the inventory model automatically deducts the corresponding product inventory).

[0049] At the same time, a two-way data feedback channel between the "twin and the physical entity" is constructed: if the twin detects that a certain device parameter is close to the threshold (such as a failure rate of 80%), it sends an early warning signal to the device's IoT control system to trigger a device maintenance reminder, effectively avoiding business losses caused by device failure;

[0050] Finally, a dynamic twin dataset of the enterprise is generated, which includes real-time status data, historical change trajectory data and correlation data of each entity in the twin. This dataset is transmitted to the enterprise operation anomaly multi-feature fusion identification module (for anomaly feature extraction), providing scenario-based and multi-dimensional data support for the enterprise operation anomaly multi-feature fusion identification module, avoiding the limitations of anomaly identification under a single data dimension, and improving the comprehensiveness of anomaly identification.

[0051] The BI visualization dimension data aggregation module aggregates pre-processed standardized data according to the analysis dimensions of the BI dashboard, generating a structured BI visualization data source to support multi-dimensional display of the BI dashboard. Specifically, the BI visualization dimension data aggregation module first determines the core analysis dimension system of the BI dashboard, including time dimension (daily / weekly / monthly / quarterly / yearly), business dimension (sales / inventory / production / finance / customer), organizational dimension (headquarters / branch offices / departments / product lines), and regional dimension (sales regions such as East China / North China / South China), which can cover the visualization needs of users at different levels of the enterprise (executives, department managers, business personnel) and meet diverse analysis scenarios.

[0052] Next, data aggregation is performed according to the "dimension-metric" mapping rules: time dimension aggregation (such as summarizing daily sales to generate monthly total sales), business dimension aggregation (such as summarizing inventory of each product line to generate total inventory turnover rate), and organizational dimension aggregation (such as summarizing the cost of each department to generate total cost of headquarters). During the aggregation process, the "rolling window algorithm" is used to process massive amounts of data to ensure aggregation efficiency, effectively shorten the data aggregation time, improve the data update efficiency of BI dashboard, and reduce user waiting time.

[0053] The aggregated data is then structured and encapsulated to generate a "BI Dimension Data Cube" that conforms to the import format of BI tools (such as Tableau and Power BI). The cube contains four elements: "dimensional label - indicator value - timestamp - data source". This reduces the difficulty and cost of connecting with different BI tools, while providing a structured and highly adaptable BI data source for the integrated visualization and interactive presentation module, ensuring the accuracy and smoothness of the BI dashboard display. Finally, the data cube is transmitted to the integrated visualization and interactive presentation module.

[0054] The enterprise operation anomaly multi-feature fusion identification module extracts operation features from the enterprise dynamic twin dataset, calculates the comprehensive anomaly degree through a multi-feature fusion algorithm, identifies enterprise operation anomalies, and sends the operation anomaly information to the fusion visualization interactive presentation module. By clearly distinguishing between positive and negative operation features and designing targeted anomaly deviation rate calculation logic, it avoids the misjudgment problems caused by not distinguishing the direction of feature quality or relying on a single feature in traditional anomaly identification, and significantly improves the accuracy of anomaly identification.

[0055] Furthermore, by calculating the comprehensive anomaly score based on feature importance weights and setting reasonable judgment thresholds in conjunction with the company's operating data from the past three years, it can accurately locate key anomaly features, providing companies with clear rectification directions, avoiding blind optimization, helping users to promptly grasp operational risks, and improving risk response efficiency. The analysis process of the enterprise operational anomaly multi-feature fusion identification module is as follows:

[0056] First, extract feature indicators that are strongly correlated with operational anomalies from the enterprise dynamic twin dataset, and clarify the direction of the advantages and disadvantages of each type of feature (i.e., divide them into positive features and negative features; the larger the value of the positive feature, the better the enterprise's operating condition, and the larger the value of the negative feature, the worse the enterprise's operating condition).

[0057] Among them, positive features include sales feature S1 (monthly actual sales revenue, the higher the sales revenue, the healthier the operation) and cash flow feature S2 (monthly actual net cash flow, the higher the net cash flow, the safer the finances). Negative features include inventory feature S3 (monthly inventory backlog rate, the lower the backlog rate, the more efficient the inventory turnover), production feature S4 (actual equipment failure rate, the lower the failure rate, the more stable the production), and customer feature S5 (monthly actual customer churn rate, the lower the churn rate, the better the customer retention).

[0058] Next, calculate the abnormal deviation rate Ri of a single feature: first determine the reasonable benchmark value Xio for each type of feature, obtain the average value of the same indicator of the same period of the companies in the neighboring 3 years and the average value of the same indicator of the same company in the industry, and take the arithmetic mean of the average value of the same indicator of the companies in the neighboring 3 years and the average value of the same indicator of the same company in the industry as the reasonable benchmark value Xio for that type of feature.

[0059] For example, S1's X1o = (Company's sales revenue in March last year was 5 million + industry average sales revenue in March was 4.8 million) / 2 = 4.9 million;

[0060] S2's X2o = (Company's inventory backlog rate in March last year was 8% + Industry average backlog rate in March was 7%) / 2 = 7.5%;

[0061] The abnormal deviation rate is calculated based on the superiority and inferiority directions: For positive features (S1, S2), the actual value Xi of the feature is collected, and the abnormal deviation rate Ri of the feature is calculated by the formula Ri=(Xio-Xi) / Xio×100%. When Xi≥Xio, it means that the feature performance is better than or meets the standard, and the abnormal deviation rate Ri=0, indicating no abnormality.

[0062] For example, if the actual value of S1 is 5.2 million ≥ X1o = 4.9 million, then R1 = 0; if the actual value of S1 is 4.2 million < 4.9 million, then R1 = (4.9 million - 4.2 million) / 4.9 million × 100% ≈ 14.3%.

[0063] For negative features (S3, S4, S5), the actual value Xi of the feature is collected, and the abnormal deviation rate Ri of the feature is calculated by the formula Ri=(Xi-Xio) / Xio×100%. When Xi≤Xio, it means that the feature performance is better than or meets the standard, and the abnormal deviation rate Ri=0, indicating no abnormality.

[0064] For example, if the actual value of S2 is 6% ≤ X2o = 7.5%, then R2 = 0; if the actual value of S2 is 10% > 7.5%, then R2 = (10 - 7.5) / 7.5 × 100% ≈ 33.3%;

[0065] If there is a feature where the abnormal deviation rate Ri exceeds the corresponding preset abnormal deviation rate threshold, the enterprise is determined to have operational abnormalities.

[0066] If there are no features with an abnormal deviation rate Ri exceeding the corresponding preset abnormal deviation rate threshold, then the importance weights wi of each feature are obtained (preferably, w1=0.2, w2=0.1, w3=0.3, w4=0.18, w5=0.22); and then the weights are obtained using the formula... The overall anomaly degree F is calculated;

[0067] For example: S1 actual value 4.2 million (R1=14.3%), S2 actual value 6% (R2=0), S3 actual value 800,000 (X3o=1 million, R3=(100-80) / 100×100%=20%), S4 actual value 5% (X4o=4%, R4=(5-4) / 4×100%=25%), S5 actual value 12% (X5o=10%, R5=(12-10) / 10×100%=20%).

[0068] Therefore, F = 14.3% × 0.2 + 0 × 0.1 + 20% × 0.3 + 25% × 0.18 + 20% × 0.22 = 2.86% + 0 + 6% + 4.5% + 4.4% = 17.76%;

[0069] Then, based on the statistical data of the company's operations over the past three years, the anomaly judgment threshold T is determined (when operating normally, F is usually within 10%, but when an actual anomaly occurs, F generally exceeds 15%). If F > T (T is set to 15%, such as 17.76% > 15% mentioned above), then the company is judged to have operational anomalies. The operational anomaly information, including the anomaly judgment result, the anomaly deviation rate Ri, and the comprehensive anomaly degree F, is transmitted to the integrated visualization interactive presentation module.

[0070] The integrated visualization and interactive presentation module merges digital twin and BI dashboard data to build an integrated visual interactive interface that supports users to view, operate, and provide feedback in real time. Specifically, it first receives twin model data (including 3D models and real-time status animations) from the enterprise entity twin dynamic construction module, data cubes (including indicator data for each dimension) from the BI visualization dimension data aggregation module, and analysis report data from the enterprise operation anomaly multi-feature fusion identification module.

[0071] Next, a unified display interface is constructed, divided into three main areas: the left "Twin Display Area" (displaying a 3D twin of the enterprise, supporting zooming, rotation, and clicking to view entity details, such as clicking on equipment to view failure rates, and clicking on departments to view cost data), the middle "BI Dashboard Area" (displaying line charts, bar charts, pie charts, etc. by time / business / organizational dimensions, such as monthly sales trend charts and pie charts of cost proportions for each department, supporting dimension switching and drill-down analysis), and the right "Anomaly Area" (displaying anomaly alerts in real time, i.e., red-marked anomaly features and a list of improvement suggestions);

[0072] Then, it supports user interaction: users can adjust the interface layout by clicking and dragging, and trigger the system to send data requests to the BI visualization dimension data aggregation module by entering query conditions (such as "sales data of East China region in Q2 2024"), and refresh the dashboard and evaluation results in real time; users can also enter the progress of abnormal rectification into the system through the "Abnormal Handling Feedback" button, and the system will transmit the feedback data back to the enterprise operation abnormal multi-feature fusion identification module for subsequent abnormality recalculation; finally, the interface operation log and user query records are stored in the system log library to provide a basis for subsequent function optimization.

[0073] Example 2: Figure 2 As shown, the difference between this embodiment and embodiment one is that the integrated visualization interactive presentation module is connected to the multi-level performance comprehensive evaluation module. The multi-level performance comprehensive evaluation module evaluates and analyzes the performance status of the integrated visualization interactive presentation module, and determines whether to generate a performance degradation alarm signal through analysis. The performance degradation alarm signal is then sent to the smart terminal of the corresponding operation and maintenance personnel to remind them to check and repair the integrated visualization interactive presentation module in a timely manner.

[0074] By evaluating the performance of the integrated visualization and interactive presentation module from multiple dimensions, potential operational risks can be accurately identified, performance issues can be detected early and alerts can be issued in a timely manner. This can prevent display failures or interactive lag caused by module performance degradation, ensuring the continuous and stable operation of the integrated visualization and interactive presentation module, maintaining the smoothness of users' visualization and interaction with business data, and ensuring the overall application effect of the system. The specific analysis process of the multi-level performance evaluation module is as follows:

[0075] A monitoring period of duration L1 is set. The number of times the integrated visualization interactive presentation module fails to display during the monitoring period is collected and marked as the presentation frequency value. The total duration during which the integrated visualization interactive presentation module fails to display during the monitoring period is marked as the presentation time value. The presentation frequency value and the presentation time value are compared with the preset presentation frequency threshold and the preset presentation time threshold, respectively. If the presentation frequency value or the presentation time value exceeds the corresponding preset threshold, it indicates that the operational reliability of the integrated visualization interactive presentation module is poor during the monitoring period, and a performance degradation alarm signal is generated.

[0076] Furthermore, if the frequency and time values ​​do not exceed the corresponding preset thresholds, when the integrated visualization interactive presentation module displays, it collects the area ratio of the blurred area and the deviation value of the display brightness and marks them as the blurred area ratio and the brightness deviation detection value, respectively. The blurred area ratio and the brightness deviation detection value are compared with the preset blurred area ratio threshold and the preset brightness deviation detection threshold, respectively. If the blurred area ratio or the brightness deviation detection value exceeds the corresponding preset threshold, it is determined that the integrated visualization interactive presentation module is in a state that needs optimization.

[0077] The total duration during which the integrated visualization interactive presentation module is in a presentation optimization state during the monitoring period is obtained and marked as the presentation optimization coefficient. The presentation optimization coefficient is compared with the preset presentation optimization coefficient threshold. If the presentation optimization coefficient exceeds the preset presentation optimization coefficient threshold, it indicates that the presentation effect of the integrated visualization interactive presentation module is poor during the monitoring period, and a performance degradation warning signal is generated.

[0078] If the presentation optimization coefficient does not exceed the preset presentation optimization coefficient threshold, the number of times the user clicks on the integrated visual interactive presentation module during the monitoring period is obtained and marked as the interaction frequency. The response delay time of the corresponding click operation is marked as the click response value. The larger the click response value, the less timely the response to the corresponding click operation. The interaction feature coefficient is obtained by calculating the ratio of the number of times the click response value exceeds the preset click response threshold to the interaction frequency during the monitoring period.

[0079] The interactive presentation decision coefficient is obtained by weighting and summing the presentation frequency value, presentation time value, presentation optimization coefficient, and interaction feature coefficient. Specifically, the presentation frequency value, presentation time value, presentation optimization coefficient, and interaction feature coefficient are each assigned a corresponding preset weight coefficient. The presentation frequency value, presentation time value, presentation optimization coefficient, and interaction feature coefficient are then multiplied by their respective preset weight coefficients, and the sum of the four product results is marked as the interactive presentation decision coefficient.

[0080] It should be noted that the larger the value of the interactive presentation decision coefficient, the worse the overall operation of the integrated visualization interactive presentation module is during the monitoring period, and the higher the probability of performance degradation. The interactive presentation decision coefficient is compared with the preset interactive presentation decision coefficient threshold. If the interactive presentation decision coefficient exceeds the preset interactive presentation decision coefficient threshold, it indicates that the overall operation of the integrated visualization interactive presentation module is poor during the monitoring period, and the probability of performance degradation is high. In this case, a performance degradation warning signal is generated.

[0081] The working principle of this invention is as follows: During use, the enterprise multi-source heterogeneous data adaptation and access module effectively supports multiple types of internal and external data sources and formats, avoiding the impact of data chaos and bias on analysis results from the source. The dynamic digital twin constructed by the enterprise entity twin dynamic construction module enables real-time synchronization of enterprise operating status and provides early warning of equipment failures, reducing operational losses. The multi-dimensional analysis system of the BI visualization dimension data aggregation module meets the needs of users at different levels, is compatible with mainstream BI tools to reduce integration costs, and provides a structured data source for visualization presentation. The enterprise operating anomaly multi-feature fusion identification module distinguishes between positive and negative features, calculates directional anomaly deviation rate and comprehensive anomaly degree, avoiding misjudgments caused by traditional single features or lack of directional distinction, helping enterprises respond to risks in a timely manner and make targeted rectifications. The integrated interface created by the visualization interactive presentation module achieves dual presentation of "digital twin scenario + BI quantitative indicators," improving users' understanding of operating status efficiency, supporting personalized interactive operations, and facilitating the optimization of enterprise management processes.

[0082] In this invention, the threshold, preset value, or preset range settings are for result comparison and analysis to determine whether the result is good or bad. The magnitude of these values ​​is determined by a combination of large-scale model analysis of sample data and human experience, and can also be appropriately adjusted based on seasonal or common-sense influence conditions. Similarly, the preset weight coefficients and influence factors are assigned specific values ​​based on the magnitude of each parameter's influence on the result, ultimately reflecting the impact on the result. These settings are also determined by a combination of large-scale model analysis of sample data and human experience, and can also be appropriately adjusted based on seasonal or common-sense influence conditions.

[0083] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, enabling those skilled in the art to better understand and utilize it. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. An enterprise data visualization interaction management system that fuses digital twin with BI dashboard, characterized in that, The application comprises an enterprise multi-source heterogeneous data adaptive access module, a data preprocessing output module, an enterprise entity twin dynamic construction module, a BI visual dimension data aggregation module, an enterprise operation abnormality multi-feature fusion identification module, and a fusion visual interactive presentation module. The enterprise multi-source heterogeneous data adaptive access module is used for safe adaptation and classified access of multi-source heterogeneous data inside and outside the enterprise, and the data preprocessing output module is used for cleaning, conversion and integration of the accessed raw data. The enterprise entity twin dynamic construction module constructs a dynamic digital twin mapped with the enterprise physical entity based on the preprocessed standardized data, and synchronizes the enterprise operation state in real time. The BI visual dimension data aggregation module aggregates data according to the analysis dimensions of the BI board based on the preprocessed standardized data, and generates structured BI visual data sources. The enterprise operation abnormality multi-feature fusion identification module identifies the enterprise operation abnormality and sends the operation abnormality information to the fusion visual interactive presentation module. The fusion visual interactive presentation module fuses the digital twin and the BI board data to construct an integrated visual interactive interface.

2. The enterprise data visualization interaction management system fusing digital twin and BI dashboard according to claim 1, characterized in that, The enterprise multi-source heterogeneous data adaptive access module accesses the enterprise internal business system and external data system through a preset multi-type data source interface, automatically identifies the data format of each data source, converts unstructured data into semi-structured data based on a preset format mapping rule, and converts semi-structured data into standardized structured data. Meanwhile, the data source access authority is verified through a triple mechanism. And the adapted data is marked according to six categories to generate a data access classification list. The data preprocessing output module receives the classified raw data, sequentially performs data cleaning, data conversion and data integration operations, divides the integrated standardized data into two paths, one of which filters key data according to the twin modeling requirements and transmits to the enterprise entity twin dynamic construction module, and the other extracts summary indicators according to the BI dimension requirements and transmits to the BI visual dimension data aggregation module.

3. The enterprise data visualisation interaction management system fusing digital twin with BI dashboard according to claim 2, characterized in that, The enterprise entity twin dynamic construction module imports the basic model of the enterprise physical entity, associates and binds the standardized data with the entity attributes in the model, dynamically updates the twin state and constructs a data bidirectional feedback channel, and generates an enterprise dynamic twin data set.

4. The enterprise data visualization interaction management system fusing digital twin and BI dashboard according to claim 2, characterized in that, The BI visual dimension data aggregation module determines the core analysis dimension system of the BI board, aggregates data according to the mapping rule, structures and encapsulates the aggregated data, and generates a BI dimension data cube conforming to the BI tool import format.

5. The enterprise data visualization interaction management system fusing digital twin and BI dashboard according to claim 1, characterized in that, The analysis process of the enterprise operation abnormality multi-feature fusion identification module is as follows: If there is a feature with an abnormal deviation rate Ri exceeding the corresponding preset abnormal deviation rate threshold, it is determined that the enterprise has operation abnormality; if there is no feature with an abnormal deviation rate Ri exceeding the corresponding preset abnormal deviation rate threshold, the comprehensive abnormality degree F is calculated, the abnormality determination threshold T is determined based on the enterprise operation data statistics of the past three years, and if F>T, it is determined that the enterprise has operation abnormality.

6. The enterprise data visualization interaction management system fusing digital twin and BI dashboard according to claim 5, characterized in that, The analysis and calculation process of the abnormal deviation rate Ri is as follows: Determine the reasonable benchmark value Xio of each type of feature, for the positive feature, collect the actual value Xi of the feature, calculate the abnormal deviation rate Ri of the feature by the formula Ri=(Xio-Xi) / Xio*100%, and when Xi is greater than or equal to Xio, the abnormal deviation rate Ri is 0; for the negative feature, calculate the abnormal deviation rate Ri of the feature by the formula Ri=(Xi-Xio) / Xio*100%, and when Xi is less than or equal to Xio, the abnormal deviation rate Ri is 0.

7. The enterprise data visualization interaction management system fusing digital twin and BI dashboard according to claim 1, characterized in that, The fusion visual interactive presentation module is connected with the multi-level performance comprehensive evaluation module, the multi-level performance comprehensive evaluation module evaluates and analyzes the performance status of the fusion visual interactive presentation module, and determines whether to generate a performance degradation alarm signal through analysis.

8. The enterprise data visualization interaction management system fusing digital twin and BI dashboard according to claim 7, characterized in that, The specific analysis process of the multi-level performance evaluation module is as follows: Set the monitoring period as L1, collect the number of times that the fusion visual interactive presentation module cannot perform display presentation in the monitoring period and mark it as a presentation abnormal frequency value, and mark the total time length of the fusion visual interactive presentation module that cannot perform presentation in the monitoring period as a presentation abnormal time value, if the presentation abnormal frequency value or the presentation abnormal time value exceeds the corresponding preset threshold, a performance degradation alarm signal is generated.

9. The enterprise data visualization interaction management system fusing digital twin and BI dashboard, according to claim 8, characterized in that, If the presentation abnormal frequency value and the presentation abnormal time value do not exceed the corresponding preset threshold, the total time length of the fusion visual interactive presentation module in the monitoring period in the presentation optimization state is obtained and marked as a presentation optimization coefficient, if the presentation optimization coefficient exceeds the preset presentation optimization coefficient threshold, a performance degradation warning signal is generated; If the presentation optimization coefficient does not exceed the preset presentation optimization coefficient threshold, an interactive presentation decision coefficient is obtained through multi-dimensional comprehensive evaluation analysis, if the interactive presentation decision coefficient exceeds the preset interactive presentation decision coefficient threshold, a performance degradation warning signal is generated.

10. The enterprise data visualization interaction management system fusing digital twin and BI dashboard according to claim 9, characterized in that, The specific analysis process of the multi-dimensional comprehensive evaluation analysis is as follows: the number of times that the user clicks on the fusion visual interactive presentation module in the monitoring period is obtained and marked as an interaction frequency, and the number of times that the click response value exceeds the preset click response threshold in the monitoring period is calculated by the ratio of the interaction frequency to obtain an interaction feature coefficient; the presentation abnormal frequency value, the presentation abnormal time value, the presentation optimization coefficient and the interaction feature coefficient are weighted and summed to obtain the interactive presentation decision coefficient.

Images