Industrial data report adaptive generation method, system and equipment and storage medium

Abstract

The invention relates to the technical field of data processing, and particularly provides an industrial data report self-adaptive generation method, system and device and a storage medium, and the method comprises the steps: obtaining heterogeneous data from a plurality of data sources, and converting the heterogeneous data into industrial data in a unified format; generating a dynamic boundary value based on the industrial data as an index by using an adaptive dynamic boundary model; and analyzing the industrial data based on the dynamic boundary value of the index, and generating a report based on an analysis result. According to the method, the report is automatically generated based on intelligent analysis of the dynamic boundary, so that the current real operation state is objectively reflected, potential risks and performance changes can be revealed in time, and a more accurate and more time-efficient decision basis is provided for management personnel.

Description

Technical Field

[0001] This invention belongs to the field of data processing technology, and specifically relates to an adaptive generation method, system, device and storage medium for industrial data reports. Background Technology

[0002] In the industrial manufacturing and operations sector, data reports are crucial for status monitoring, performance evaluation, and decision support. With the development of the Industrial Internet of Things (IIoT), data sources are becoming increasingly diverse and heterogeneous, encompassing various devices, sensors, and business systems. Their formats, protocols, and sampling frequencies differ significantly, leading to difficulties in data integration and low report generation efficiency.

[0003] Traditional report generation methods typically rely on pre-set static thresholds or rules to determine indicator status and generate reports. However, industrial site conditions are complex and constantly changing, with equipment status, production load, and environmental factors all fluctuating dynamically. Static thresholds struggle to adapt to these real-time changes, often leading to inaccurate report results: either generating numerous invalid alarms (false alarms) for normal fluctuations, or failing to capture genuine anomalies in a timely manner (missed reports). This not only reduces the report's reference value but also affects managers' real-time judgment and response efficiency regarding production conditions.

[0004] Therefore, there is an urgent need for a method that can automatically adapt to the dynamic characteristics of industrial data and intelligently generate accurate reports. Summary of the Invention

[0005] In view of the above-mentioned shortcomings of the prior art, the present invention provides an adaptive generation method, system, device and storage medium for industrial data reports to solve the above-mentioned technical problems.

[0006] In a first aspect, the present invention provides an adaptive generation method for industrial data reports, comprising: Obtain heterogeneous data from multiple data sources and convert the heterogeneous data into industrial data in a unified format; An adaptive dynamic boundary model is used to generate dynamic boundary values ​​based on the industrial data as an indicator. The industrial data is analyzed based on the dynamic boundary value of the indicators, and reports are generated based on the analysis results.

[0007] In one optional implementation, heterogeneous data is acquired from multiple data sources and converted into industrial data in a unified format, including: A strategy combining incremental and full data collection is adopted to acquire heterogeneous data from multiple data sources; The metadata management system enables unified modeling and storage of metadata from various data sources. Define corresponding data transformation rules for various data structures; The data structure of the heterogeneous data is determined based on the metadata of the data source to which the heterogeneous data belongs, and the corresponding data conversion rules are called to convert the heterogeneous data into industrial data in a preset format.

[0008] In an optional implementation, an adaptive dynamic boundary model is used to generate dynamic boundary values ​​for indicators based on the industrial data, including: Identify the associated parameter items of the indicators, and extract the parameter values ​​of the associated parameter items from the industrial data; Obtain the historical boundary values ​​of the indicator; Get system status; Dynamic boundary values ​​of indicators are generated based on the parameter values ​​of associated parameter items, historical boundary values, and system status.

[0009] In one optional implementation, the adaptive dynamic boundary model includes:

[0010] in, This is a real-time data vector for the associated parameter items; A vector of historical data for the indicator; Θ, Φ, Ψ are the system state vectors; Θ, Φ, Ψ are the model parameters; λ, γ are the mode switching parameters; It belongs to the nonparametric regression model; It belongs to the mixed-scale correction function; This represents the neural network correction term.

[0011] In an optional implementation, generating dynamic boundary values ​​for the metric based on the parameter values ​​of the associated parameter items and historical boundary values ​​includes: Obtain the indicator type, and determine the values ​​of the mode switch parameters λ and γ and the model parameters Θ, Φ, Ψ based on the indicator type; Substituting the values ​​of the mode switching parameters λ and γ and the values ​​of the model parameters Θ, Φ, and Ψ into the adaptive dynamic boundary model, we obtain the specific model for the index type. The dynamic boundary value of the indicator is generated by substituting the parameter values ​​of the associated parameter items and historical boundary values ​​into the dedicated model to obtain the dynamic boundary value of the indicator type.

[0012] In one optional implementation, the industrial data is analyzed based on the dynamic boundary values ​​of the indicators, and reports are generated based on the analysis results, including: Establish the correspondence between parameters and indicators in industrial data; Compare the dynamic boundary values ​​of the indicator with the parameter values ​​of the corresponding parameter items; Summarize the comparison results of all indicators and parameter values, and generate a report based on the comparison results.

[0013] In one optional implementation, a report is generated based on the comparison results, including: Determine the report type based on the indicator type and the corresponding parameter item type; Reports are generated based on report type and comparison results.

[0014] Secondly, the present invention provides an adaptive industrial data report generation system, comprising: The data conversion module is used to acquire heterogeneous data from multiple data sources and convert the heterogeneous data into industrial data in a unified format. The indicator update module is used to generate dynamic boundary values ​​for the indicators based on the industrial data using an adaptive dynamic boundary model. The report generation module is used to analyze the industrial data based on the dynamic boundary values ​​of the indicators and generate reports based on the analysis results.

[0015] Thirdly, a device is provided, comprising: Memory is used to store the adaptive generation program for industrial data reports; A processor is configured to implement the steps of the industrial data report adaptive generation method as provided in the first aspect when executing the industrial data report adaptive generation program.

[0016] Fourthly, a computer-readable storage medium is provided, on which an industrial data report adaptive generation program is stored, wherein when the industrial data report adaptive generation program is executed by a processor, the steps of the industrial data report adaptive generation method provided in the first aspect are implemented.

[0017] The beneficial effects of this invention are as follows: the industrial data report adaptive generation method, system, device, and storage medium provided by this invention significantly improve the compatibility and processing efficiency of data sources by integrating multi-source heterogeneous data and unifying the format. The core lies in using an adaptive dynamic boundary model to generate dynamic boundary values ​​for indicators, effectively overcoming the limitations of traditional static thresholds. This model can adapt in real time to the complex and ever-changing production loads, equipment status, and environmental conditions in industrial settings, greatly improving the accuracy of anomaly detection and significantly reducing false alarms and missed alarms. Based on intelligent analysis of dynamic boundaries, the automated report generation not only objectively reflects the current real operating status but also promptly reveals potential risks and performance changes, providing managers with more accurate and timely decision-making basis, ultimately improving the intelligence level and response speed of industrial operations. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic flowchart of a method according to an embodiment of the present invention.

[0020] Figure 2 This is a schematic block diagram of a system according to an embodiment of the present invention.

[0021] Figure 3 This is a schematic diagram of the structure of a device provided in an embodiment of the present invention. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

[0024] The adaptive generation method for industrial data reports provided in this embodiment of the invention is executed by a computer device, and correspondingly, the adaptive generation system for industrial data reports runs on the computer device.

[0025] Figure 1 This is a schematic flowchart illustrating a method according to an embodiment of the present invention. Wherein, Figure 1 The executing entity can be an adaptive industrial data report generation system. Depending on different needs, the order of steps in this flowchart can be changed, and some can be omitted.

[0026] like Figure 1 As shown, the method includes: S1. Obtain heterogeneous data from multiple data sources and convert the heterogeneous data into industrial data in a unified format; S2. Using an adaptive dynamic boundary model, dynamic boundary values ​​are generated based on the industrial data as indicators; S3. Analyze the industrial data based on the dynamic boundary values ​​of the indicators, and generate reports based on the analysis results.

[0027] In one embodiment of the present invention, based on step S1, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0028] S101. A strategy combining incremental and full data collection is adopted to obtain heterogeneous data from multiple data sources.

[0029] The data acquisition process employs a hybrid acquisition strategy to balance data integrity and resource consumption. For relational databases (such as MySQL and PostgreSQL), incremental data is obtained by polling the database transaction logs (such as MySQL's Bin log and PostgreSQL's WAL log), and the data update status is marked using timestamp or version number fields. Full data snapshot backups are performed periodically. For time-series databases (such as Influx DB and Timescale DB), incremental acquisition is performed based on a time window mechanism, pulling the latest data in minutes or hours, and full historical data archiving is performed monthly. For unstructured data acquisition, Flume's Tail Dir Source component is used to monitor file appending in real time for log files generated by industrial IoT devices, and key information is extracted using regular expressions. For historical documents in file systems, full data acquisition is achieved through a recursive scanning mechanism of a distributed file system (such as HDFS). During the acquisition process, a Kafka message queue is introduced as a data buffer layer. Data throughput and reliability are optimized by setting the number of partitions and the replication factor. The Kafka Connect framework is used to achieve parallel acquisition and asynchronous transmission of multi-source data.

[0030] S102. Based on the metadata management system, perform unified modeling and storage of metadata from various data sources.

[0031] The metadata management system adopts a three-tier metadata model architecture: the base layer stores data source connection information (such as IP address, port number, and authentication credentials); the structure layer describes data field attributes (data type, length, precision, and constraints); and the semantic layer defines business concept mapping relationships (such as the correspondence between device codes and product models). The system is built on Apache Atlas, using a plug-in architecture to adapt to different data sources and leveraging a RESTful API for dynamic registration and updates of metadata. During the modeling process, for relational databases, table structure and field information are automatically extracted via JDBC metadata interface; for unstructured data, NLP technology is used to analyze document content, extract keywords and topic tags to construct semantic metadata. Metadata storage uses Neo4j graph database, which establishes entity-relationship graph models (such as hierarchical relationships between data sources, data tables, and data fields) to achieve metadata association queries and traceability analysis, while using a version control mechanism to record metadata change history.

[0032] S103. Define corresponding data transformation rules for various data structures.

[0033] The data transformation rules are designed following a layered and decoupled principle, constructing a three-tiered rule system: "Data Source Adaptation Layer - Format Conversion Layer - Semantic Mapping Layer". In the data source adaptation layer, dedicated connectors are designed for different data source types. For example, relational databases use an SQL transformation engine to map table structures, while file systems use regular expression templates to parse data formats. The format conversion layer leverages Apache NiFi's dynamic data flow orchestration capabilities, configuring processor chains to perform data type conversions (e.g., string to date format), field splitting and merging, and missing value imputation (mean imputation, interpolation). The semantic mapping layer utilizes ontology engineering technology, building an industrial data ontology model based on the OWL language. By defining mapping rules (e.g., SKOS concept mapping), it unifies the business terms of heterogeneous data into standard terms.

[0034] The rules are stored in JSON Schema format. Each rule contains elements such as triggering conditions (e.g., data source type, data structure identifier), transformation operations (e.g., field renaming, data aggregation), and output format definition. The rules are dynamically loaded and prioritized through a rule engine (e.g., Drools).

[0035] S104. Determine the data structure of the heterogeneous data based on the metadata of the data source to which the heterogeneous data belongs, and call the corresponding data conversion rules to convert the heterogeneous data into industrial data in a preset format.

[0036] The data transformation process employs a metadata-driven intelligent decision-making mechanism. The system first obtains the structural and semantic metadata of the data source through the metadata management system, and then uses pattern matching algorithms (such as regular expression matching and tree structure comparison) to identify the data structure type. For data sources identified as relational data structures, SQL transformation rules are invoked for field mapping and data cleaning; for semi-structured data in JSON format, key nodes are extracted using JSON Path expressions and converted to the target format; for unstructured text data, natural language processing techniques (such as named entity recognition and information extraction) are used to extract structured information before format conversion. During the transformation execution phase, the Apache Spark distributed computing framework is used to achieve parallel data processing, and transformation operations are performed through RDD (Resilient Distributed Dataset) or Data Frame API. A data quality monitoring module is embedded during the transformation process, using predefined data validation rules (such as uniqueness constraints and value range validation) to detect data quality in real time. Abnormal data is triggered to trigger cleaning processes (such as deduplication, error correction, and data completion). Finally, the standardized industrial data is stored in a data warehouse or data lake, providing a foundation for subsequent data analysis and applications.

[0037] In one embodiment of the present invention, based on step S2, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0038] S201. Determine the associated parameter items of the indicator, and extract the parameter values ​​of the associated parameter items from the industrial data; S202. Obtain the historical boundary values ​​of the indicator; S203. Obtain system status; S204. Generate dynamic boundary values ​​for indicators based on the parameter values ​​of associated parameter items, historical boundary values, and system status.

[0039] Adaptive dynamic boundary models include:

[0040] in, This is a real-time data vector for the associated parameter items; A vector of historical data for the indicator; Θ, Φ, Ψ are the system state vectors; Θ, Φ, Ψ are the model parameters; λ, γ are the mode switching parameters; It belongs to the nonparametric regression model; It belongs to the mixed-scale correction function; This represents the neural network correction term.

[0041] This belongs to a nonparametric regression model:

[0042] in, This represents the reference vector.

[0043] Kernel Gaussian kernel function: .

[0044] These are learnable parameters.

[0045] This function can establish a nonlinear mapping relationship between variables and boundaries.

[0046] Belongs to mixed-scale correction functions:

[0047] Φ is the trend decomposition parameter (such as λ in HPF, wavelet basis type).

[0048] Representing neural network correction terms:

[0049] Ψ represents the GNN parameters, including the adjacency matrix A and the propagation weights.

[0050] (1) Obtain the index type, and determine the values ​​of the mode switch parameters λ and γ and the model parameters Θ, Φ, Ψ based on the index type.

[0051] Based on the physical characteristics and data features of the indicators, they are divided into: Slowly changing indicators: such as equipment temperature and pressure; Abrupt change indicators: such as fault alarms and abnormal events; Cyclical indicators: such as production cycle time and energy consumption fluctuations.

[0052] Preset initial values ​​for parameters based on the indicator type.

[0053] (2) Substitute the values ​​of the mode switch parameters λ and γ and the values ​​of the model parameters Θ, Φ, Ψ into the adaptive dynamic boundary model to obtain the specific model of the index type.

[0054] By using transfer learning, the current model is initialized with historical training parameters of similar metrics, thus accelerating convergence.

[0055] (3) Substitute the parameter values ​​of the associated parameter items and the historical boundary values ​​into the specific model to obtain the dynamic boundary values ​​of the indicator type.

[0056] Data preprocessing: processing the input vector , and Normalization is performed; the outputs of the three model components are computed in parallel; weighted fusion is used to obtain dynamic boundary values; and physical constraints are applied to correct unreasonable boundary predictions.

[0057] In one embodiment of the present invention, based on step S3, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0058] S301. Establish the correspondence between parameters and indicators in industrial data.

[0059] Semantic Web technology is used to construct an industrial data knowledge graph, serving as the foundation for storing and managing the correspondence between parameters and indicators. Using the Neo4j graph database as the storage medium, parameters and indicators in the industrial data are abstracted as nodes in the knowledge graph, and the relationships between them are abstracted as edges. Node attributes include information such as the name, data type, unit, and description of the parameter or indicator, while edge attributes record the association strength and association type (e.g., direct influence, indirect influence). For structured data, the relationships between parameters and indicators are automatically extracted by parsing the database table structure and field comments. For example, in a production equipment monitoring database, there is a positive correlation between equipment speed and equipment capacity indicators. This relationship can be determined by analyzing the calculation logic between fields using SQL statements (e.g., capacity = speed × production quantity per unit time). For unstructured data, such as equipment maintenance logs, named entity recognition (NER) and relation extraction (RE) algorithms from Natural Language Processing (NLP) are used to extract parameters (e.g., "bearing temperature") and indicators (e.g., "equipment failure probability") and their relationships from the text.

[0060] A relationship verification mechanism was established, employing a combination of expert experience and data mining to validate the initially established correspondences. Industry experts were invited to manually review the relationships between key parameters and indicators, correcting obvious errors or unreasonable associations. Simultaneously, association rule mining algorithms (such as the Apriori algorithm) were used to analyze historical industrial data, calculating the support and confidence levels between parameters and indicators to identify combinations with strong correlations. For relationships with low confidence, further analysis of data characteristics or remodeling of the relationships was conducted to ensure their accuracy and reliability.

[0061] S302. Compare the dynamic boundary values ​​of the indicator with the parameter values ​​of the corresponding parameter items.

[0062] Before comparison, the dynamic boundary values ​​of the indicators and the parameter values ​​of the parameters need to be preprocessed. Appropriate standardization methods are adopted for different data types: for numerical data, Z-score standardization is used to map the data to the same scale; for categorical data, one-hot encoding or label encoding is used for numerical conversion. Simultaneously, missing values ​​are handled: for numerical data, mean imputation, median imputation, or machine learning-based predictive imputation (such as random forest regression) are used; for categorical data, mode imputation or the creation of new "missing" categories is used.

[0063] Different comparison algorithms are designed based on the type of indicator and the characteristics of the parameter items: Comparison of numerical indicators and parameter items: For continuous numerical indicators, such as equipment temperature indicators, calculate the difference between the actual temperature value of the parameter item and the dynamic boundary value of the indicator (ΔT=T). actual -T boundaryThe system determines whether the temperature exceeds the boundary based on the difference between the values. Multiple warning levels are set, such as ≤5℃ for normal conditions, 5℃ < ≤10℃ for a mild warning, and ≤10℃ for a severe warning. For discrete numerical indicators, such as production batch pass rate, the actual pass rate is directly compared with the dynamic boundary value; if the actual pass rate is lower than the boundary value, a warning is triggered. Classification indicators and parameter comparison: A pattern matching algorithm, such as the Longest Common Subsequence (LCS) algorithm, is used to calculate the degree of matching between the classification results of the parameter items and the dynamic boundary values ​​of the indicators. For example, in the product quality classification indicators, the actual product quality categories (such as "qualified", "defective", "scrap") are compared with the expected categories in the dynamic boundary values. If a mismatch occurs, the difference information is recorded and an early warning is generated. Time-series indicator and parameter comparison: Utilizing the Dynamic Time Warping (DTW) algorithm, the problem of incomplete alignment of time-series data on the time axis is addressed by calculating the similarity distance between the parameter time series and the indicator's dynamic boundary value time series. When the distance exceeds a set threshold, an abnormal deviation between the parameter and the indicator is identified, and an early warning signal is issued.

[0064] S303. Summarize the comparison results of all indicators and parameter values, and generate a report based on the comparison results.

[0065] (1) Determine the report type based on the indicator type and the corresponding parameter item type; Based on the indicator type and the corresponding parameter item type, reports are divided into several types: Status monitoring reports: These are suitable for indicators with high real-time requirements, such as equipment operating status indicators and key parameters of the production process. These reports are displayed in dashboard format and include elements such as real-time data, dynamic trend graphs, and warning status indicators, allowing operators to quickly grasp the system's operating status. Anomaly analysis reports: These reports are generated for indicators and parameters that exhibit anomalies, including detailed information such as the time of occurrence, anomaly type, involved parameters, and specific differences. Fault tree analysis (FTA) or root cause analysis (RCA) methods are used to trace the root cause of the anomalies and provide handling recommendations. Trend forecasting reports: Based on historical comparison results and time series forecasting algorithms (such as ARIMA and LSTM), these reports predict the future trends of indicators and parameters. The reports are displayed in the form of line charts, forecast interval charts, etc., helping decision-makers to plan production resources and develop maintenance plans in advance. Comprehensive statistical reports: These reports summarize the comparison results of various indicators and parameters by time period (e.g., daily, weekly, monthly) and perform statistical analysis, such as calculating pass rate, failure rate, mean, standard deviation, and other statistical quantities. The reports are presented in the form of tables, bar charts, pie charts, etc., and are used to evaluate production performance and optimize production processes.

[0066] (2) Generate reports based on report type and comparison results.

[0067] Report generation is achieved using a template-driven approach combined with data visualization technology. The specific process is as follows: Data filtering and integration: Based on the report type and time range requirements, relevant data is filtered from the comparison results database. For complex reports, it may be necessary to integrate data from multiple data sources, such as linking equipment operation data with production order data, in order to analyze production from multiple dimensions. Template Selection and Configuration: Select a template from the report template library that matches the report type. For example, a bar chart template is used to compare the output of different production lines, and a heat map template is used to present the temperature distribution of various components of the equipment. Configure the template parameters according to the actual data characteristics and display requirements, such as axis labels, color scheme, and data label display method. Data population and visualization: The filtered and integrated data is populated into the report template, and visual charts are generated using data visualization libraries (such as EC Harts and D3.js). For dynamic reports, Web Socket technology is used to achieve real-time data updates, ensuring that operators receive the latest information. Report generation and output: Combines visual charts with relevant text descriptions and statistical data into complete reports, supporting multiple output formats such as PDF, Excel, and HTML. Provides report download, printing, and sharing functions for convenient use by users with different roles. In practical industrial applications, taking an automobile manufacturing company as an example, the above method was used to establish a correspondence between production equipment parameters (such as stamping press pressure and welding robot current) and product quality indicators (such as parts qualification rate and welding strength). Through dynamic boundary value comparison and report generation, the company can promptly identify quality problems caused by equipment anomalies, effectively improving production efficiency and product quality.

[0068] This embodiment provides an adaptive generation method for industrial data reports, including: Data acquisition and transformation: Step 1: Mixed Data Acquisition Relational database (MySQL example): # Monitor incremental data using Binlog binlog_stream = MysqlBinlogConnector(host="192.168.1.100", interval=60) , # Daily full snapshot (executed at midnight) if current_time == "00:00": execute_snapshot("CREATE SNAPSHOT AS SELECT * FROM sensor_data").

[0069] Time-series database (InfluxDB example): Retrieve time window data every minute: SELECT * FROM pressure WHERE time>now() - 1m, and archive all data to HDFS monthly.

[0070] Unstructured data: Configuring TailDir Source monitoring log files using Flume: agent.sources = tailSrc, agent.sources.tailSrc.type = TAILDIR, agent.sources.tailSrc.positionFile = / opt / flume / taildir_position.json.

[0071] Step 2: Metadata-driven transformation Extracting metadata (field names, types) from MySQL table structures using Apache Atlas Invoking a pre-defined JSON Schema rule (example): { "rule_id": "TEMP_CONV_001", "trigger": {"source_type": "MySQL", "table": "sensor_raw"}, "operations": [ {"action": "rename_field", "old_name": "temp_val", "new_name": "temperature"}, {"action": "convert_unit", "field": "temperature", "from": "°F", "to": "°C"} ]}.

[0072] Performing distributed transformations using Spark: val df = spark.read.format("jdbc").load("sensor_raw") , val converted = df.withColumn("temperature", (col("temp_val")-32)*5 / 9).

[0073] Dynamic boundary value generation (taking equipment temperature as an example): Step 1: Parameter Extraction Related parameters: cooling water flow rate, ambient humidity, equipment load rate; Real-time data vector: X_t = [flow rate = 120L / min, humidity = 45%, load rate = 85%]; Step 2: Model Parameter Configuration Indicator type: Slow-change indicator → Mode switch parameters λ=0.7, γ=0.3.

[0074] Model parameter initialization: Θ=[0.12,-0.05,0.08]^T, Φ={HPF_λ=14400, wavelet basis='db4'}, Ψ = {adjacency matrix A ∈ R^{3×3}, propagation weight W ∈ R^{64×64}} Step 3: Boundary Value Calculation Nonparametric regression term: f_{NPR}(X_t)=\sum_{i=1}^{n}\exp(-\frac{\|X_t- X_{hist_i}\|^2}{2\sigma^2}) \cdot Y_{hist_i}, Output dynamic boundary value: T_boundary = 68.5±2.3°C (corrected by physical constraints).

[0075] Indicator monitoring and report generation: Step 1: Parameter-Indicator Comparison Real-time temperature: T_actual = 72.1℃; Deviation calculation: ΔT = 72.1 - 68.5 = 3.6℃; Warning triggered: 5℃ < |ΔT | ≤ 10℃ → Mild warning.

[0076] Step 2: Knowledge Graph-Assisted Analysis Query relationships using Neo4j: MATCH(p:Parameter{name:"Cooling water flow rate"})-[r:IMPACTS]->(m:Metric {name:"Equipment temperature"}); RETURN r.strength, r.type / / Returns the strength of the effect, 0.85 (direct effect).

[0077] Step 3: Generate anomaly analysis report

[0078] In the visualization component, the real-time trend chart is overlaid with a dynamic boundary band, and warning points are marked in red.

[0079] In some embodiments, the adaptive industrial data report generation system may include multiple functional modules composed of computer program segments. The computer programs for each program segment in the adaptive industrial data report generation system may be stored in the memory of a computer device and executed by at least one processor to perform (see details). Figure 1 (Description) The function of adaptive generation of industrial data reports.

[0080] In this embodiment, the industrial data report adaptive generation system can be divided into multiple functional modules based on its functions, such as... Figure 2 As shown. The module referred to in this invention is a series of computer program segments that can be executed by at least one processor and perform a fixed function, and is stored in memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0081] The data conversion module is used to acquire heterogeneous data from multiple data sources and convert the heterogeneous data into industrial data in a unified format. The indicator update module is used to generate dynamic boundary values ​​for the indicators based on the industrial data using an adaptive dynamic boundary model. The report generation module is used to analyze the industrial data based on the dynamic boundary values ​​of the indicators and generate reports based on the analysis results.

[0082] Figure 3 The adaptive generation method for industrial data reports provided in this application embodiment can be applied to devices. Those skilled in the art will understand that the device structure involved in the embodiments of this invention does not constitute a limitation on the device. A device may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. In the embodiments of this invention, the device includes, but is not limited to, laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of this application described and / or claimed herein.

[0083] The device 300 may include a processor 310, a memory 320, and a communication unit 330. These components communicate via one or more buses. Those skilled in the art will understand that the server structure shown in the figure does not constitute a limitation of the present invention. It may be a bus topology or a star topology, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0084] The memory 320 can be used to store execution instructions of the processor 310. The memory 320 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. When the execution instructions in the memory 320 are executed by the processor 310, the device 300 is able to perform some or all of the steps in the above method embodiments.

[0085] The processor 310 serves as the control center of the storage device, connecting various parts of the electronic device via various interfaces and lines. It executes software programs and / or modules stored in the memory 320, and calls data stored in the memory to perform various functions of the electronic device and / or process data. The processor can be composed of integrated circuits (ICs), such as a single packaged IC or multiple packaged ICs with the same or different functions connected together. For example, the processor 310 may consist only of a central processing unit (CPU). In this embodiment of the invention, the CPU may have a single processing core or include multiple processing cores.

[0086] The communication unit 330 is used to establish a communication channel, enabling the storage device to communicate with other devices. It can receive user data sent by other devices or send user data to other devices.

[0087] The present invention also provides a computer storage medium, wherein the computer storage medium may store a program, which, when executed, may include some or all of the steps provided in the embodiments of the present invention. The storage medium may be a magnetic disk, an optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0088] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or any other medium capable of storing program code. It includes several instructions to cause a computer device (which may be a personal computer, a server, or a second device, network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0089] The same or similar parts between the various embodiments in this specification can be referred to mutually. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple, and the relevant parts can be referred to the description in the method embodiments.

[0090] In the embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or modules may be electrical, mechanical, or other forms.

[0091] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0092] In addition, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0093] Although the present invention has been described in detail with reference to the accompanying drawings and preferred embodiments, the present invention is not limited thereto. Various equivalent modifications or substitutions can be made to the embodiments of the present invention by those skilled in the art without departing from the spirit and essence of the invention, and such modifications or substitutions should all be within the scope of the present invention. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should also be covered within the protection scope of the present invention.

Claims

1. An adaptive generation method for industrial data reports, characterized in that, include: Obtain heterogeneous data from multiple data sources and convert the heterogeneous data into industrial data in a unified format; An adaptive dynamic boundary model is used to generate dynamic boundary values ​​based on the industrial data as an indicator. The industrial data is analyzed based on the dynamic boundary value of the indicators, and reports are generated based on the analysis results.

2. The method according to claim 1, characterized in that, Acquire heterogeneous data from multiple data sources and transform it into industrial data in a unified format, including: A strategy combining incremental and full data collection is adopted to acquire heterogeneous data from multiple data sources; The metadata management system enables unified modeling and storage of metadata from various data sources. Define corresponding data transformation rules for various data structures; The data structure of the heterogeneous data is determined based on the metadata of the data source to which the heterogeneous data belongs, and the corresponding data conversion rules are called to convert the heterogeneous data into industrial data in a preset format.

3. The method according to claim 1, characterized in that, Using an adaptive dynamic boundary model to generate dynamic boundary values ​​for indicators based on the industrial data, including: Identify the associated parameter items of the indicators, and extract the parameter values ​​of the associated parameter items from the industrial data; Obtain the historical boundary values ​​of the indicator; Get system status; Dynamic boundary values ​​of indicators are generated based on the parameter values ​​of associated parameter items, historical boundary values, and system status.

4. The method according to claim 3, characterized in that, The adaptive dynamic boundary model includes: in, This is a real-time data vector for the associated parameter items; A vector of historical data for the indicator; Θ, Φ, Ψ are the system state vectors; Θ, Φ, Ψ are the model parameters; λ, γ are the mode switching parameters; It belongs to the nonparametric regression model; It belongs to the mixed-scale correction function; This represents the neural network correction term.

5. The method according to claim 4, characterized in that, The dynamic boundary values ​​of the indicator are generated based on the parameter values ​​of the associated parameter items and historical boundary values, including: Obtain the indicator type, and determine the values ​​of the mode switch parameters λ and γ and the model parameters Θ, Φ, Ψ based on the indicator type; Substituting the values ​​of the mode switching parameters λ and γ and the values ​​of the model parameters Θ, Φ, and Ψ into the adaptive dynamic boundary model, we obtain the specific model for the index type. The dynamic boundary value of the indicator is generated by substituting the parameter values ​​of the associated parameter items and historical boundary values ​​into the dedicated model to obtain the dynamic boundary value of the indicator type.

6. The method according to claim 1, characterized in that, The industrial data is analyzed based on the dynamic boundary values ​​of the indicators, and reports are generated based on the analysis results, including: Establish the correspondence between parameters and indicators in industrial data; Compare the dynamic boundary values ​​of the indicator with the parameter values ​​of the corresponding parameter items; Summarize the comparison results of all indicators and parameter values, and generate a report based on the comparison results.

7. The method according to claim 6, characterized in that, Reports are generated based on the comparison results, including: Determine the report type based on the indicator type and the corresponding parameter item type; Reports are generated based on report type and comparison results.

8. An adaptive industrial data report generation system, characterized in that, include: The data conversion module is used to acquire heterogeneous data from multiple data sources and convert the heterogeneous data into industrial data in a unified format. The indicator update module is used to generate dynamic boundary values ​​for the indicators based on the industrial data using an adaptive dynamic boundary model. The report generation module is used to analyze the industrial data based on the dynamic boundary values ​​of the indicators and generate reports based on the analysis results.

9. A device, characterized in that, include: Memory is used to store the adaptive generation program for industrial data reports; A processor, configured to implement the steps of the industrial data report adaptive generation method as described in any one of claims 1-7 when executing the industrial data report adaptive generation program.

10. A computer-readable storage medium storing a computer program, characterized in that, The readable storage medium stores an industrial data report adaptive generation program, which, when executed by a processor, implements the steps of the industrial data report adaptive generation method as described in any one of claims 1-7.

Images