Method for Heterogeneous Industrial Data Fusion Based on Artificial Intelligence

Abstract

The present invention relates to the technical field of data analysis, and discloses a method for heterogeneous industrial data fusion based on artificial intelligence. The present invention conducts structural analysis of data fusion targets, obtains data structure information, constructs a data fusion target framework for data filling processing, constructs a data fusion channel, collects element data, extracts key data features, substitutes the key features into the fusion target framework to complete heterogeneous data fusion, improves the fusion efficiency of heterogeneous industrial data, realizes automatic feature extraction and data filling processing, enhances the intelligence and adaptability of the system, and solves the problem of low efficiency in processing large-scale multi-source heterogeneous data in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular, to a method for heterogeneous industrial data fusion based on artificial intelligence. Background Art

[0002] In the industrial field, data comes from various heterogeneous systems, such as different sensors, machine equipment, management systems, etc. These data usually have different formats, structures, and semantics. Most traditional data fusion methods rely on manually defined data mapping and conversion rules. This approach is overwhelmed when dealing with large-scale data and multi-source heterogeneous data. Traditional data fusion frameworks usually lack flexibility and scalability, and it is difficult to adapt to the changing data fusion requirements in different industrial scenarios. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for heterogeneous industrial data fusion based on artificial intelligence, aiming to solve the problem of low efficiency in processing large-scale multi-source heterogeneous data in the prior art.

[0004] The present invention is implemented as follows. In the first aspect, the present invention provides a method for heterogeneous industrial data fusion based on artificial intelligence, including:

[0005] Conducting a structural analysis of the data fusion target according to several data fusion elements to be subjected to heterogeneous industrial data fusion, so as to obtain the data structure information of the fusion target; wherein, the data fusion element is a data source participating in heterogeneous industrial data fusion.

[0006] Based on the data structure information, constructing a data fusion target framework corresponding to each of the data fusion elements; wherein, the data fusion target framework is used for data filling processing after heterogeneous industrial data fusion of each of the data fusion elements.

[0007] On the basis of the data fusion target framework, using a data feature analysis intelligent model to construct corresponding data fusion channels for each of the data fusion elements; wherein, the data fusion channel is used for data acquisition of the data fusion element.

[0008] Obtaining an element data set from the data fusion element through the data fusion channel, and enabling the data feature analysis intelligent model to extract key data from the element data set to obtain data key features corresponding to the element data set.

[0009] Substituting the data key features into the data fusion target framework to complete the heterogeneous industrial data fusion of each of the data fusion elements.

[0010] The present invention provides a method for heterogeneous industrial data fusion based on artificial intelligence, which has the following beneficial effects:

[0011] The present invention conducts a structural analysis of the data fusion target, obtains data structure information, constructs a data fusion target framework for data filling processing, constructs a data fusion channel, collects feature data, extracts key data features, substitutes the key features into the fusion target framework to complete heterogeneous data fusion, improves the fusion efficiency of heterogeneous industrial data, realizes automatic feature extraction and data filling processing, enhances the intelligence and adaptability of the system, and solves the problem of low efficiency in processing large-scale multi-source heterogeneous data in the prior art. Description of the Drawings

[0012] Figure 1 is a schematic diagram of the steps of a method for heterogeneous industrial data fusion based on artificial intelligence provided by an embodiment of the present invention. Detailed Embodiments

[0013] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0014] The implementation of the present invention will be described in detail below with reference to specific embodiments.

[0015] Refer to Figure 1 as shown, which is a preferred embodiment provided by the present invention.

[0016] In a first aspect, the present invention provides a method for heterogeneous industrial data fusion based on artificial intelligence, including:

[0017] S1: Conduct a structural analysis of the data fusion target according to several data fusion elements to be used for heterogeneous industrial data fusion, so as to obtain the data structure information of the fusion target; wherein, the data fusion element is a data source participating in heterogeneous industrial data fusion;

[0018] S2: Based on the data structure information, construct a data fusion target framework corresponding to each of the data fusion elements; wherein, the data fusion target framework is used for data filling processing after heterogeneous industrial data fusion of each of the data fusion elements;

[0019] S3: On the basis of the data fusion target framework, use a data feature analysis intelligent model to construct corresponding data fusion channels for each of the data fusion elements; wherein, the data fusion channel is used for data collection of the data fusion element;

[0020] S4: Obtain the set of feature data from the data fusion features through the data fusion channel, and let the data feature analysis intelligent model extract key data from the set of feature data to obtain the data key features corresponding to the set of feature data.

[0021] S5: Substitute the data key features into the data fusion target framework to complete the heterogeneous industrial data fusion of each data fusion feature.

[0022] Specifically, in step S1 of the embodiment provided by the present invention, first clarify all data sources participating in data fusion. These data sources include data from different systems, different formats, or different structures. For example, sensor data, production line control data, ERP system data, etc., so as to determine all features to be involved in data fusion for subsequent structural analysis.

[0023] More specifically, conduct in-depth structural analysis on each data fusion feature, collect its data structure information, including but not limited to data type, field name, field meaning, data relationship (such as primary key, foreign key), constraint conditions (such as non-null, uniqueness), etc., comprehensively master the structural information of each data fusion feature, and lay a foundation for the structural design of the fusion target.

[0024] More specifically, compare the structural information of different data features collected, analyze the similarities and differences between each data feature, especially the structural differences, determine which data fields can be directly matched, which need to be converted or mapped, identify possible conflicts (such as duplicate field names but different meanings), identify and resolve the differences between the structures of different data features, and clear obstacles for the smooth progress of data fusion.

[0025] More specifically, based on the previous analysis, design a unified data structure as the data structure of the fusion target. This structure should be able to accommodate the data in all data fusion features and solve the structural difference problem. For example, define unified field names and data types, design association relationships, etc., and construct a unified data structure that can cover all data sources as the target structure of the final fused data.

[0026] More specifically, verify the designed fusion target structure to ensure that it can effectively accommodate all data features and optimize it if necessary. The verification methods can include: comparing with actual data for testing, filling with simulated data, etc., to ensure the rationality, integrity, and adaptability of the fusion target structure, and to provide support for subsequent data fusion.

[0027] It is understandable that through detailed structural analysis, a clear understanding of the internal structure of each data fusion element is ensured. To avoid data loss, conflicts, or errors caused by unclear structural information during the data fusion process, a unified fusion target structure is designed, which can effectively integrate data from different sources. This structure provides a standardized basis for subsequent data filling and fusion, ensuring that the fused data has good operability and consistency. By means of structural analysis, possible structural problems are resolved in advance, reducing trial and error and rework during the data fusion process, and improving the efficiency and accuracy of fusion.

[0028] Specifically, in step S2 of the embodiment provided by the present invention, according to the data structure information of each obtained data fusion element, a unified data model is designed. This model should include all data fields that need to be fused and resolve differences between data structures, such as unifying field types, determining field naming rules, adjusting data relationships, etc. A unified data model is created to ensure that the data of all data fusion elements can be integrated into this model.

[0029] More specifically, mapping rules are established for the data fields in each data fusion element and the fields in the unified data model. These rules should clarify how to map data fields from different sources to the corresponding fields in the target framework, including data conversion, type conversion, unit unification, etc., to ensure that fields from different data sources can be accurately mapped into the unified data model, avoiding data loss or conflicts.

[0030] More specifically, necessary data integrity constraints are set in the data fusion target framework, including primary key constraints, foreign key constraints, uniqueness constraints, non-null constraints, etc. These constraints ensure the consistency and integrity of the data after data fusion and prevent data inconsistency or violations during the data filling process through the setting of constraint conditions.

[0031] More specifically, according to the requirements of data fusion, the preliminarily designed data fusion target framework is optimized, including eliminating data redundancy, improving performance (such as setting indexes), and designing the scalability of the framework (such as reserving expansion space for future data sources), to ensure that the data fusion target framework can not only meet the current data fusion requirements but also adapt to future expansion requirements, improving the performance of the overall system.

[0032] More specifically, through actual data filling tests, the constructed data fusion target framework is verified to ensure that it can fill data accurately and achieve the expected fusion effect. Simulated data or real data can be used during the verification process to ensure that the data fusion target framework can operate stably and reliably in actual applications, laying a foundation for subsequent industrial data fusion.

[0033] It is understandable that by designing a unified data model, heterogeneous industrial data is integrated into a unified framework to achieve standardized data management, which facilitates subsequent data query, analysis, and application. Through the optimization of the framework, not only is the system performance improved, making the data processing speed faster, but also space is reserved for future expansion requirements, enabling the system to have good scalability.

[0034] Specifically, in step S3 of the embodiment provided by the present invention, a data feature analysis intelligent model is used to extract features from the data in each data fusion element. Feature extraction includes but is not limited to data distribution, data type, statistical attributes of data, as well as time series features, spatial features, etc. of data, comprehensively understanding the features of each data fusion element, and extracting data features that can represent the accurate meaning and key characteristics of the data from the data fusion elements to achieve normalized statistical processing of the data of various different formats of fusion elements.

[0035] More specifically, the data feature analysis intelligent model is trained using an artificial intelligence model to achieve feature extraction for each data fusion element, and the data fusion channel is a data transmission channel constructed between the data feature analysis intelligent model and each data fusion element, so as to collect the original data corresponding to each data fusion element and let the data feature analysis intelligent model perform feature analysis on the original data.

[0036] More specifically, with the support of the intelligent model, the construction of the data fusion channel has good scalability. Even if the data fusion elements increase or change in the future, the channel system can still adapt to the new data acquisition requirements through model adjustment, ensuring the flexibility of the system.

[0037] Specifically, in step S4 of the embodiment provided by the present invention, through the already constructed data fusion channel, a complete set of element data is obtained from each data fusion element. These data come from different sources, including time series data, spatial data, sensor data, etc. Therefore, these data have different data formats and are obtained based on different data standards. In order to uniformly summarize these data with different formats and standards, it is necessary to use a data feature analysis intelligent model to extract features from each data, so as to obtain data features with consistent formats and standards.

[0038] More specifically, initialize the data feature analysis intelligent model, and set the parameters and analysis dimensions of the model according to the attributes of the element data set. The model can be a model based on machine learning algorithms, such as clustering analysis, principal component analysis (PCA), factor analysis (FA), neural network, etc., and select a suitable analysis model for specific data fusion elements to accurately extract key data features.

[0039] More specifically, the initialized data feature analysis intelligent model is used to analyze the element data set and extract key data features. This step involves dimensionality reduction of the high-dimensional features of the data to identify the most representative and influential features in the data. For example, the main components are extracted through PCA, and the clustering patterns of the data are identified through cluster analysis, etc. Key features with representativeness and decision-making significance are extracted from a large amount of raw data, reducing data redundancy and improving the efficiency and effectiveness of data analysis.

[0040] More specifically, the extracted key data features are sorted according to their importance to identify the key data that can best represent the core features of the element data set. Methods such as feature importance scoring and weight analysis can be used to evaluate the role of each feature in the overall data set.

[0041] More specifically, the extracted key data features are verified to ensure that these features can indeed represent the core information of the element data set. The verification methods can include cross-validation, model prediction accuracy evaluation, etc. According to the verification results, the parameters of the intelligent model are optimized to improve the model's analysis and feature extraction capabilities. Through verification and optimization, it is ensured that the extracted key features have high representativeness and high accuracy, enabling the model to work stably and efficiently in practical applications.

[0042] It can be understood that through feature analysis and extraction, the originally high-dimensional element data set is compressed into more concise and representative key features. This not only reduces the complexity of data storage and calculation but also improves the efficiency of subsequent data processing. Through the analysis of the intelligent model, the extracted key data features can more accurately reflect the core characteristics of the data fusion elements. This provides a more reliable and meaningful input for subsequent data analysis and decision-making.

[0043] Specifically, in step S5 of the embodiment provided by the present invention, the already extracted data key features are mapped to the corresponding modules of the data fusion target framework. The mapping process needs to consider the relevance between each feature and different functional modules in the target framework. Through feature mapping, each key feature is assigned to a suitable fusion logic and processing unit to ensure that the data features can effectively participate in the overall fusion process, realizing the seamless docking of the key features and the data fusion target framework and laying a foundation for subsequent fusion processing.

[0044] More specifically, according to the set rules, the fusion processing of heterogeneous data is performed in the data fusion target framework. The fusion processing may include steps such as multi-source data merging, feature weighting, conflict resolution, data calibration, etc. The key features of each data fusion element are fused into comprehensive information in this process, providing a basis for the final decision support. Through precise fusion algorithms and processing means, heterogeneous industrial data from different sources and of different types are fused into a consistent and available information set.

[0045] More specifically, the verified fusion data is integrated into the industrial system for further analysis, decision-making, and application. The integration process may include data storage, real-time application, report generation, etc. Ensure that the fusion results can effectively support the various functions and business requirements of the system, realize the landing application of the data fusion results, provide accurate and reliable data support for the industrial system, and improve the intelligence and automation level of the system.

[0046] It can be understood that through the mapping and fusion of the key features of the data, the integration problem of heterogeneous industrial data from different data sources and in different formats is solved. The fused data has consistency, integrity, and availability, and can support more complex analysis and decision-making. The fused data supports higher-level intelligent analysis and automated decision-making, bringing stronger intelligent capabilities to the industrial system. This includes key applications such as real-time monitoring, predictive maintenance, and intelligent scheduling.

[0047] The present invention provides a method for heterogeneous industrial data fusion based on artificial intelligence, having the following beneficial effects:

[0048] The present invention conducts a structural analysis of the data fusion target, obtains data structure information, constructs a data fusion target framework for data filling processing, constructs a data fusion channel, collects element data, extracts the key features of the data, substitutes the key features into the fusion target framework, and completes the heterogeneous data fusion, improving the fusion efficiency of heterogeneous industrial data, realizing automatic feature extraction and data filling processing, enhancing the intelligence and adaptability of the system, and solving the problem of low efficiency in processing large-scale multi-source heterogeneous data in the prior art.

[0049] Preferably, the step of performing a structural analysis of the data fusion target to obtain the data structure information of the fusion target according to several data fusion elements to be subjected to heterogeneous industrial data fusion includes:

[0050] S11: Obtain a historical data set of several of the data fusion elements to be subjected to heterogeneous industrial data fusion;

[0051] S12: Extract data features from the historical data sets of each of the data fusion elements through a pre-trained intelligent data feature extraction model, so as to obtain the element data features of each of the data fusion elements; wherein, the element data features are used to describe the data features of the data fed back by the data fusion elements;

[0052] S13: Conduct a commonality analysis on the element data features of each of the data fusion elements to obtain the common format distribution of each of the element data features; wherein, the common format distribution includes a number of common data features, and the common data features are the features of the data formats that can be converted through each of the element data features;

[0053] S14: Conduct a multi-dimensional analysis of commonality conversion on the element data features of each of the data fusion elements based on the common format distribution to obtain the common reference features of each of the data fusion elements with respect to each of the common data features of the common format distribution; wherein, the common reference features include common conversion fitness, common conversion confidence, and common conversion loss degree;

[0054] S15: Select the best common data feature in the common format distribution according to the common reference features of each of the data fusion elements with respect to each of the common data features of the common format distribution as the data structure information of the fusion target.

[0055] Specifically, first, extract historical data sets from each element to be subjected to heterogeneous industrial data fusion. These data sets usually include data in various formats and structures recorded in the past, which are used to reflect the data representation forms at different times and in different environments, providing a comprehensive historical data basis for subsequent data feature extraction and analysis.

[0056] More specifically, analyze the historical data sets of each data fusion element through a pre-trained intelligent data feature extraction model to extract element data features. These features include the core attributes of the data, such as statistical distribution, trend, periodicity, outliers, etc., which are used to describe the unique features of the data fed back by the elements, and extract representative and descriptive element data features from the historical data to facilitate subsequent commonality analysis and conversion.

[0057] More specifically, conduct a commonality analysis on the extracted element data features to identify and define the common format distribution of these features. The common format distribution refers to the standardized data formats that multiple element data features can be uniformly converted into. The key to this step is to find the commonalities between different element data features, thereby simplifying the subsequent data fusion process.

[0058] More specifically, based on the common format distribution, multi-dimensional analysis is performed on the feature data of each data fusion element to calculate its common reference features relative to the common format distribution. These common reference features include: common conversion fitness: measuring the degree of adaptation of the feature data of the element to the common data features; common conversion confidence: indicating the reliability and credibility of the conversion result; common conversion loss: evaluating the possible information loss during the conversion process. Through multi-dimensional analysis, the matching degree between each feature data of the element and the common data features is quantified, providing a basis for selecting the best common data features.

[0059] More specifically, according to the common reference features of each element relative to the common format distribution, the common data features that best represent the data structure of the fusion target are evaluated and selected. The selection process considers the features with the highest common conversion fitness, the highest common conversion confidence, and the lowest common conversion loss to ensure that the final data structure information is both accurate and effective. The best common data features are determined as the final data structure information, providing a unified structure standard for the fusion of heterogeneous industrial data.

[0060] It can be understood that through the establishment of commonality analysis and common format distribution, the standardization of different feature data of elements is achieved, which greatly improves the consistency of data during the data fusion process and avoids data fusion obstacles caused by inconsistent formats. Through multi-dimensional analysis and processing, the relationship between each feature data of the element and the common data features is accurately quantified, making the data fusion process not only accurate but also highly confident, thus improving the reliability of the fusion result. The application of the common format distribution simplifies the data fusion process and no longer requires dealing with the conversion problems between complex heterogeneous data, thus improving the overall fusion efficiency. During the common conversion process, by evaluating the conversion loss, the information loss is minimized, maintaining the high fidelity of the data and ensuring that the final data fusion result has high-quality information retention.

[0061] Preferably, the steps of constructing a data fusion target framework corresponding to each of the data fusion elements based on the data structure information include:

[0062] S21: Construct a corresponding first-layer data framework based on each of the data fusion elements; wherein, the first-layer data framework includes a first time coordinate axis and a number of element raw data axes sequentially arranged on the first time coordinate axis, and the element raw data axes are used for storing and processing the element data sets of each of the data fusion elements.

[0063] S22: Construct a second - layer data framework based on the data - structure information in the first - layer data framework. The second - layer data framework includes a second time coordinate axis and a number of element - feature data axes arranged in sequence on the second time coordinate axis. The element - feature data axes are used to store and process the key data features of each data - fusion element, and the data structure of the key data features stored in the element - feature data axes is consistent with the data - structure information.

[0064] S23: Construct a third - layer data framework based on the second - layer data framework. The third - layer data framework includes a feature - correlation analysis unit and a feature - change - trend analysis unit. The feature - correlation analysis unit is used to analyze the correlation between the key data features of each data - fusion element in the second - layer data framework, and the feature - change - trend analysis unit is used to analyze the change trend of the key data features of each data - fusion element in the second - layer data framework.

[0065] Specifically, based on each data - fusion element, construct a first - layer data framework, which includes a first time coordinate axis and a number of element - original data axes arranged in sequence on this time coordinate axis. Each element - original data axis corresponds to a data - fusion element and is used to store the original data set of this element. These data axes are arranged along the time axis, reflecting the evolution of data over time, providing a structured storage mechanism for saving and managing the historical original data of each data - fusion element, and providing a basis for subsequent data processing.

[0066] More specifically, based on the first - layer data framework, use the data - structure information to construct a second - layer data framework, which includes a second time coordinate axis and a number of element - feature data axes arranged in sequence on this time coordinate axis. Each element - feature data axis is used to store the key data features of the corresponding data - fusion element, and the data structure of these features is consistent with the pre - determined data - structure information. Based on the time axis, systematically store the key data features of each data - fusion element, providing an organized data structure for higher - level analysis.

[0067] More specifically, on top of the second-layer data framework, a third-layer data framework is constructed, which includes two main units: Feature Relevance Analysis Unit: Conducts relevance analysis on the key data features of each data fusion element in the second-layer data framework, identifies and quantifies the relationships between different elements, and these relationships may include linear correlation, non-linear correlation, temporal correlation, etc. Feature Change Trend Analysis Unit: Analyzes the change trends of the key data features of each data fusion element over time to help identify the long-term change patterns or short-term fluctuation rules of the features. Through the relevance and trend analysis of the key data features, the internal connections and evolution laws between the data fusion elements are discovered, providing important references for further decision-making support.

[0068] It can be understood that through the construction of the three-layer data framework, a hierarchical structure from raw data to key features and then to high-level analysis is achieved, making the storage, management, and analysis of data more systematic and organized. Each layer of the data framework has a clear functional division, enabling the system to execute more efficiently during data storage, feature extraction, and correlation analysis, reducing unnecessary calculations and data redundancy. The relevance and trend analysis units in the third-layer data framework can handle complex multi-dimensional data analysis, helping to identify the deep relationships and dynamic changes between data fusion elements and providing more comprehensive analysis results for the system. This hierarchical data framework design has strong scalability and can flexibly add new data axes or analysis units according to new data fusion elements or new analysis requirements, thus continuously enriching and optimizing the data fusion target framework. By storing and managing the key data features with a unified structure in the second-layer data framework, the data consistency during the analysis process is ensured, thereby improving the accuracy and reliability of data fusion. The analysis units in the third-layer data framework can generate detailed relevance and trend reports, providing accurate decision-making support for managers and decision-making systems and facilitating the optimization and upgrade of industrial systems.

[0069] Preferably, based on the data fusion target framework, the steps of constructing corresponding data fusion channels for each of the data fusion elements using a data feature analysis intelligent model include:

[0070] S31: Pre-train a data feature analysis intelligent model to obtain the data feature analysis model that has completed model training;

[0071] S32: Obtain the data sources of each of the data fusion elements, and construct a data acquisition unit based on the data sources of each of the data fusion elements to collect data on the data fusion elements through the data acquisition unit, thereby obtaining the element data set of the data fusion elements;

[0072] S33: Deploy the intelligent model for data feature analysis on the data fusion target framework, and connect the intelligent model for data feature analysis to the data fusion target framework for data connection;

[0073] S34: Connect the intelligent model for data feature analysis to each of the data acquisition units, so that the intelligent model for data feature analysis can extract data key features from the set of elemental data obtained from the data acquisition units, and transmit the data key features to the corresponding positions on the data fusion target framework.

[0074] Specifically, first, pre-train the intelligent model for data feature analysis. By using a large-scale historical data set and labeled data, the model learns various data feature extraction methods and pattern recognition algorithms to accurately extract key data features. After training, a high-precision data feature analysis model that can be effectively applied to actual data is obtained, ensuring that the intelligent model for data feature analysis has high precision and high robustness and can accurately identify and extract various data features.

[0075] More specifically, obtain the specific data sources of each data fusion element (such as sensors, databases, APIs, etc.), and build corresponding data acquisition units based on these sources. The data acquisition units are responsible for continuously collecting and processing data from the data sources, converting the raw data into a structured set of elemental data, and establishing an effective data acquisition channel for each data fusion element to ensure the real-time and accuracy of data acquisition.

[0076] More specifically, deploy the trained intelligent model for data feature analysis into the data fusion target framework. During the deployment process, ensure that the intelligent model can be seamlessly connected to the data fusion target framework, enabling the framework to call the model for data feature analysis and extraction, endowing the data fusion target framework with automated data feature analysis capabilities, and enhancing the intelligence level of the system.

[0077] More specifically, connect the data acquisition units to the intelligent model for data feature analysis. Through this connection, the model can obtain real-time or batch sets of elemental data from each data acquisition unit, extract features from these data, and transmit the extracted data key features to the corresponding positions in the data fusion target framework for storage and further processing, establishing an automated process among data acquisition, feature extraction, and storage to ensure the efficiency and stability of data flow.

[0078] More specifically, the key data features extracted from the intelligent model are stored on the corresponding data axes of the data fusion target framework. These feature data can then be used for advanced data analysis tasks such as correlation analysis and trend prediction, ensuring the efficient storage and management of key data features and providing a high-quality data foundation for further decision support and system optimization.

[0079] It can be understood that by introducing the intelligent model for data feature analysis, the system can automatically extract key features from the original data, reducing manual intervention and improving the automation level of data processing. The efficient connection between the data acquisition unit and the intelligent model ensures the real-time acquisition and processing of data, enabling the system to quickly respond and accurately extract key features to meet the needs of real-time analysis and decision-making.

[0080] Preferably, the steps of pre-training the intelligent model for data feature analysis and jointly deploying the trained intelligent model for data feature analysis with the data fusion target framework include:

[0081] S311: Prepare a convolutional neural network model, a recurrent neural network model, and a training data set; wherein, the training data set includes a visual data type set and a sequence data type set; wherein, the visual data type set includes visual training data of data fusion elements corresponding to the visual data type in the data format and first target training data corresponding to the data key features, and the sequence data type set includes sequence inspection data of data fusion elements corresponding to the sequence data type in the data format and second target training data corresponding to the data key features. The data formats of the first target training data and the second target training data are both consistent with the data structure information;

[0082] S312: Substitute the visual data type set into the convolutional neural network model, and let the convolutional neural network model perform model training according to the visual data type set, so that the convolutional neural network model can analyze and obtain data key features corresponding to the data structure information in the data format according to the element data set;

[0083] S313: Substitute the sequence data type set into the recurrent neural network model, and let the recurrent neural network model perform model training according to the sequence data type set, so that the recurrent neural network model can analyze and obtain data key features corresponding to the data structure information in the data format according to the element data set;

[0084] S314: Construct a data type discrimination model based on the convolutional neural network model and the recurrent neural network model. The data type discrimination model is used to discriminate the data format of the data fusion elements, and deliver the element data set of the data fusion elements to the convolutional neural network model or the recurrent neural network model according to the discrimination result;

[0085] S315: Jointly use the convolutional neural network model, the recurrent neural network model, and the data type discrimination model that have completed model training as the data feature analysis intelligent model.

[0086] Specifically, first, select and initialize the convolutional neural network model and the recurrent neural network model, and prepare the corresponding training data set. The training data set should include two types of data: Visual data type set: Training data of data fusion elements containing visual data types, which are usually in the form of images, videos, or other two-dimensional data, along with the corresponding first target training data, that is, the key data features that have been labeled; Sequence data type set: Training data of data fusion elements containing sequence data types, which are usually in the form of time series, text, or other one-dimensional data, along with the corresponding second target training data, that is, the key data features that have been labeled, to ensure that the data required for model training covers different types of features, and the format of these data is consistent with the data structure information required in the final application.

[0087] More specifically, substitute the visual data type set into the convolutional neural network model for training. Through multiple iterations of optimization, the model will learn how to extract key features corresponding to the data structure information from the visual data. Finally, the trained convolutional neural network model can automatically extract the corresponding key features when new visual data types are input, obtaining a convolutional neural network model that can efficiently process visual data types, ensuring that it can accurately extract the key features in the visual data.

[0088] More specifically, substitute the sequence data type set into the recurrent neural network model for training. The model learns how to extract key features corresponding to the data structure information from the sequence data through iterative training. Finally, the trained recurrent neural network model can automatically extract the corresponding key features when new sequence data is input, obtaining a recurrent neural network model that can efficiently process sequence data types, ensuring that it can accurately extract the key features in the sequence data.

[0089] More specifically, based on a convolutional neural network model and a recurrent neural network model, a data type discrimination model is constructed. This model is used to determine the data format (visual data or sequential data) of the input data fusion elements, and deliver the element data set to the corresponding model for processing according to the discrimination result, ensuring that the system can automatically select an appropriate model based on the data type, and optimizing the accuracy and efficiency of data feature extraction.

[0090] More specifically, the trained convolutional neural network model, recurrent neural network model, and data type discrimination model are jointly used as a data feature analysis intelligent model, and are deployed into the data fusion target framework. These models will function as a whole to automatically process the input element data set, extract key data features, and transmit them to the corresponding positions in the data fusion target framework. By integrating multiple models and jointly deploying them with the data fusion target framework, intelligent and automated data feature extraction and fusion are achieved.

[0091] It can be understood that through models specifically trained for different data types (visual data and sequential data), the system can extract key features more accurately, reducing errors caused by data type mismatches. The convolutional neural network model and the recurrent neural network model are trained for different types of data respectively, ensuring the generalization ability of the models and enabling them to adapt to inputs of multiple data formats. The data type discrimination model automatically judges the data format and selects an appropriate model for processing, avoiding unnecessary waste of computing resources and improving the overall computing efficiency of the system. By jointly deploying the trained models with the data fusion target framework, the system can automatically complete the entire process from data acquisition to feature extraction and then to data fusion, reducing the need for human intervention.

[0092] Preferably, the steps of obtaining an element data set from the data fusion elements through the data fusion channel and enabling the data feature analysis intelligent model to perform key data extraction on the element data set to obtain the data key features corresponding to the element data set include:

[0093] S41: Obtain an element data set from the data fusion elements through the data fusion channel;

[0094] S42: Perform data format judgment processing on the element data set according to the data type discrimination model of the data feature analysis intelligent model, and allocate the element data set to the convolutional neural network model or the recurrent neural network model for processing according to the result of the judgment processing;

[0095] S43: Perform key data extraction on the element data set through the convolutional neural network model or the recurrent neural network model to obtain the data key features corresponding to the element data set.

[0096] Specifically, through pre-constructed data fusion channels, element data sets from different data fusion elements are obtained in real-time or in batches. These data may include visual data (such as images, videos), sequence data (such as time series, text), etc., covering various data formats, ensuring the timeliness and integrity of data collection, and providing a sufficient data basis for subsequent data processing and feature extraction.

[0097] More specifically, the data type discrimination model in the data feature analysis intelligent model is used to judge and process the data format of the obtained element data set. The data type discrimination model will automatically judge whether the data belongs to the visual data type or the sequence data type according to the structure, dimension and content of the data, and allocate the data set to the corresponding neural network model (convolutional neural network or recurrent neural network) for processing. Through automated data format discrimination, it is ensured that the data set can be sent to the most suitable model for feature extraction, optimizing the accuracy and efficiency of the processing flow.

[0098] More specifically, according to the data format discrimination results, the element data sets are sent to the convolutional neural network model or the recurrent neural network model for processing respectively: Visual data type set: Sent to the convolutional neural network model, which will use convolutional operations to extract key features in visual data such as images and videos. Sequence data type set: Sent to the recurrent neural network model, which analyzes data such as time series and text recursively to extract key features corresponding to the data structure information. Through in-depth analysis of the data set by a specific model, valuable key data features are extracted for subsequent fusion and decision support.

[0099] More specifically, the key features extracted from the convolutional neural network model or the recurrent neural network model are summarized to form the complete data key features corresponding to the element data set. These key features will be transmitted to the corresponding positions in the data fusion target framework for further data fusion, analysis and decision support, generating accurate key data features to ensure that the system can use these features to achieve more intelligent analysis and decision-making during the data fusion process.

[0100] It can be understood that through the automated processing of the data type discrimination model, the processes of data format judgment and feature extraction are seamlessly connected, reducing manual intervention and enhancing the automation level of the system. The convolutional neural network model and the recurrent neural network model respectively extract features for different types of data, ensuring the efficiency and accuracy of processing, and can quickly generate key data features of value. The system can automatically identify different types of data and select the best processing path, enhancing the intelligence level of the data fusion process. Especially when dealing with multi-source heterogeneous data, it can achieve more efficient fusion and analysis. Through in-depth feature extraction for different data types, the system can not only obtain more comprehensive data information but also deeply explore the hidden features in the data, providing a solid foundation for subsequent advanced data analysis and decision support. This way of multi-model collaborative work enables the system to flexibly respond to changes in different data types, adapt to various application scenarios and business requirements, and enhance the versatility and scalability of the system.

[0101] Preferably, the step of substituting the data key features into the data fusion target framework to complete the heterogeneous industrial data fusion of each data fusion element includes:

[0102] S51: Substitute the data key features into the second-layer data framework of the data fusion target framework, and let the second-layer data framework store and process the data key features of each data fusion element;

[0103] S52: Let the third-layer data framework of the data fusion target framework perform in-depth analysis and processing on the second-layer data framework to obtain the data correlation features analyzed by the feature correlation analysis unit, obtain the change trend features analyzed by the feature change trend analysis unit, and let the third-layer data framework store and process the data correlation features and the change trend features.

[0104] Specifically, the data key features that have been extracted by the convolutional neural network and the recurrent neural network model are imported into the second-layer data framework of the data fusion target framework. The function of the second-layer data framework is to store and manage the data key features of each data fusion element (such as different types of industrial equipment or sensor data), ensuring that the key features of each industrial data fusion element are systematically stored so that subsequent processing can quickly and accurately access and utilize these features.

[0105] More specifically, in the second-layer data framework, all received key features of data from different sources are stored and processed. During the storage and processing, the format of the data is ensured to be unified and the structure is clear, so that the subsequent processing and analysis modules can be smoothly connected, and a reliable data storage mechanism is established to ensure the unified management of the features of multi-source heterogeneous data and reduce the risk of inconsistent data storage or data loss.

[0106] More specifically, in the third-layer data framework, in-depth analysis is performed on the key features stored in the second-layer data framework. The third-layer data framework mainly includes two analysis units: Feature correlation analysis unit: Analyze the correlation features between different data fusion elements. By mining the relationships between the key features of the data, identify the potential associations between various elements, such as the collaborative relationship between devices or the mutual influence of sensor data; Feature change trend analysis unit: Analyze the change trends of data features, identify the trends of key features changing over time or operating conditions, such as the gradual decline of device performance or the significant fluctuations of certain variables during the production process. Through in-depth analysis, extract the correlation features and change trend features between different industrial data, providing a data basis for subsequent decision support and predictive analysis.

[0107] More specifically, store and process the data correlation features and change trend features analyzed in the third-layer data framework. These features will serve as the basis for further analysis and decision support, ensuring the continuity and integrity of the data, and ensuring that the data analysis results are systematically saved so that these data can be quickly accessed and used to provide support for decision-making when needed.

[0108] It can be understood that by importing the key features of multi-source heterogeneous industrial data into a unified data framework, the efficient fusion of industrial data is achieved. This fusion can establish connections between multiple different data sources, ensuring the data interoperability and consistency in the entire industrial system. The second-layer data framework provides a unified data storage mechanism to ensure the reliable storage and management of multi-source data in the system, reducing the risk of data loss or inconsistent format and ensuring the integrity of the data. Through the in-depth analysis function of the third-layer data framework, the system can extract the correlation features and change trends between the data. The mining of these features not only helps to understand the internal relationships of the data but can also be used to predict potential problems in the system or optimize industrial processes.

[0109] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for heterogeneous industrial data fusion based on artificial intelligence, characterized in that, Including: Performing a structural analysis of the data fusion objective based on several data fusion elements for heterogeneous industrial data fusion to obtain the data structure information of the fusion objective; wherein, the data fusion element is a data source participating in the heterogeneous industrial data fusion. Based on the data structure information, constructing a data fusion objective framework corresponding to each of the data fusion elements; wherein, the data fusion objective framework is used for data filling processing after heterogeneous industrial data fusion by each of the data fusion elements. On the basis of the data fusion objective framework, using a data feature analysis intelligent model to construct corresponding data fusion channels for each of the data fusion elements; wherein, the data fusion channel is used for data collection of the data fusion element. Obtaining an element data set from the data fusion element through the data fusion channel, and enabling the data feature analysis intelligent model to extract key data from the element data set to obtain the data key features corresponding to the element data set. Substituting the data key features into the data fusion objective framework to complete the heterogeneous industrial data fusion of each of the data fusion elements. The step of constructing a data fusion objective framework corresponding to each of the data fusion elements based on the data structure information includes: Constructing a corresponding first-layer data framework based on each of the data fusion elements; wherein, the first-layer data framework includes a first time axis and several element raw data axes sequentially arranged on the first time axis, and the element raw data axis is used for storing and processing the element data set of each of the data fusion elements. Constructing a second-layer data framework on the basis of the data structure information in the first-layer data framework; wherein, the second-layer data framework includes a second time axis and several element feature data axes sequentially arranged on the second time axis, the element feature data axis is used for storing and processing the key data features of each of the data fusion elements, and the data structure of the key data features stored in the element feature data axis is consistent with the data structure information. Constructing a third-layer data framework on the basis of the second-layer data framework; wherein, the third-layer data framework includes a feature correlation analysis unit and a feature change trend analysis unit, the feature correlation analysis unit is used for analyzing the correlation between the key data features of each of the data fusion elements in the second-layer data framework, and the feature change trend analysis unit is used for analyzing the change trend of the key data features of each of the data fusion elements in the second-layer data framework.

2. The method for heterogeneous industrial data fusion based on artificial intelligence according to claim 1, wherein, The step of performing a structural analysis of the data fusion objective based on several data fusion elements for heterogeneous industrial data fusion to obtain the data structure information of the fusion objective includes: Obtaining a historical data set of several of the data fusion elements for heterogeneous industrial data fusion. Through a pre-trained data feature extraction intelligent model, perform data feature extraction processing on the historical data sets of each of the data fusion elements to obtain the element data features of each of the data fusion elements; wherein, the element data features are used to describe the data features of the data fed back by the data fusion elements. Perform commonality analysis processing on the element data features of each of the data fusion elements to obtain the common format distribution of each of the element data features; wherein, the common format distribution includes a number of common data features, and the common data features are the features of the data formats that can be converted through each of the element data features. Perform multi-dimensional analysis processing of commonality conversion on the element data features of each of the data fusion elements based on the common format distribution to obtain the common reference features of each of the data fusion elements with respect to each of the common data features of the common format distribution; wherein, the common reference features include common conversion fitness, common conversion confidence, and common conversion loss degree. According to the common reference features of each of the data fusion elements with respect to each of the common data features of the common format distribution, select the best of the common data features in the common format distribution as the data structure information of the fusion target.

3. The method for heterogeneous industrial data fusion based on artificial intelligence according to claim 1, wherein On the basis of the data fusion target framework, the steps of constructing corresponding data fusion channels for each of the data fusion elements by using a data feature analysis intelligent model include: Pre-train a data feature analysis intelligent model to obtain the data feature analysis model that has completed model training. Obtain the data sources of each of the data fusion elements, and construct a data acquisition unit based on the data sources of each of the data fusion elements to collect data on the data fusion elements through the data acquisition unit, so as to obtain the element data sets of the data fusion elements. Deploy the data feature analysis intelligent model on the data fusion target framework, and make the data feature analysis intelligent model be data-connected to the data fusion target framework. Connect the data feature analysis intelligent model to each of the data acquisition units so that the data feature analysis intelligent model can extract data key features from the element data sets obtained by the data acquisition units and transmit the data key features to the corresponding positions on the data fusion target framework.

4. The method for heterogeneous industrial data fusion based on artificial intelligence according to claim 3, characterized in that, The steps of pre-training a data feature analysis intelligent model and jointly deploying the data feature analysis model that has completed training with the data fusion target framework include: Prepare a convolutional neural network model, a recurrent neural network model, and a training data set; wherein, the training data set includes a visual data type set and a sequence data type set; wherein, the visual data type set includes visual training data of data fusion elements corresponding to the visual data type in terms of data format and first target training data corresponding to the key data features, and the sequence data type set includes sequence inspection data of data fusion elements corresponding to the sequence data type in terms of data format and second target training data corresponding to the key data features, and the data formats of the first target training data and the second target training data are both consistent with the data structure information; Substitute the visual data type set into the convolutional neural network model, and let the convolutional neural network model perform model training according to the visual data type set, so that the convolutional neural network model can analyze and obtain key data features corresponding to the data structure information in terms of data format based on the element data set; Substitute the sequence data type set into the recurrent neural network model, and let the recurrent neural network model perform model training according to the sequence data type set, so that the recurrent neural network model can analyze and obtain key data features corresponding to the data structure information in terms of data format based on the element data set; Construct a data type discrimination model based on the convolutional neural network model and the recurrent neural network model. The data type discrimination model is used to discriminate the data format of the data fusion elements, and convey the element data set of the data fusion elements to the convolutional neural network model or the recurrent neural network model according to the discrimination result; Collectively use the convolutional neural network model, the recurrent neural network model, and the data type discrimination model that have completed model training as the data feature analysis intelligent model.

5. The method for heterogeneous industrial data fusion based on artificial intelligence according to claim 4, characterized in that, The steps of obtaining an element data set from the data fusion elements through the data fusion channel and letting the data feature analysis intelligent model extract key data from the element data set to obtain key data features corresponding to the element data set include: Obtain an element data set from the data fusion elements through the data fusion channel; Perform judgment processing on the data format of the element data set according to the data type discrimination model of the data feature analysis intelligent model, and allocate the element data set to the convolutional neural network model or the recurrent neural network model for processing according to the result of the judgment processing; Extract key data from the element data set through the convolutional neural network model or the recurrent neural network model to obtain key data features corresponding to the element data set.

6. The method for heterogeneous industrial data fusion based on artificial intelligence according to claim 1, wherein The steps of substituting the key data features into the data fusion target framework to complete the heterogeneous industrial data fusion of each data fusion element include: Substitute the key data features into the second-layer data framework of the data fusion target framework, and let the second-layer data framework perform storage processing on the key data features of each data fusion element; Let the third-layer data framework of the data fusion target framework perform in-depth analysis and processing on the second-layer data framework to obtain the data association features analyzed by the feature correlation analysis unit, obtain the change trend features analyzed by the feature change trend analysis unit, and store and process the data association features and the change trend features by the third-layer data framework.

Images