An assembly gap optimization system for knowledge base and data model fusion analysis

Abstract

The application belongs to the technical field of intelligent manufacturing, and discloses an assembly gap optimization system based on knowledge base and data model fusion analysis. The system realizes accurate prediction of the gap parameters in the product assembly process by coupling the model and the knowledge base, converting and fusing the knowledge base data and the data model, and optimizing the assembly gap prediction model. The system identifies the assembly gap that needs to be calculated, matches the model similar to the current assembly part from the model and knowledge base coupling, predicts the assembly gap value under different parameters, and outputs the optimal parameter configuration suggestion and the assembly adjustment gage size scheme.

Description

Technical Field

[0001] This invention belongs to the field of intelligent manufacturing technology, and in particular relates to an assembly gap optimization system that integrates knowledge base and data model analysis. Background Technology

[0002] Traditional assembly clearance optimization mainly relies on manual experience, trial and error, and simple statistical analysis. Specific steps typically include: manual measurement, using tooling or measuring instruments to measure part dimensions and clearances; experience-based adjustments, adjusting part positions and using shims based on worker experience; iterative testing, repeatedly adjusting and testing to achieve the target clearance value; and simple data analysis, performing statistical analysis based on limited historical data to optimize parameter configurations.

[0003] In modern manufacturing, assembly clearance is a crucial indicator of product quality and performance. Traditional methods for optimizing assembly clearance primarily rely on manual experience or single data analysis techniques, resulting in low efficiency and poor accuracy. With the development of Industry 4.0 and artificial intelligence, how to combine knowledge bases and model analysis techniques to achieve efficient and intelligent assembly clearance optimization has become an urgent problem to be solved. Summary of the Invention

[0004] The technical problem this invention aims to solve is: how to combine knowledge base and model analysis technology to achieve efficient and intelligent assembly gap optimization.

[0005] To solve the above-mentioned technical problems, the specific technical solution of the present invention is as follows: An assembly gap optimization system that integrates knowledge base and data model analysis includes a data acquisition module, a data preprocessing module, a data association module, a knowledge base module, a feature engineering module, a model training module, a model-knowledge base coupling module, and an assembly gap optimization module. The multi-source data acquisition module is used to acquire assembly parameters in real time. After the preprocessing module cleans and standardizes the data, it is transmitted to the data association module to establish a logical mapping and form structured data. The associated structured data is then processed by the feature engineering module to extract key features and optimize their distribution before being sent to the model training module to build a prediction model. At the same time, the process experience stored in the knowledge base module is embedded into a knowledge graph through the encoder of the coupling module. The decoder of the coupling module feeds back the knowledge association information to the model training module, realizing data-knowledge dual-driven optimization. The optimization module calculates gap parameters based on the coupling module's matching of historical cases and real-time assembly data, and generates personalized adjustment suggestions.

[0006] Furthermore, the data acquisition module collects assembly parameters in real time through various sensors, cameras, and related interface devices.

[0007] Furthermore, the data preprocessing module performs integrity checks and anomaly detection on the input assembly parameter set, then removes noise interference and retains valid information; finally, it transforms features of different dimensions into a consistent expression form.

[0008] Furthermore, the data association module is used to extract key feature information from each data source and construct logical mapping relationships between data based on preset association rules.

[0009] Furthermore, the feature engineering module is used to extract key size deviation features, then perform wavelet transform for noise reduction, and finally optimize the data feature distribution to construct feature vectors.

[0010] Furthermore, the knowledge base module is used for data management and knowledge integration in industrial manufacturing scenarios.

[0011] Furthermore, the coupling module is used to transform and integrate knowledge base data and data models.

[0012] This invention offers the following advantages: By employing multi-source data acquisition, deep feature extraction, and knowledge-driven model coupling, it achieves accurate prediction and optimization of assembly gaps, improving the accuracy of assembly gap prediction and reducing the need for manual experience and repeated trials. It also enables automated optimization of the assembly process, improving production efficiency and product quality. Attached Figure Description

[0013] Figure 1 This is a flowchart illustrating the assembly gap optimization system based on the fusion analysis of the knowledge base and data model of this invention. Detailed Implementation

[0014] To better understand the purpose, structure, and function of this invention, the invention will be described in further detail below with reference to the accompanying drawings.

[0015] like Figure 1 As shown, the assembly gap optimization system of the present invention, which integrates knowledge base and data model analysis, mainly consists of a data acquisition module, a data preprocessing module, a data association module, a knowledge base module, a feature engineering module, a model training module, a model and knowledge base coupling module, and an assembly gap optimization module.

[0016] The system acquires assembly process parameters in real time through a multi-source data acquisition module. After data cleaning and standardization by a preprocessing module, the data is transmitted to a data association module to establish a logical mapping. The associated structured data is then processed by a feature engineering module to extract key features and optimize their distribution before being fed into a model training module to build a predictive model. Simultaneously, the process experience stored in the knowledge base module is embedded into a knowledge graph through the encoder of the coupling module, and the decoder feeds back the knowledge association information to the model training, achieving data-knowledge dual-driven optimization. Finally, the optimization module calculates gap parameters based on the coupled model, matching historical cases and combining them with real-time assembly data, to generate personalized adjustment suggestions.

[0017] The data acquisition module's main function is to achieve real-time acquisition and integration of multi-source data during the front drive assembly process. This module is an integrated multi-source data acquisition component, equipped with various sensors, cameras, and related interface devices, possessing powerful real-time data acquisition and processing capabilities. This module effectively improves the digitalization level of the assembly process, providing reliable data support for intelligent manufacturing.

[0018] The data preprocessing module is used to systematically and standardizedly clean and optimize the assembled dataset. Specifically, the module includes the following core technical steps: First, it performs integrity checks and anomaly detection on the input data; second, it removes noise interference and retains valid information; finally, it transforms features of different dimensions into a consistent expression. This module can significantly improve data quality, eliminate potential sources of bias, and ensure the accuracy and stability of subsequent analysis tasks (such as regression modeling and classification prediction), providing a solid preliminary guarantee for intelligent decision-making.

[0019] The data association module is used to establish relationships between different data sources. This module can extract key feature information from each data source and construct logical mapping relationships between data based on preset association rules. This module has high scalability and flexibility, and is compatible with various mainstream databases and unstructured data formats. It supports visual display and dynamic adjustment of association relationships, and can adapt to changes in business needs.

[0020] The feature engineering module is used to extract key features and construct feature vectors. This module includes: first, extracting key size deviation features; then, performing wavelet transform for denoising; and finally, optimizing the data feature distribution. This module significantly improves model training efficiency and prediction accuracy.

[0021] The model training module is used to build an assembly gap prediction model to predict gap parameters during product assembly. During training, the module performs feature engineering and optimization adjustments on the input data to improve the accuracy and robustness of the prediction results. This module can efficiently handle complex nonlinear relationships and establish a high-precision prediction model.

[0022] The knowledge base module is used for data management and knowledge integration in industrial manufacturing scenarios. This module supports the storage and management of various production parameters, processing procedures, and performance indicators, building an intelligent data management and knowledge sharing platform. It consolidates processing, assembly, and testing processes and parameter knowledge into a historical knowledge base, thus forming reusable knowledge assets. This module significantly improves data utilization efficiency in industrial manufacturing processes, laying a solid foundation for enterprises to achieve intelligent production and quality optimization.

[0023] The model-knowledge base coupling module transforms and integrates knowledge base data and data models to optimize the assembly gap prediction model, enabling accurate prediction of gap parameters during product assembly. This module includes an encoder and a decoder. The encoder embeds the knowledge graph, while the decoder enhances the accuracy and diversity of model recommendations by using associated information from the knowledge base. This module provides technical support for industrial assembly processes, helping enterprises achieve intelligent and automated optimization of production processes, thereby improving product quality, reducing rework and scrap rates, minimizing resource waste, and significantly increasing production efficiency.

[0024] The assembly clearance optimization module outputs optimal assembly parameters and adjustment suggestions. This module includes a data recognition subsystem, a parameter calculation subsystem, and an adjustment suggestion generation subsystem. This module provides solutions to assembly clearance problems caused by factors such as part dimensional deviations and fit relationships during complex mechanical assembly processes.

[0025] The data analysis subsystem is used to identify the assembly gap that needs to be calculated.

[0026] The parameter calculation subsystem is used to match and calculate models similar to the currently assembled parts from the coupling of the model and the knowledge base, and calculate the optimal assembly gap parameters.

[0027] The aforementioned adjustment suggestion generation subsystem is used to calculate the subsystem's output based on parameters, select the most suitable assembly pad size scheme, and automatically generate specific suggestions for modifying part dimensions, changing assembly sequence, or other optimization measures to recommend to the user.

[0028] The assembly clearance optimization system, which combines knowledge base and model analysis, provides a system workflow, such as... Figure 1 As shown, it includes the following steps: S1. Data Acquisition and Preprocessing By configuring various sensors, cameras, and related interface devices, it is possible to simultaneously acquire mechanical parameters (such as torque, angle, and displacement), image information, environmental data (temperature, humidity, etc.), and quality results during the assembly process, forming a complete assembly dataset. The collected data is efficiently transmitted to a database, and through a multi-step combination method, efficient and reliable data cleaning, noise reduction, and feature standardization functions are achieved.

[0029] S2. Data Linking Step Extract key information from each data source of the collected multi-source data, construct logical mapping relationships between data, form a complete data chain, establish correlation relationships between different data sources, and establish a link between isolated data that were originally scattered in different systems, forming a unified data management view.

[0030] S3. Feature Engineering Steps To analyze the correlated data and create new features or transform existing features, we first need to extract key size deviation features, then perform wavelet transform for noise reduction, and finally optimize the data feature distribution.

[0031] S4. Model Training Phase By optimizing the distribution of data features and using deep learning algorithms (such as neural networks and random forests) to train feature vectors, and performing feature engineering and optimization on the input data, complex nonlinear relationships can be efficiently processed and high-precision prediction models can be established.

[0032] S5. Knowledge Base Phase Processing, assembly, and testing techniques and related knowledge are accumulated into a historical knowledge base, thereby forming reusable knowledge assets. Historical production clearance tolerance standards and process specifications can be accessed at any time.

[0033] S6. Model-Knowledge Base Coupling Link Transforming and integrating knowledge base data and data models, and incorporating the knowledge base into the model, can enhance the model's understanding of complex concepts. Embedding rules and experiences from the knowledge base into the model enables synergistic optimization through data-driven and knowledge-driven approaches. By employing a multi-model fusion strategy, the assembly gap prediction model can be optimized to achieve accurate prediction of gap parameters during product assembly.

[0034] S7. Assembly Clearance Optimization When new critical components need to be assembled, users can identify the assembly gap that needs to be calculated based on the part size information used in the current assembly and the trained model. The model is matched and calculated from the coupling of the model and the knowledge base to predict the assembly gap value under different parameters. The optimal parameter configuration suggestion and assembly adjustment pad size scheme are output, and specific suggestions for modifying part size, changing assembly sequence or other optimization measures are automatically generated and recommended to the user.

[0035] This embodiment of the assembly gap optimization system, which integrates knowledge base and data model analysis, couples the model with the knowledge base to transform and fuse the knowledge base data and the data model, optimizing the assembly gap prediction model to achieve accurate prediction of gap parameters during product assembly. By identifying the assembly gap that needs to be calculated, it matches and calculates models similar to the currently assembled parts from the coupled model and knowledge base, predicts assembly gap values ​​under different parameters, and outputs optimal parameter configuration suggestions and assembly adjustment shim size schemes.

[0036] Although embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art will be able to make various modifications and improvements without departing from the principles of the present invention, and these modifications and improvements should also be considered to fall within the scope of protection of the present invention.

Claims

1. A system for optimizing assembly clearances by integrating knowledge base and data model analysis, characterized in that, It includes a data acquisition module, a data preprocessing module, a data association module, a knowledge base module, a feature engineering module, a model training module, a model-knowledge base coupling module, and an assembly gap optimization module; The multi-source data acquisition module is used to acquire assembly parameters in real time. After the preprocessing module cleans and standardizes the data, it is transmitted to the data association module to establish a logical mapping and form structured data. The associated structured data is then processed by the feature engineering module to extract key features and optimize the distribution before being sent to the model training module to build a prediction model. At the same time, the process experience stored in the knowledge base module is embedded into the knowledge graph through the encoder of the coupling module. The decoder of the coupling module feeds back the knowledge association information to the model training module, realizing data-knowledge dual-driven optimization. The optimization module matches historical cases with the coupled module and calculates gap parameters based on real-time assembly data to generate personalized adjustment suggestions.

2. The assembly gap optimization system based on the fusion analysis of knowledge base and data model according to claim 1, characterized in that, The data acquisition module collects assembly parameters in real time through various sensors, cameras, and related interface devices.

3. The assembly gap optimization system based on the fusion analysis of knowledge base and data model according to claim 1, characterized in that, The data preprocessing module is used to perform integrity checks and anomaly detection on the input assembly parameter set, then remove noise interference and retain valid information; finally, the features of different dimensions are transformed into a consistent expression form.

4. The assembly gap optimization system based on the fusion analysis of knowledge base and data model according to claim 1, characterized in that, The data association module is used to extract key feature information from various data sources and construct logical mapping relationships between data based on preset association rules.

5. The assembly gap optimization system based on the fusion analysis of knowledge base and data model according to claim 1, characterized in that, The feature engineering module is used to extract key size deviation features, then perform wavelet transform for noise reduction, and finally optimize the data feature distribution to construct feature vectors.

6. The assembly gap optimization system based on the fusion analysis of knowledge base and data model according to claim 1, characterized in that, The knowledge base module is used for data management and knowledge integration in industrial manufacturing scenarios.

7. The assembly gap optimization system based on the fusion analysis of knowledge base and data model according to claim 1, characterized in that, The coupling module is used to transform and integrate knowledge base data and data models.

Images