Method and platform for assembling pole-mounted circuit breaker

Abstract

The invention discloses an assembly method and platform for a pole-mounted circuit breaker, and relates to the technical field of low-voltage electric appliances, and the method comprises the steps: obtaining parts and a connection relation thereof, carrying out the assembly neighborhood recognition, sequentially assembling the parts according to a preset assembly scheme, synchronously monitoring the parts and the neighborhood thereof in the assembly process, generating an assembly monitoring data set, and carrying out the recognition of the assembly neighborhood. The method comprises the following steps: extracting monitoring data and carrying out iterative correlation analysis, identifying an assembly reliability factor, if the part reliability factor is lower than a threshold value, adding the part reliability factor to an abnormal part set, and generating an assembly early warning instruction. The technical problems that in the prior art, global and sequential analysis of the incidence relation between the pole-mounted circuit breaker parts is lacked, real-time deviation correction cannot be conducted in the assembling process, and the assembling efficiency and reliability are affected are solved, the potential assembling problem is recognized in real time in the assembling process, deviation correction is conducted in time, and the assembling efficiency and reliability are improved. And the technical effects of high efficiency and reliability in the assembly process are ensured.

Description

Technical Field

[0001] The invention relates to the technical field of low-voltage electrical appliances, and in particular to an assembly method and a platform for a pole-mounted circuit breaker. Background Art

[0002] Pole-mounted circuit breakers are one of the important devices in the power system, mainly used for short circuit, overload and other fault protection in power lines to ensure the safety of power equipment and lines. Pole-mounted circuit breakers are composed of multiple parts. The assembly sequence and mutual relationship of these parts are complex. Traditional methods are difficult to effectively deal with the relationship between parts. In addition, the monitoring of the assembly quality of parts during the assembly process is often limited to the detection of a single part. There is a lack of global and temporal analysis of the relationship between parts, and it is impossible to make real-time corrections and optimizations during the assembly process. Summary of the invention

[0003] The present application provides an assembly method and platform for a pole-mounted circuit breaker, which is used to solve the technical problems in the prior art of lacking global and temporal analysis of the relationship between the components of the pole-mounted circuit breaker, being unable to correct deviations in real time during the assembly process, and affecting assembly efficiency and reliability.

[0004] In a first aspect of the present application, a method for assembling a pole-mounted circuit breaker is provided, the method comprising: obtaining a plurality of components of a target pole-mounted circuit breaker, identifying assembly neighborhoods of the plurality of components according to connection relationships between the components in a preset assembly scheme, and obtaining a plurality of component assembly neighborhoods; assembling the plurality of components in sequence according to the preset assembly scheme, and synchronously monitoring the plurality of components and the plurality of component assembly neighborhoods during the assembly process, and obtaining a plurality of component assembly monitoring data sets and a plurality of neighborhood synchronous monitoring data sets; taking the components as indexes, extracting monitoring data from the plurality of neighborhood synchronous monitoring data sets according to the sequence of monitoring time windows, and obtaining a plurality of component neighborhood synchronous monitoring data sets. A monitoring data set sequence, wherein each component neighborhood synchronous monitoring data set sequence corresponds to a component, and each component neighborhood synchronous monitoring data corresponds to a monitoring time window; performing data iterative correlation analysis on the multiple component assembly monitoring data sets and the multiple component neighborhood synchronous monitoring data set sequences to determine multiple component long-term correlation monitoring data sets; traversing the multiple component long-term correlation monitoring data sets to identify assembly reliability factors to obtain multiple component assembly reliability factors; judging whether the multiple component assembly reliability factors are greater than or equal to a preset reliability factor threshold, if not, adding the corresponding components into an abnormal component set, and generating an assembly warning instruction based on the abnormal component set.

[0005] In a second aspect of the present application, an assembly platform for a pole-mounted circuit breaker is provided, the platform comprising: an assembly neighborhood identification module, the assembly neighborhood identification module is used to obtain multiple components of a target pole-mounted circuit breaker, and perform assembly neighborhood identification on the multiple components according to the connection relationship between the components in a preset assembly scheme to obtain multiple component assembly neighborhoods; a synchronous monitoring module, the synchronous monitoring module is used to assemble the multiple components in sequence according to the preset assembly scheme, and synchronously monitor the multiple components and the multiple component assembly neighborhoods during the assembly process to obtain multiple component assembly monitoring data sets and multiple neighborhood synchronous monitoring data sets; a monitoring data extraction module, the monitoring data extraction module is used to extract monitoring data from the multiple neighborhood synchronous monitoring data sets based on the order of monitoring time windows with the components as indexes, and obtain multiple component neighborhood synchronous monitoring data sets. A data set sequence, wherein each component neighborhood synchronous monitoring data set sequence corresponds to a component, and each component neighborhood synchronous monitoring data corresponds to a monitoring time window; an iterative association analysis module, wherein the iterative association analysis module is used to perform data iterative association analysis on the multiple component assembly monitoring data sets and the multiple component neighborhood synchronous monitoring data set sequences to determine multiple component long-term correlation monitoring data sets; an assembly reliability factor identification module, wherein the assembly reliability factor identification module is used to traverse the multiple component long-term correlation monitoring data sets to perform assembly reliability factor identification to obtain multiple component assembly reliability factors; an assembly warning module, wherein the assembly warning module is used to determine whether the multiple component assembly reliability factors are greater than or equal to a preset reliability factor threshold, and if not, add the corresponding component into an abnormal component set, and generate an assembly warning instruction based on the abnormal component set.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The present application provides an assembly method and platform for a pole-mounted circuit breaker, which relates to the field of low-voltage electrical appliance technology. The method and platform identify the connection relationship between components and synchronously monitor the components and neighborhood data during the assembly process, perform data extraction and iterative analysis, identify assembly reliability factors, and if the reliability factor of a component is lower than a threshold, mark it as abnormal, and generate an assembly warning instruction to achieve real-time monitoring and fault warning. The method solves the technical problems in the prior art of lacking global and temporal analysis of the association relationship between the components of the pole-mounted circuit breaker, being unable to make real-time corrections during the assembly process, and affecting the assembly efficiency and reliability. The method realizes real-time identification of potential assembly problems during the assembly process, and timely corrections, thereby ensuring the technical effect of high efficiency and reliability of the assembly process. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1 A schematic flow chart of an assembly method for a pole-mounted circuit breaker provided in an embodiment of the present application.

[0009] Figure 2 A schematic diagram of the structure of an assembly platform for a pole-mounted circuit breaker provided in an embodiment of the present application.

[0010] Explanation of the reference numerals: assembly neighborhood identification module 11 , synchronous monitoring module 12 , monitoring data extraction module 13 , iterative correlation analysis module 14 , assembly reliability factor identification module 15 , assembly early warning module 16 . DETAILED DESCRIPTION

[0011] The present application provides an assembly method and platform for a pole-mounted circuit breaker, which is used to solve the technical problems in the prior art of lacking global and temporal analysis of the relationship between the components of the pole-mounted circuit breaker, being unable to correct deviations in real time during the assembly process, and affecting assembly efficiency and reliability.

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0014] Embodiment 1, as Figure 1 As shown, the present application provides a method for assembling a pole-mounted circuit breaker, the method comprising: P10: Acquire multiple components of the target pole-mounted circuit breaker, identify assembly neighborhoods of the multiple components according to the connection relationship between the components in a preset assembly scheme, and obtain multiple component assembly neighborhoods.

[0015] Specifically, we first need to obtain multiple parts of the target pole-mounted circuit breaker. Pole-mounted circuit breakers are usually composed of multiple complex mechanical and electrical components, which need to be accurately assembled in a specific order and connection relationship during the assembly process. In order to ensure the efficiency and accuracy of the assembly process, the assembly neighborhood of each component must be identified according to the preset assembly plan.

[0016] The assembly neighborhood refers to the associated area between a component and other surrounding components during the assembly process. The neighborhood of each component is not only the physical contact area, but also includes its interaction relationship with other components. In order to accurately identify the assembly neighborhood of components, it is first necessary to clarify the connection relationship between the components through a preset assembly scheme. In this scheme, the function, position, connection method and other information of each component have been pre-defined, and this information includes the assembly order, connection method and adaptation requirements of the components.

[0017] In order to achieve neighborhood recognition, each component and its connection relationship can be represented in a digital environment through modeling methods, such as using CAD (computer-aided design) data or 3D modeling technology. This modeling process can be further optimized through computer vision and image processing technology to identify the position, orientation and connection between components during the assembly process. In this way, the assembly neighborhood of multiple components is obtained, and the assembly requirements and surrounding environment of each component are further evaluated.

[0018] In the assembly process, neighborhood identification is not just a static step, but also needs to be dynamically adjusted in real time according to the assembly progress. For example, when a component has been assembled and connected to other components, the system needs to update the neighborhood information of the component to ensure that the subsequent assembly components can be correctly connected to it. Therefore, the identification of assembly neighborhoods not only depends on preset solutions, but also needs to be optimized and adjusted in combination with real-time data monitoring and assembly progress.

[0019] P20: Assemble the multiple parts in sequence according to the preset assembly plan, and synchronously monitor the multiple parts and the multiple parts assembly neighborhoods during the assembly process to obtain multiple parts assembly monitoring data sets and multiple neighborhood synchronous monitoring data sets.

[0020] Furthermore, step P20 of the embodiment of the present application further includes: P21: According to the multiple assembly sequences corresponding to the multiple parts in the preset assembly scheme, the multiple parts are serialized to obtain a part sequence; P22: Based on the part sequence and the multiple part assembly neighborhoods, and the multiple assembly positions of the multiple parts in the preset assembly scheme, a monitoring sensor array analysis is performed to obtain a preset monitoring sensor array sequence; P23: Based on the part sequence and the multiple assembly positions, parts are assembled in sequence, and sensors are arranged according to the preset monitoring sensor array sequence during assembly, and synchronous monitoring is performed using the arranged preset monitoring sensor array sequence to obtain the multiple parts assembly monitoring data sets and multiple neighborhood synchronous monitoring data sets.

[0021] It should be understood that the core of the assembly process is to assemble multiple parts in sequence according to the preset assembly plan, and to synchronously monitor each part and its assembly neighborhood during this process, so as to collect assembly monitoring data in real time. The realization of this process not only depends on the precise control of the assembly sequence, but also requires the layout of the sensor array and data collection to ensure the efficiency and quality of the assembly.

[0022] First, multiple parts are serialized according to the assembly sequence in the preset assembly scheme. The goal of this step is to arrange the parts in the assembly in the correct assembly order and generate a part sequence. This serialization process is based on the assembly requirements of each part and their connection relationship to ensure that each part can be assembled in the appropriate order. Usually, this process is completed through a dedicated assembly management system, which automatically adjusts the assembly sequence based on assembly drawings, technical specifications and real-time feedback.

[0023] Next, based on the generated sequence of parts, the monitoring sensor array is analyzed in combination with each part’s assembly neighborhood (i.e., the area of ​​mutual relationship between the part and other parts) and assembly position. The assembly position refers to the precise spatial position of each part during the assembly process, which can be used to determine what kind of sensors are needed to monitor each part. The analysis of the monitoring sensor array means to configure the sensor array reasonably according to the assembly sequence and spatial arrangement of the parts, so as to monitor the status of each part and its neighborhood synchronously during the assembly process.

[0024] The sensor array includes temperature, pressure, vibration and other sensors, which are placed at key locations where stress, temperature changes or other physical phenomena may affect the assembly process. Through array analysis, it is possible to determine which areas need more monitoring data support and accurately place sensors at these locations.

[0025] After completing the analysis of the monitoring sensor array, the specific operation stage of component assembly begins. According to the assembly order and assembly position, the components will be assembled in sequence, and in each assembly step, the sensors will be arranged according to the preset sensor array sequence. These sensors will work synchronously with the components and continuously collect data during the assembly process. Synchronous monitoring means that all sensors can work in parallel in real time to continuously track the status of the components and their neighborhoods. During the assembly process, the data obtained by the sensors is transmitted to the monitoring system in real time to generate assembly monitoring data sets for multiple components and synchronous monitoring data sets for multiple component neighborhoods. These data sets contain the assembly status information of each component, such as the accuracy, displacement, stress, temperature and other parameters of the components during the assembly process, as well as environmental change information of the surrounding neighborhood, such as temperature fluctuations, pressure changes, etc.

[0026] During the assembly process, the sensor array can be connected to the central control system via wireless network or wired communication to transmit real-time data to the assembly monitoring system. These real-time data can not only help judge the assembly quality, but also provide data support for subsequent assembly process optimization and problem correction, thus improving the efficiency and accuracy of the entire assembly process.

[0027] P30: Taking the parts as indexes, the monitoring data of the multiple neighborhood synchronous monitoring data sets are extracted according to the order of the monitoring time windows to obtain multiple parts neighborhood synchronous monitoring data set sequences, wherein each part neighborhood synchronous monitoring data set sequence corresponds to a part, and each part neighborhood synchronous monitoring data corresponds to a monitoring time window.

[0028] Optionally, monitoring data is extracted through time series analysis to extract valuable information related to component assembly quality from real-time assembly monitoring data.

[0029] First, using the parts as indexes, multiple neighborhood synchronous monitoring data sets obtained in the previous step are extracted according to the order of the monitoring time windows. The purpose of this process is to organize the dynamic data of each part during the assembly process according to the time sequence to form an orderly sequence of part neighborhood synchronous monitoring data sets.

[0030] The component neighborhood synchronous monitoring data set contains real-time monitoring data of each component and its neighborhood. These data usually include real-time changes in physical parameters such as temperature, vibration, pressure, stress, and possible abnormal signals. In order to ensure that each component in the assembly process can be effectively monitored at the right time, it is necessary to organize these monitoring data according to the time sequence of the assembly process to form a complete monitoring data sequence.

[0031] Each component neighborhood synchronous monitoring dataset sequence corresponds to all monitoring data of a specific component during the assembly process. This dataset sequence reflects the long-term state and dynamic changes of the component during the assembly process through the arrangement and sorting of time windows. Each data record in each dataset sequence is obtained within a specific monitoring time window, and the data in each time window corresponds to the physical state of the component and its neighborhood at that moment.

[0032] In this way, a complete monitoring time series is established for each component, providing detailed, time-based assembly process data support for subsequent analysis. The establishment of the data series enables all key links in the assembly process to be systematically tracked and recorded, and provides complete historical data for subsequent analysis.

[0033] P40: Perform data iterative correlation analysis on the multiple component assembly monitoring data sets and the multiple component neighborhood synchronous monitoring data set sequences to determine multiple component long-term correlation monitoring data sets.

[0034] Furthermore, step P40 of the embodiment of the present application also includes: P41: extracting a plurality of first component neighborhood synchronous monitoring data sets from the plurality of component neighborhood synchronous monitoring data set sequences, performing data iterative association analysis with the corresponding plurality of component assembly monitoring data sets, and obtaining a plurality of first associated monitoring data sets; P42: extracting a plurality of second component neighborhood synchronous monitoring data sets from the plurality of component neighborhood synchronous monitoring data set sequences again, performing data iterative association analysis with the corresponding plurality of second associated monitoring data sets, and obtaining a plurality of second associated monitoring data sets; P43: performing multiple data iterative association analyses until reaching the end of the plurality of component neighborhood synchronous monitoring data set sequences, and obtaining the plurality of component long-term associated monitoring data sets.

[0035] It should be understood that through data iterative association analysis, the monitoring data of each component and its neighborhood during the assembly process is deeply analyzed, and finally a long-term correlation monitoring data set of components is obtained. The core purpose of data iterative association analysis is to identify the long-term association rules between components and their neighborhoods during the assembly process through repeated analysis of multiple data sets. This analysis helps to reveal the mutual influence of various components in the assembly process and their long-term impact on assembly quality, thereby providing a strong basis for refined management, assembly optimization, and early warning of potential problems.

[0036] First, multiple first component neighborhood synchronous monitoring data sets are extracted from multiple component neighborhood synchronous monitoring data set sequences. These data sets contain monitoring information of components and their neighborhoods within a specific time window, such as temperature, pressure, vibration, etc. At this time, the data in the first component neighborhood synchronous monitoring data set is usually in units of time windows, and the interactive changes between the components and their surroundings are gradually obtained in a time series manner.

[0037] Next, these extracted neighborhood monitoring data sets will be subjected to data iterative correlation analysis with the corresponding parts assembly monitoring data sets. Through this analysis, the parts assembly status is combined with the environmental changes of the surrounding neighborhood, and the dynamic relationship between the parts assembly process and the neighborhood changes is identified. This process uses a variety of data analysis methods, such as time series data analysis and data mining techniques, to find the correlation and influence pattern between parts assembly and neighborhood data.

[0038] After completing the first data iteration association analysis, the second round of analysis begins. At this point, multiple second component neighborhood synchronous monitoring data sets are extracted from multiple component neighborhood synchronous monitoring data set sequences again, and data iteration association analysis is performed with the multiple first associated monitoring data sets obtained previously. In this process, the goal of the second iteration analysis is to further deepen the understanding of the assembly process, especially to identify potential trends and anomalies in long-term assembly through multiple iteration analyses. For example, through data analysis of continuous time windows, it is possible to discover slight deviations in the long-term assembly of certain components, which are difficult to detect in the short term, but will gradually emerge with the iteration and association of data.

[0039] After multiple data iterations and correlation analysis, the analysis gradually deepens until the end of the entire component neighborhood synchronous monitoring data set sequence is completed. In this process, each iteration will further refine the relationship between the data and identify the long-term correlation pattern between each component and its neighborhood. This information is eventually aggregated into multiple component long-term correlation monitoring data sets, which contain deep correlation information between components and their neighborhoods during the assembly process, and can provide an accurate data basis for subsequent assembly quality assessment, problem warning, and decision support.

[0040] Furthermore, step P41 of the embodiment of the present application further includes: P41-1: Perform inner product mapping identification on the multiple first component neighborhood synchronous monitoring data sets and the corresponding multiple component assembly monitoring data sets respectively to obtain multiple first data similarity sets; P41-2: Traverse the multiple first data similarity sets to perform similarity normalization processing, and embed the processing results into an initially empty matrix to obtain multiple first iterative association matrices; P41-3: Convolve the multiple first iterative association matrices with the multiple first component neighborhood synchronous monitoring data sets to obtain multiple first associated monitoring data sets.

[0041] In a possible embodiment of the present application, the data analysis process is further refined, especially the association analysis between the first component neighborhood synchronous monitoring data set and the component assembly monitoring data set.

[0042] First, inner product mapping is performed on multiple first component neighborhood synchronous monitoring data sets and corresponding multiple component assembly monitoring data sets respectively. The core task of this step is to measure the similarity between component assembly data and its neighborhood monitoring data through inner product mapping. Inner product mapping is a mathematical method used to measure the similarity between two vectors in a high-dimensional space. In this step, the component assembly monitoring data and the neighborhood synchronous monitoring data will be converted into vector form, and each data point contains multidimensional features in the time series, such as temperature, pressure, vibration and other monitoring dimensions. By calculating the inner product of these data sets, the similarity value between the two sets of data can be obtained, that is, multiple first data similarity sets are obtained. These similarity values ​​reflect the dynamic relationship between the component assembly and its neighborhood environment.

[0043] Next, similarity normalization is performed on the obtained multiple first data similarity sets. Normalization is a common method in data preprocessing, which aims to adjust the data to a standard range to avoid deviations in the analysis of data with different dimensions or numerical ranges. Here, the normalization operation ensures that the data of different parts or monitoring time windows are consistent when calculating similarity. After the normalization process is completed, these normalized results are embedded into an initially empty matrix to form multiple first iterative association matrices. These matrices will be used for further analysis and provide a data basis for convolution calculations.

[0044] Finally, multiple first-iteration association matrices are convolved with multiple first component neighborhood synchronous monitoring data sets. Convolution is a mathematical operation commonly used in signal processing and image processing. It can perform weighted summation of data through a sliding window to extract key features from the data. In this step, the purpose of the convolution calculation is to combine the association matrix with the component neighborhood data in order to more accurately capture the dynamic changes and related patterns in the assembly process. Through the convolution operation, multiple first-association monitoring data sets can be obtained. These data sets reflect the long-term correlation characteristics with the neighborhood data during the component assembly process, which can provide a scientific basis for subsequent assembly quality assessment, early warning and optimization.

[0045] Furthermore, step P41-2 of the embodiment of the present application also includes: P41-21: Pre-construct a similarity normalization function, wherein the similarity normalization function is: ;in, is the normalized value corresponding to the i-th first data similarity in the first data similarity set, e is the base of the natural logarithm, m is the total number of first data similarities in the first data similarity set, is the first data similarity between the i-th first component neighborhood synchronous monitoring data in the first component neighborhood synchronous monitoring data set and the i-th component assembly monitoring data in the component assembly monitoring data set, is the i-th first component neighborhood synchronous monitoring data in the first component neighborhood synchronous monitoring data set, The i-th component assembly monitoring data in the component assembly monitoring data set; P41-22: using the similarity normalization function to perform similarity normalization processing on the multiple first data similarity sets, and embedding the processing results into an initially empty matrix to obtain multiple first iterative association matrices.

[0046] Specifically, the similarity normalization process is further refined, and a similarity normalization function is used to normalize multiple first data similarity sets. The purpose of this process is to convert the similarity values ​​of different component assembly data and neighborhood synchronous monitoring data into a standardized range through an accurate mathematical formula, thereby ensuring the consistency between different data and providing more reliable input for subsequent association analysis and convolution calculation.

[0047] First, a similarity normalization function is pre-constructed to normalize the similarity values ​​in the first data similarity set. The function is in the form of: ;in, is the normalized value corresponding to the i-th first data similarity in the first data similarity set, e is the base of the natural logarithm, m is the total number of first data similarities in the first data similarity set, is the first data similarity between the i-th first component neighborhood synchronous monitoring data in the first component neighborhood synchronous monitoring data set and the i-th component assembly monitoring data in the component assembly monitoring data set, is the i-th first component neighborhood synchronous monitoring data in the first component neighborhood synchronous monitoring data set, It is the i-th component assembly monitoring data in the component assembly monitoring data set.

[0048] The similarity normalization function mentioned above is used to perform similarity normalization processing on multiple first data similarity sets. Each similarity value is mapped to a new range to ensure the consistency and comparability of the data. The processed similarity data will be embedded in an initially empty matrix, which is called the first iterative association matrix. These iterative association matrices are the basic data structures in subsequent analysis, which are used to describe the relationship between different component assembly data and neighborhood data. Each matrix element represents the degree of association between a component and its corresponding neighborhood monitoring data. Each row of the matrix represents the assembly process of a component, and the column represents the corresponding neighborhood data. These matrices provide the necessary foundation for subsequent convolution calculations and long time series data analysis, helping to accurately capture potential anomalies and assembly quality problems in complex assembly processes.

[0049] Furthermore, step P41-3 of the embodiment of the present application also includes: P41-31: Using graph neural network as the basic framework, obtain multiple sample iterative association matrices and multiple sample first component neighborhood synchronous monitoring data sets, and the corresponding multiple sample first associated monitoring data sets as training data for the basic framework; P41-32: Based on the training data, use the K-fold cross-validation method to perform supervised training on the basic framework until the training converges to obtain a trained associated convolutional network layer; P41-33: Use the associated convolutional network layer to perform convolution calculations on the multiple first iterative association matrices and the multiple first component neighborhood synchronous monitoring data sets to obtain multiple first associated monitoring data sets.

[0050] Optionally, deep learning processing of iterative association matrix and parts assembly data can be performed through the combination of graph neural network and convolutional neural network to further improve the intelligence and accuracy of the assembly process.

[0051] First, the graph structure of the assembly data is constructed through a graph neural network (GNN). A graph neural network is a deep learning model suitable for processing data with a graph structure, and is used here to represent the relationship between parts and their assembly neighborhoods. Specifically, the iterative association matrix of multiple samples (representing the association between time series data between different parts) and the first part neighborhood synchronous monitoring data set (recording the data of the assembly neighborhood around the parts) and the corresponding first association monitoring data set (representing the association relationship between the parts and their neighborhood data) are obtained as training data. Through these training data, the graph neural network can capture the complex relationship between parts and provide a basis for subsequent deep learning analysis.

[0052] Next, based on the collected training data, supervised training of the graph neural network is performed using the K-fold cross-validation method. K-fold cross-validation is a commonly used machine learning model validation method that divides the training data into K subsets, trains the model with K-1 subsets each time, and uses the remaining 1 subset for validation. Finally, the validation results of each subset are combined to ensure the generalization ability of the model. Here, the iterative association matrix, the synchronous monitoring data set of the component neighborhood, and the associated monitoring data set are used to guide the training of the model during the training process. The goal is to enable the network to learn the potential laws and characteristics of the component assembly process. Through K-fold cross-validation, the model can reduce the risk of overfitting and improve the robustness of the training process.

[0053] After training, the resulting association convolutional network layer will be used to process new monitoring data. In this stage, the trained convolutional network layer is used to perform convolution calculations on multiple first-iteration association matrices and first-component neighborhood synchronous monitoring data sets to extract high-level features from the data. Convolution calculations can help the model identify and extract spatial and temporal correlations in the assembly process, enabling it to extract key association information from complex temporal data.

[0054] Finally, after convolution calculation, multiple first-association monitoring data sets are generated, which contain the key features of each component and its neighborhood assembly process, and are used to further analyze and evaluate the assembly quality. This intelligent analysis method based on deep learning can more accurately identify and correct assembly problems, and improve the reliability and intelligence level of the assembly process.

[0055] P50: Traverse the plurality of component long-term correlation monitoring data sets to identify assembly reliability factors, and obtain a plurality of component assembly reliability factors.

[0056] Furthermore, step P50 of the embodiment of the present application also includes: P51: Calculate the maximum data difference of the long-term correlation monitoring data sets of the multiple components respectively to obtain the maximum difference of the long-term correlation monitoring data of the multiple components; P52: Traverse the multiple long-term correlation monitoring data sets of the multiple components to perform variance calculation to obtain multiple fluctuation variances; P53: Perform weighted calculation on the maximum difference of the long-term correlation monitoring data of the multiple components and the multiple fluctuation variances to obtain the assembly reliability factors of the multiple components.

[0057] It should be understood that the long time series correlation monitoring data set of multiple parts is traversed, and the assembly reliability factor of each part is calculated one by one to quantify and evaluate the reliability of the assembly process. This process mainly analyzes the time series information in the monitoring data, evaluates the stability and reliability of each part in the assembly process, and identifies the key factors that may affect the assembly quality.

[0058] First, the maximum data difference is calculated from the long time series correlation monitoring data set of multiple components. The maximum data difference reflects the maximum fluctuation of each component during the assembly process. Specifically, it is the maximum change between consecutive monitoring data points of each component during the monitoring period. A large data difference usually means the instability of the component during the assembly process or possible assembly problems, so this indicator is an important basis for evaluating assembly reliability.

[0059] Next, the variance of the long-term correlation monitoring data set of multiple parts is calculated to obtain the fluctuation variance of each part. Variance is a common statistical indicator to measure the amplitude of data fluctuation, which indicates the degree of dispersion between the monitoring data points and the data average. A higher fluctuation variance usually indicates that there is a greater uncertainty in the assembly process of the part, which may affect the final assembly quality. Therefore, variance calculation helps to identify potential quality fluctuation risks in the assembly process.

[0060] Finally, the maximum data difference and fluctuation variance are weighted to obtain the assembly reliability factor of each component. The purpose of weighted calculation is to comprehensively consider the influence of the two and obtain a comprehensive reliability evaluation value. Specifically, different weights can be assigned to the maximum difference and variance to reflect their different effects on reliability during the assembly process. Through this weighted calculation, the assembly reliability of parts can be quantified more accurately, providing a basis for subsequent assembly warning and quality optimization. This method based on data analysis can identify potential assembly risks in advance, ensure high precision and high reliability of the assembly process, and help optimize the assembly process and improve overall product quality.

[0061] P60: Determine whether the assembly reliability factors of the multiple parts are greater than or equal to a preset reliability factor threshold. If not, add the corresponding parts to the abnormal parts set, and generate an assembly warning instruction based on the abnormal parts set.

[0062] Optionally, the assembly reliability factors of multiple parts are judged one by one, specifically judging whether the assembly reliability factor of each part meets a preset reliability factor threshold. The threshold is usually set based on factors such as assembly process requirements, part performance standards, or historical assembly data, and is used to distinguish which parts meet assembly requirements and which parts may have assembly risks.

[0063] Specifically, for each component, first obtain its assembly reliability factor, and then compare it with the preset reliability factor threshold. If the assembly reliability factor of the component is greater than or equal to the threshold, it is considered that the reliability of the component in the assembly process meets the requirements and the assembly process can continue. If the assembly reliability factor is lower than the threshold, it indicates that there may be problems in the assembly process of the component, the reliability is insufficient, and special attention is required.

[0064] Components with assembly reliability factors below the threshold are marked as abnormal components and added to the abnormal component set. Components in this set usually require further inspection and processing to determine the specific cause of their assembly problems and take necessary corrective measures. For example, the assembly process may need to be re-evaluated, or these components may need to be replaced or repaired.

[0065] Next, an assembly warning instruction is generated based on the abnormal parts set, which is intended to notify relevant personnel (such as assembly workers, quality inspectors or equipment operators) about the existence of abnormal parts so that timely measures can be taken. The warning instruction may include a specific description of the abnormal parts, expected assembly problems and possible subsequent processing steps, thereby helping staff to quickly identify problems and take corrective measures to ensure assembly quality.

[0066] In summary, the embodiments of the present application have at least the following technical effects: This application identifies the assembly neighborhood by acquiring parts and their connection relationships, and assembles parts in sequence according to a preset assembly plan. During the assembly process, the parts and their neighborhoods are synchronously monitored to generate an assembly monitoring data set. The assembly reliability factor is identified by extracting the monitoring data and performing iterative correlation analysis. If the reliability factor of a part is lower than a threshold, it is added to the abnormal part set, and an assembly warning instruction is generated to achieve real-time monitoring and fault warning.

[0067] The technical effect of real-time identification of potential assembly problems during the assembly process and timely correction is achieved to ensure the efficiency and reliability of the assembly process.

[0068] Embodiment 2 is based on the same inventive concept as the assembly method of a pole-mounted circuit breaker in the above embodiment. Figure 2 As shown, the present application provides an assembly platform for a pole-mounted circuit breaker, and the platform in the embodiment of the present application and the method embodiment are based on the same inventive concept. The platform includes: The assembly neighborhood identification module 11 is used to obtain multiple components of the target pole-mounted circuit breaker, and perform assembly neighborhood identification on the multiple components according to the connection relationship between the components in the preset assembly scheme to obtain multiple component assembly neighborhoods.

[0069] The synchronous monitoring module 12 is used to assemble the multiple parts in sequence according to the preset assembly plan, and synchronously monitor the multiple parts and the multiple parts assembly neighborhoods during the assembly process to obtain multiple parts assembly monitoring data sets and multiple neighborhood synchronous monitoring data sets.

[0070] The monitoring data extraction module 13 is used to extract monitoring data from the multiple neighborhood synchronous monitoring data sets based on the order of monitoring time windows with the parts as indexes, and obtain multiple parts neighborhood synchronous monitoring data set sequences, wherein each part neighborhood synchronous monitoring data set sequence corresponds to a part, and each part neighborhood synchronous monitoring data corresponds to a monitoring time window.

[0071] The iterative association analysis module 14 is used to perform data iterative association analysis on the multiple component assembly monitoring data sets and multiple component neighborhood synchronous monitoring data set sequences to determine multiple component long-term association monitoring data sets.

[0072] The assembly reliability factor identification module 15 is used to traverse the plurality of component long-term correlation monitoring data sets to perform assembly reliability factor identification and obtain a plurality of component assembly reliability factors.

[0073] The assembly warning module 16 is used to determine whether the assembly reliability factors of the multiple parts are greater than or equal to a preset reliability factor threshold. If not, the corresponding parts are added to the abnormal parts set, and an assembly warning instruction is generated according to the abnormal parts set.

[0074] Furthermore, the synchronization monitoring module 12 is also used to perform the following steps: According to the multiple assembly sequences corresponding to the multiple parts in the preset assembly scheme, the multiple parts are serialized to obtain a part sequence; based on the part sequence and the multiple part assembly neighborhoods, and the multiple assembly positions of the multiple parts in the preset assembly scheme, a monitoring sensor array analysis is performed to obtain a preset monitoring sensor array sequence; based on the part sequence and the multiple assembly positions, parts are assembled in sequence, and sensors are arranged according to the preset monitoring sensor array sequence during assembly, and synchronous monitoring is performed using the arranged preset monitoring sensor array sequence to obtain the multiple parts assembly monitoring data sets and multiple neighborhood synchronous monitoring data sets.

[0075] Furthermore, the iterative association analysis module 14 is further configured to perform the following steps: Extract multiple first component neighborhood synchronous monitoring data sets from the multiple component neighborhood synchronous monitoring data set sequences, perform data iterative association analysis with the corresponding multiple component assembly monitoring data sets, and obtain multiple first associated monitoring data sets; extract multiple second component neighborhood synchronous monitoring data sets from the multiple component neighborhood synchronous monitoring data set sequences again, perform data iterative association analysis with the corresponding multiple second associated monitoring data sets, and obtain multiple second associated monitoring data sets; perform multiple data iterative association analyses until reaching the end of the multiple component neighborhood synchronous monitoring data set sequences, and obtain the multiple component long-term associated monitoring data sets.

[0076] Furthermore, the iterative association analysis module 14 is further configured to perform the following steps: Inner product mapping is performed on the multiple first component neighborhood synchronous monitoring data sets and the corresponding multiple component assembly monitoring data sets respectively to obtain multiple first data similarity sets; the multiple first data similarity sets are traversed to perform similarity normalization processing, and the processing results are embedded in an initially empty matrix to obtain multiple first iterative association matrices; convolution calculation is performed on the multiple first iterative association matrices and the multiple first component neighborhood synchronous monitoring data sets to obtain multiple first associated monitoring data sets.

[0077] Furthermore, the iterative association analysis module 14 is further configured to perform the following steps: A similarity normalization function is pre-constructed, wherein the similarity normalization function is: ;in, is the normalized value corresponding to the i-th first data similarity in the first data similarity set, e is the base of the natural logarithm, m is the total number of first data similarities in the first data similarity set, is the first data similarity between the i-th first component neighborhood synchronous monitoring data in the first component neighborhood synchronous monitoring data set and the i-th component assembly monitoring data in the component assembly monitoring data set, is the i-th first component neighborhood synchronous monitoring data in the first component neighborhood synchronous monitoring data set, The i-th component assembly monitoring data in the component assembly monitoring data set is used; the similarity normalization function is used to perform similarity normalization processing on the multiple first data similarity sets, and the processing results are embedded in an initially empty matrix to obtain multiple first iterative association matrices.

[0078] Furthermore, the iterative association analysis module 14 is further configured to perform the following steps: Taking the graph neural network as the basic framework, multiple sample iterative association matrices and multiple sample first component neighborhood synchronous monitoring data sets, as well as corresponding multiple sample first associated monitoring data sets are obtained as training data for the basic framework; based on the training data, the basic framework is supervised trained using the K-fold cross-validation method until the training converges to obtain a trained associated convolutional network layer; the associated convolutional network layer is used to perform convolution calculations on the multiple first iterative association matrices and the multiple first component neighborhood synchronous monitoring data sets to obtain multiple first associated monitoring data sets.

[0079] Furthermore, the assembly reliability factor identification module 15 is also used to perform the following steps: The maximum data difference of the long-term correlation monitoring data sets of the multiple components is calculated respectively to obtain the maximum difference of the long-term correlation monitoring data of the multiple components; the variance calculation is performed on the multiple long-term correlation monitoring data sets of the multiple components to obtain multiple fluctuation variances; and the maximum difference of the long-term correlation monitoring data of the multiple components and the multiple fluctuation variances are weightedly calculated to obtain the assembly reliability factors of the multiple components.

[0080] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0081] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0082] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for assembling a pole mounted circuit breaker, characterized in that: The method comprises: Acquire multiple components of the target pole-mounted circuit breaker, identify assembly neighborhoods of the multiple components according to connection relationships between the components in a preset assembly scheme, and obtain multiple component assembly neighborhoods; Assembling the multiple parts in sequence according to the preset assembly scheme, and synchronously monitoring the multiple parts and the multiple parts assembly neighborhoods during the assembly process to obtain multiple parts assembly monitoring data sets and multiple neighborhood synchronous monitoring data sets; Taking the parts as indexes, extracting monitoring data from the multiple neighborhood synchronous monitoring data sets according to the order of monitoring time windows, and obtaining multiple parts neighborhood synchronous monitoring data set sequences, wherein each part neighborhood synchronous monitoring data set sequence corresponds to a part, and each part neighborhood synchronous monitoring data corresponds to a monitoring time window; Performing data iteration association analysis on the multiple component assembly monitoring data sets and multiple component neighborhood synchronous monitoring data set sequences to determine multiple component long-term association monitoring data sets; Traversing the plurality of component long-term correlation monitoring data sets to identify assembly reliability factors, and obtaining a plurality of component assembly reliability factors; It is determined whether the assembly reliability factors of the plurality of parts are greater than or equal to a preset reliability factor threshold value. If not, the corresponding parts are added into an abnormal parts set, and an assembly warning instruction is generated according to the abnormal parts set.

2. The method for assembling a pole mounted circuit breaker according to claim 1, characterized in that: Performing data iteration association analysis on the multiple component assembly monitoring data sets and multiple component neighborhood synchronous monitoring data set sequences to determine multiple component long-term association monitoring data sets, including: Extracting a plurality of first component neighborhood synchronous monitoring data sets from the plurality of component neighborhood synchronous monitoring data set sequences, and performing data iterative correlation analysis with the corresponding plurality of component assembly monitoring data sets to obtain a plurality of first correlated monitoring data sets; Extracting a plurality of second component neighborhood synchronous monitoring data sets from the plurality of component neighborhood synchronous monitoring data set sequences again, and performing data iterative association analysis with the corresponding plurality of second associated monitoring data sets to obtain a plurality of second associated monitoring data sets; After multiple data iteration association analyses, until the end of the plurality of component neighborhood synchronous monitoring data set sequences is reached, the plurality of component long-term association monitoring data sets are obtained.

3. The method for assembling a pole mounted circuit breaker according to claim 2, characterized in that: Extracting a plurality of first component neighborhood synchronous monitoring data sets from the plurality of component neighborhood synchronous monitoring data set sequences, and performing data iterative association analysis with the corresponding plurality of component assembly monitoring data sets to obtain a plurality of first associated monitoring data sets, including: Performing inner product mapping identification on the plurality of first component neighborhood synchronous monitoring data sets and the corresponding plurality of component assembly monitoring data sets respectively to obtain a plurality of first data similarity sets; Traversing the plurality of first data similarity sets to perform similarity normalization processing, and embedding the processing results into an initially empty matrix to obtain a plurality of first iterative association matrices; Convolution calculation is performed on the multiple first iterative association matrices and the multiple first component neighborhood synchronous monitoring data sets to obtain multiple first associated monitoring data sets.

4. The method for assembling a pole mounted circuit breaker according to claim 3, characterized in that: include: A similarity normalization function is pre-constructed, wherein the similarity normalization function is: ； in, is the normalized value corresponding to the i-th first data similarity in the first data similarity set, e is the base of the natural logarithm, m is the total number of first data similarities in the first data similarity set, is the first data similarity between the i-th first component neighborhood synchronous monitoring data in the first component neighborhood synchronous monitoring data set and the i-th component assembly monitoring data in the component assembly monitoring data set, is the i-th first component neighborhood synchronous monitoring data in the first component neighborhood synchronous monitoring data set, It is the i-th component assembly monitoring data in the component assembly monitoring data set; The similarity normalization function is used to perform similarity normalization processing on the multiple first data similarity sets, and the processing results are embedded in an initially empty matrix to obtain multiple first iterative correlation matrices.

5. The method for assembling a pole mounted circuit breaker according to claim 3, characterized in that: include: Taking the graph neural network as the basic framework, multiple sample iterative association matrices and multiple sample first component neighborhood synchronous monitoring data sets, as well as corresponding multiple sample first association monitoring data sets are obtained as training data for the basic framework; Based on the training data, the basic framework is supervised trained using a K-fold cross validation method until the training converges, thereby obtaining a trained associated convolutional network layer; The association convolutional network layer is used to perform convolution calculations on the multiple first iterative association matrices and the multiple first component neighborhood synchronous monitoring data sets to obtain multiple first association monitoring data sets.

6. The method for assembling a pole mounted circuit breaker according to claim 1, characterized in that: Traversing the plurality of component long-term correlation monitoring data sets to identify assembly reliability factors, and obtaining a plurality of component assembly reliability factors, including: Calculate the maximum data difference of the plurality of component long-series correlation monitoring data sets respectively to obtain the maximum data difference of the plurality of component long-series correlation monitoring data sets; Traversing the plurality of component long-term correlation monitoring data sets to perform variance calculation to obtain a plurality of fluctuation variances; The maximum difference of the long-term correlated monitoring data of the multiple components and the multiple fluctuation variances are weightedly calculated to obtain the assembly reliability factors of the multiple components.

7. The method for assembling a pole mounted circuit breaker according to claim 1, characterized in that: The plurality of parts are assembled sequentially according to the preset assembly scheme, and the plurality of parts and the plurality of parts assembly neighborhoods are synchronously monitored during the assembly process to obtain a plurality of parts assembly monitoring data sets and a plurality of neighborhood synchronous monitoring data sets, including: According to a plurality of assembly sequences corresponding to the plurality of components in the preset assembly scheme, serializing the plurality of components to obtain a component sequence; Performing a monitoring sensor array analysis based on the component sequence and the plurality of component assembly neighborhoods, and the plurality of assembly positions of the plurality of components in the preset assembly scheme to obtain a preset monitoring sensor array sequence; Based on the component sequence and multiple assembly positions, components are assembled in sequence, and during the assembly, sensors are arranged according to the preset monitoring sensor array sequence. Synchronous monitoring is performed using the arranged preset monitoring sensor array sequence to obtain the multiple component assembly monitoring data sets and multiple neighborhood synchronous monitoring data sets.

8. An assembly platform for a pole-mounted circuit breaker, characterized in that: The platform includes: An assembly neighborhood identification module, the assembly neighborhood identification module is used to obtain multiple components of the target pole-mounted circuit breaker, perform assembly neighborhood identification on the multiple components according to the connection relationship between the components in the preset assembly scheme, and obtain multiple component assembly neighborhoods; A synchronous monitoring module, wherein the synchronous monitoring module is used to assemble the plurality of parts in sequence according to the preset assembly scheme, and synchronously monitor the plurality of parts and the plurality of parts assembly neighborhoods during the assembly process to obtain a plurality of parts assembly monitoring data sets and a plurality of neighborhood synchronous monitoring data sets; A monitoring data extraction module, wherein the monitoring data extraction module is used to extract monitoring data from the multiple neighborhood synchronous monitoring data sets based on the order of monitoring time windows with the parts as indexes, and obtain multiple parts neighborhood synchronous monitoring data set sequences, wherein each part neighborhood synchronous monitoring data set sequence corresponds to a part, and each part neighborhood synchronous monitoring data corresponds to a monitoring time window; An iterative association analysis module, the iterative association analysis module is used to perform data iterative association analysis on the multiple component assembly monitoring data sets and the multiple component neighborhood synchronous monitoring data set sequences to determine multiple component long-term association monitoring data sets; An assembly reliability factor identification module, wherein the assembly reliability factor identification module is used to traverse the plurality of component long-term correlation monitoring data sets to perform assembly reliability factor identification and obtain a plurality of component assembly reliability factors; The assembly warning module is used to determine whether the assembly reliability factors of the multiple parts are greater than or equal to a preset reliability factor threshold. If not, the corresponding parts are added to the abnormal parts set, and an assembly warning instruction is generated according to the abnormal parts set.

Images