Intelligent optimization method and system for electric energy metering equipment production line

Abstract

The application provides an intelligent optimization method and system for an electric energy metering equipment production line, relates to the technical field of industrial automation and intelligent manufacturing, and comprises the following steps: obtaining production data of the production line in a current fault-free time period, including sensor data and product image data; performing smoothing processing on the sensor data and feature extraction on the product image data on the production data to obtain current time period production parameters; solving optimal production parameters with the minimum error in the production process as the target; dynamically adjusting the production process based on the optimal production parameters, monitoring and comparing the difference between the current time period production parameters and the optimal production parameters in real time, adopting a model predictive control algorithm for optimization to obtain control input of a next time period, which is used for adjusting operation parameters of production equipment in the production line; and the application optimizes the control input of the production line based on the sensor data and the product image data, improves the stability, efficiency, response timeliness and the like of the optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation and intelligent manufacturing, and particularly relates to an intelligent optimization method and system for an electric energy metering device production line. BACKGROUND

[0002] In modern industrial production, electric energy metering devices, as one of the key components, are widely used in energy management, power monitoring and smart grid fields. With the development of industrial automation and intelligence, higher requirements are put forward for the production efficiency, quality control and cost management of electric energy metering devices. Traditional electric energy metering device production lines mainly rely on manual monitoring and operation, and the production efficiency and product quality are difficult to meet the growing market demand. In order to cope with this challenge, it is urgent to introduce advanced intelligent technology to improve the automation level and intelligent degree of the production process.

[0003] With the rapid development of sensor technology, image processing technology and big data analysis technology, the data acquisition, processing and analysis capability in industrial production process has been significantly improved. The application of these technologies makes it possible to realize real-time monitoring, intelligent fault handling and production optimization in the production process. By deploying various types of sensors on the production line, multi-dimensional data such as temperature, humidity, pressure and product quality can be collected in real time, providing detailed production environment and product state information. At the same time, the camera system can capture the appearance and assembly of the product, and realize real-time detection of product quality through image processing technology.

[0004] The existing technology at least has the following technical problems:

[0005] Traditional production lines rely on manual monitoring and operation, which is low in efficiency and prone to errors, resulting in low production efficiency; due to the lack of real-time monitoring and intelligent fault handling mechanism, production delay and product quality problems often occur;

[0006] The existing technology lacks systematicness and scientificalness in optimizing the production process, and it is difficult to find the bottlenecks and optimization points in the production process through data analysis and optimization algorithm. The adjustment of production parameters mainly depends on experience and manual operation, and it is difficult to realize dynamic optimization, and the product quality is difficult to guarantee.

[0007] Therefore, the existing electric energy metering device production line optimization has deficiencies in stability, efficiency and response timeliness. SUMMARY

[0008] In order to solve the above problems, the present application provides an intelligent optimization method and system for an electric energy metering device production line, which optimizes the control input of the production line based on sensor data and product image data, improves the stability, efficiency and response timeliness of the optimization.

[0009] According to some embodiments, the present application adopts the technical solutions as follows:

[0010] An intelligent optimization method of an electric energy metering device production line, comprising:

[0011] Obtaining production data of the electric energy metering device production line in a current failure-free time period, including sensor data and product image data;

[0012] Performing smoothing processing on the sensor data and feature extraction on the product image data on the production data to obtain current time period production parameters;

[0013] Solving optimal production parameters with the objective of minimizing errors in the production process;

[0014] Based on the optimal production parameters, dynamically adjusting the production process, real-time monitoring and comparing the differences between the current time period production parameters and the optimal production parameters, and using a model predictive control algorithm for optimization to obtain control inputs of the next time period for adjusting the operating parameters of the production equipment in the production line.

[0015] According to some embodiments, the present application adopts the technical solutions as follows:

[0016] An intelligent optimization system of an electric energy metering device production line, comprising:

[0017] A production data acquisition module configured to obtain production data of the electric energy metering device production line in a current failure-free time period, including sensor data and product image data;

[0018] A production data processing module configured to perform smoothing processing on the sensor data and feature extraction on the product image data on the production data to obtain current time period production parameters;

[0019] A production parameter solving module configured to solve optimal production parameters with the objective of minimizing errors in the production process;

[0020] A control input optimization module configured to, based on the optimal production parameters, dynamically adjust the production process, real-time monitor and compare the differences between the current time period production parameters and the optimal production parameters, and use a model predictive control algorithm for optimization to obtain control inputs of the next time period for adjusting the operating parameters of the production equipment in the production line.

[0021] According to some embodiments, the present application adopts the technical solutions as follows:

[0022] A computer program product comprising a computer program, which, when executed by a processor, implements the intelligent optimization method of the electric energy metering device production line.

[0023] According to some embodiments, the present application adopts the technical solutions as follows:

[0024] A non-transitory computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the intelligent optimization method for an electric energy metering device production pipeline.

[0025] According to some embodiments, the present application adopts the technical solutions as follows:

[0026] An electronic device comprising a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory; when the electronic device is running, the processor executes the computer program stored in the memory to make the electronic device execute the intelligent optimization method for an electric energy metering device production pipeline.

[0027] Compared with the prior art, the present application has the beneficial effects that:

[0028] 1、The present application identifies bottlenecks and optimization points in the production process by analyzing collected data, constructs a target function to solve the optimal solution, and dynamically adjusts the production process based on the optimal production parameters to ensure the stability and efficiency of production; the application of model predictive control algorithm optimizes the production parameters to ensure that the production process tends to the optimal state.

[0029] 2、The present application monitors environmental parameters and overall product appearance in the production process in real time by deploying sensors and cameras on the production line, discovers and handles production abnormalities in a timely manner, reduces production delays, and improves production efficiency; the application of multi-dimensional adaptive smoothing algorithm and multi-scale adaptive feature extraction algorithm realizes efficient processing of sensor data and image data, improves the accuracy and efficiency of data processing, and ensures the accuracy and response speed of data processing.

[0030] 3、The present application continuously adjusts control input through real-time monitoring and feedback cycles to ensure the stability and efficiency of the production process; the establishment of fault detection and handling mechanism improves the robustness and reliability of the system, ensuring the continuity and stability of production.

[0031] 4、The present application improves the efficiency and accuracy of fault detection and handling by using a multi-layer Bayesian decision network to judge and handle faults, responds immediately when detecting production abnormalities, automatically suspends and handles problems, and reduces the need for manual intervention.

[0032] 5、The application accurately detects product surface defects through a multi-scale adaptive feature extraction algorithm, improves the accuracy of product quality detection, and ensures the consistency and reliability of product quality; through real-time monitoring and data analysis, production parameters are optimized, resource waste and production loss are reduced, and production cost is reduced; the application of hierarchical adaptive data compression algorithm improves the data transmission efficiency, reduces the data transmission delay and storage cost. BRIEF DESCRIPTION OF DRAWINGS

[0033] The accompanying drawings, which form a part of the present description, are used to provide further understanding of the application, and the illustrative embodiments of the application and their description serve to explain the application. The drawings do not constitute an improper limitation of the application.

[0034] Figure 1 The method flowchart of example 1. DETAILED DESCRIPTION

[0035] The application will be further described below in conjunction with the drawings and examples.

[0036] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the application. Unless otherwise indicated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the application belongs.

[0037] It should be noted that the terms used herein are only for the purpose of describing specific embodiments, and are not intended to limit the exemplary embodiments according to the application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and furthermore, it should be understood that when the terms "comprise" and / or "comprise" are used in the specification, they mean the presence of a feature, step, operation, device, component and / or combination thereof.

[0038] Example 1

[0039] An embodiment of the application provides an intelligent optimization method for an electric energy metering device production line, as shown in Figure 1 , comprising:

[0040] Step S1: Obtain the production data of the electric energy metering device production line in the current fault-free time period, including sensor data and product image data;

[0041] Step S2: Smooth the sensor data and extract the features of the product image data from the production data to obtain the production parameters in the current time period;

[0042] Step S3: Solve the optimal production parameters with the objective of minimizing the error in the production process;

[0043] Step S4: Based on the optimal production parameters, dynamically adjust the production process, monitor the difference between the current time period production parameters and the optimal production parameters in real time, use model predictive control algorithm for optimization, get the control input of the next time period, and adjust the operation parameters of the production equipment in the production line.

[0044] As an embodiment, an intelligent optimization method of an electric energy metering equipment production line of the application optimizes the control input of the production line based on sensor data and product image data, improves the stability, efficiency, and timeliness of the optimization, and the specific implementation process is as follows:

[0045] I. Obtain sensor data and product image data of the electric energy metering equipment, smooth the sensor data and extract image features through a multi-dimensional adaptive smoothing algorithm and a multi-scale adaptive feature extraction algorithm, and detect faults through a multi-layer Bayesian decision network.

[0046] 1. Sensor data and smoothing processing

[0047] The sensors are deployed at key points of the production line, which can reflect the state of the key process, including but not limited to temperature detection points, humidity detection points, pressure detection points, and product quality detection points. The sensors real-time monitor the temperature, humidity, pressure and other environmental parameters in the production process and the key quality indicators of the product, including temperature sensors, humidity sensors, pressure sensors, and quality sensors. Each sensor reflects the current production environment and product quality status through real-time data collection.

[0048] For example, temperature sensors are deployed in the cooling area after the shell of the electric energy metering equipment is formed, humidity sensors are deployed in the environmental monitoring points of the assembly process or coating curing process, pressure sensors are deployed in the pressure collection points of the press-fitting or testing station, and quality sensors are deployed in the final inspection station for detecting product size, weight, precision, etc.; The collected parameters are directly related to the production process of the electric energy metering equipment, including environmental parameters, reflecting the temperature and humidity of the air or tooling fixture around the production station, pressure (for controlling process stability); device / product parameters, reflecting the temperature of the processed electric energy metering equipment body or assembly (such as PCB welding temperature, shell surface temperature); the pressure applied during assembly; the weight / size of the finished product and other quality indicators.

[0049] The multi-dimensional adaptive smoothing algorithm processes the collected data by applying an adaptive smoothing strategy in a multi-dimensional space, eliminates data noise and maintains response speed through a multiple weighting and adjustment mechanism, as follows:

[0050] Each sensor collects data at different time points, including temperature data, humidity data, pressure data, and quality data, forming a multi-dimensional data set Set initial weights for each sensor data point. The initial weights reflect the impact of different sensor data on the overall smoothed data. At each time point t, the smoothed data is calculated using a weighted average. The formula for smoothing data is:

[0051]

[0052] in, This represents the smoothed data at time t. As the initial weights, For the first Time of the first Dimensional sensor data, It is the size of the time window. It refers to the number of dimensions in the sensor data.

[0053] The weight adjustment mechanism is as follows:

[0054]

[0055] in, For the first The standard deviation of dimensional sensor data To adjust the coefficients and control the sensitivity of the weights, real-time smoothing of the data is achieved, providing a reliable data foundation for subsequent monitoring and analysis modules.

[0056] 2. Product image data and feature extraction

[0057] Cameras are positioned at different locations on the production line to capture product images at each stage of production, showing the product's appearance and assembly status in real time, including the product's appearance, assembly process, and final product.

[0058] Specifically, cameras are deployed along the production line at key processes, including:

[0059] Incoming materials / component loading area: Inspect the appearance of components, batch markings, and solder joint integrity, etc.

[0060] PCB assembly and soldering area: Monitor solder joint quality, component position deviation, solder amount, etc.

[0061] Shell and structure assembly area: Inspect the shell for scratches, deformation, and assembly gaps;

[0062] Key component integration and calibration areas (such as the installation locations of sensing modules and metering modules), etc.

[0063] The production stage refers to a processing / assembly process segment divided according to the process sequence on the assembly line, which is set according to the actual situation. Each stage corresponds to an independent station or process unit of the assembly line in space, and constitutes a continuous segment of the production process in time.

[0064] Using a multi-scale adaptive feature extraction algorithm, image features are extracted by multi-scale image features and adaptive weight mechanisms to analyze product images to detect product surface defects. The extracted image features are expressed by the formula:

[0065]

[0066] wherein, is the feature value after convolution operation at the image coordinates , is the adaptive convolution kernel parameter, representing the parameter of the convolution kernel in the cth row and dth column, is the number of rows of the multi-scale convolution kernel, is the number of columns of the multi-scale convolution kernel, is the image pixel value, , is the multi-scale offset, is the bias term, and the offset adjustment mechanism is:

[0067]

[0068] wherein, and are the standard deviations of the pixel intensity of the convolution kernel in the cth row and dth column in the vertical direction and horizontal direction, respectively; and are the adjustment factors in the vertical direction and horizontal direction, respectively.

[0069] The adaptive weight mechanism here mainly reflects the dynamic adjustment mode of (convolution kernel parameter) and , (multi-scale offset) in the convolution feature extraction process. The adaptation is reflected in the dynamic adjustment of the weight and offset with the local features of the image, rather than a fixed convolution kernel.

[0070] Multi-scale feature extraction, positioning and identification of product surface defects improve the accuracy of product quality detection.

[0071] 3. Fault detection and fault handling

[0072] Based on the smoothed sensor data and the extracted image features, a multi-layer Bayesian decision network is used for real-time fault detection. When a fault is detected, production is automatically suspended, and after the fault is resolved, production is restarted and production data is collected, including two sub-steps: fault detection and fault handling.

[0073] Specifically, the multi-layer Bayesian decision network uses the extracted feature values and the smoothed sensor data to make fault judgments, and detects a fault to immediately issue an alarm and notify the controller of the production line to automatically suspend production. The multi-layer Bayesian decision network is used to calculate the probability of a fault, which is expressed by the formula:

[0074]

[0075] wherein, represents the probability of a fault occurring under the feature data and the smoothed sensor data represents the set of all observation data used, represents the probability of observing the feature and the smoothed sensor data when a fault occurs, is an adjustment factor, is a normalization constant, is a prior fault probability, is a prior non-fault probability.

[0076] After calculating , a fault is detected by a pre-set threshold value. If exceeds the pre-set threshold value, it is considered that there is a fault. Once a fault is detected, production is automatically suspended, and the fault is handled. If the fault cannot be automatically handled, an alarm is issued and an operator is notified to intervene. After the fault is handled, the production line is restarted.

[0077] II. After the fault is handled, production data is collected, data compression is performed by a hierarchical adaptive data compression algorithm, a first objective function is constructed to solve optimal production parameters, and target optimization control inputs are used to optimize the production process, thereby ensuring the stability and efficiency of production.

[0078] After the fault is handled and normal production is resumed, the smoothed data of the sensors and the image feature data in the production process are collected, and the collected data is compressed by a hierarchical adaptive data compression algorithm. The formula is:

[0079]

[0080] wherein, is a data compression rate, is the probability of occurrence of the th data, is an adjustment coefficient, is the comprehensive data of the smoothed data and the image feature data, ​The average of the comprehensive data. Through efficient data compression, reduce data transmission, improve data transmission efficiency.

[0081] The collected data is analyzed to identify bottlenecks and optimization points in the production process, and the optimal solution of the production parameters is solved by constructing a first objective function, the first objective function aims to minimize the error in the production process, the production parameters are variables that can be adjusted and controlled, and the first objective function is as follows:

[0082]

[0083] wherein, a set of production parameters to be optimized, such as equipment operating speed, temperature setting value, production line scheduling, etc. 、 is the weight in the optimization process; is the i-th sensor data under the production parameter , is the target value of the sensor data; is the image feature data under the production parameter , is the target value of the image feature data, and n and m are the number of sensor data and image feature data, respectively.

[0084] In order to minimize the first objective function , the optimization problem can be expressed as:

[0085]

[0086] To solve the optimal solution of the first objective function , an optimization algorithm such as gradient descent or genetic algorithm is needed, which is relatively mature in the prior art, and this embodiment will not be described in detail.

[0087] After obtaining the optimal production parameters , the production process is dynamically adjusted based on the optimal production parameters to ensure the stability and efficiency of the production, the optimal production parameters including sensor data and image feature data , the sensor data is used to represent the controllable production parameters (such as temperature, humidity, pressure and physical mass measurement values), and the image feature data is used to represent the overall appearance of the product.

[0088] By real-time monitoring, the difference between the current production parameters and the optimal production parameters is compared, and the deviation calculation formula is:

[0089]

[0090]

[0091] wherein, and is the deviation between the current production parameters and the optimal production parameters, and is the current sensor data and image feature data, and is the optimal production parameters.

[0092] By calculating the deviation, the difference between the current production state and the optimal state can be quantified, so as to determine whether adjustment is needed; the model predictive control algorithm is adopted to adjust the production parameters according to the deviation value, to ensure that the production process tends to be optimal, and the optimization objective of the model predictive control is to minimize the deviation and the change of control input in the future period of time, which is represented by the second objective function:

[0093]

[0094] wherein, , are the weights of the production parameter deviation, is the weight of the control input, is the change of the control input, is the next period of time to be predicted, by minimizing the deviation and the change of the control input, the production parameters are optimized, so as to realize the optimal production state.

[0095] By optimizing the second objective function , a new control input is obtained and applied to the production process, and the update formula of the control input is:

[0096]

[0097] The updated control input is used to adjust the operating parameters of the production equipment, to ensure that the production process is carried out according to the optimal parameters.

[0098] A specific example is provided:

[0099] Firstly, sensors and cameras are installed at key positions of the production line to ensure comprehensive monitoring of the entire production process, the sensors detect environmental parameters and product quality indicators, and the cameras capture product appearance images.

[0100] Then, the data of the sensors and cameras are monitored in real time, once an abnormality is found, an alarm is immediately issued, the production is automatically suspended, the fault cause is analyzed, and corresponding handling measures are executed.

[0101] Secondly, after the fault handling is completed and the normal production is resumed, the monitoring data is processed to obtain the optimal production parameters;

[0102] Finally, according to the optimal production parameters, the control input is dynamically adjusted to optimize the production process and ensure the stability and efficiency of production.

[0103] Embodiment 2

[0104] In an embodiment of the present application, an intelligent optimization system for an electric energy metering device production line is provided, comprising:

[0105] The production data acquisition module is configured to acquire production data of the electric energy metering device production line in a current fault-free time period, including sensor data and product image data;

[0106] The production data processing module is configured to perform sensor data smoothing processing and product image data feature extraction on the production data to obtain current time period production parameters;

[0107] The production parameter solving module is configured to solve the optimal production parameters by minimizing the error in the production process;

[0108] The control input optimization module is configured to dynamically adjust the production process based on the optimal production parameters, monitor and compare the difference between the current time period production parameters and the optimal production parameters in real time, use a model predictive control algorithm for optimization to obtain control input for the next time period, and use the control input to adjust the operating parameters of the production equipment in the production line.

[0109] Embodiment 3

[0110] In an embodiment of the present application, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the intelligent optimization method for an electric energy metering device production line.

[0111] Embodiment 4

[0112] In an embodiment of the present application, a non-transitory computer readable storage medium is provided for storing computer instructions, which, when executed by a processor, implement the intelligent optimization method for an electric energy metering device production line.

[0113] Embodiment 5

[0114] In an embodiment of the present application, an electronic device is provided, comprising a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory; when the electronic device is running, the processor executes the computer program stored in the memory to make the electronic device execute the intelligent optimization method for an electric energy metering device production line.

[0115] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps in one or more flow or blocks.

[0116] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 steps for functionally implementing the steps in one or more flow or blocks.

[0117] Although the present application has been described in connection with the preferred embodiments thereof with reference to the accompanying drawings, it is to be noted that various changes and modifications can be made thereto without departing from the scope of the present application as set forth in the appended claims.

Claims

1. A method for intelligent optimization of an electricity metering equipment production line, characterized in that, include: The production data of the power metering equipment production line during the current fault-free period is obtained, including sensor data and product image data. The production data during the current fault-free period is based on smoothed sensor data and extracted image features. A multi-layer Bayesian decision network is used for real-time fault detection. When a fault is detected, production is automatically suspended. Production is restarted and production data is collected after the fault is resolved. The production data is smoothed using sensor data and features are extracted from product image data to obtain production parameters for the current time period. The sensor data is smoothed using a multi-dimensional adaptive smoothing algorithm, and the product image data is extracted using a multi-scale adaptive feature extraction algorithm. The goal is to minimize errors in the production process, and then solve for the optimal production parameters. The objective of minimizing errors in the production process is expressed by the first objective function as follows: Where s represents a set of production parameters to be optimized; , These are the weights in the optimization process; s is the data from the i-th sensor under the production parameters. It is the target value of the sensor data; It is image feature data under production parameters. is the target value of the image feature data, and n and m are the number of sensor data and image feature data, respectively. Based on the optimal production parameters, the production process is dynamically adjusted, and the difference between the current time period production parameters and the optimal production parameters is monitored and compared in real time. The model predictive control algorithm is used for optimization to obtain the control input for the next time period, which is used to adjust the operating parameters of the production equipment in the production line. The real-time monitoring compares the differences between the current production parameters and the optimal production parameters, and optimizes them using a model predictive control algorithm, expressed by the second objective function: in, , All are weights of production parameter deviations. It controls the weights of the inputs. and For the bias of sensor data and the bias of image feature data, The variable is the change in the control input, and T is the next time period to be predicted. The control input is optimized by minimizing the deviation and the change in the control input.

2. The intelligent optimization method for a production line of electricity metering equipment as described in claim 1, characterized in that, The sensor data includes temperature data, humidity data, pressure data, and mass data, which are collected by sensors installed on the production line. The product image data is captured by cameras arranged along the production line at preset processes, showing the product's appearance and assembly status.

3. An intelligent optimization system for an electricity metering equipment production line, characterized in that, include: The production data acquisition module is configured to: acquire production data of the power metering equipment production line during the current fault-free period, including sensor data and product image data; the production data of the current fault-free period is based on smoothed sensor data and extracted image features, and uses a multi-layer Bayesian decision network for real-time fault detection. When a fault is detected, production is automatically suspended, and production is restarted and production data is collected after the fault is resolved. The production data processing module is configured to: perform smoothing processing on sensor data and feature extraction on product image data to obtain production parameters for the current time period; the sensor data is smoothed using a multi-dimensional adaptive smoothing algorithm; and the product image data is extracted using a multi-scale adaptive feature extraction algorithm. The production parameter solution module is configured to solve for the optimal production parameters with the goal of minimizing errors in the production process. The objective of minimizing errors in the production process is expressed by the first objective function as follows: Where s represents a set of production parameters to be optimized; , These are the weights in the optimization process; s is the data from the i-th sensor under the production parameters. It is the target value of the sensor data; It is image feature data under production parameters. is the target value of the image feature data, and n and m are the number of sensor data and image feature data, respectively. The control input optimization module is configured to: dynamically adjust the production process based on the optimal production parameters, monitor and compare the difference between the current time period production parameters and the optimal production parameters in real time, optimize using a model predictive control algorithm, and obtain the control input for the next time period to adjust the operating parameters of the production equipment in the production line. The real-time monitoring compares the differences between the current production parameters and the optimal production parameters, and optimizes them using a model predictive control algorithm, expressed by the second objective function: in, , All are weights of production parameter deviations. It controls the weights of the inputs. and For the bias of sensor data and the bias of image feature data, The variable is the change in the control input, and T is the next time period to be predicted. The control input is optimized by minimizing the deviation and the change in the control input.

4. The intelligent optimization system for an electricity metering equipment production line as described in claim 3, characterized in that, The sensor data, including temperature, humidity, pressure, and mass data, is collected by sensors installed on the production line; the product image data is captured by cameras arranged along the production line at predetermined stages, showing the product's appearance and assembly status.

5. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the intelligent optimization method for a production line of electricity metering equipment as described in any one of claims 1-2, including: The production data of the power metering equipment production line during the current fault-free period is obtained, including sensor data and product image data. The production data during the current fault-free period is based on smoothed sensor data and extracted image features. A multi-layer Bayesian decision network is used for real-time fault detection. When a fault is detected, production is automatically suspended. Production is restarted and production data is collected after the fault is resolved. The production data is smoothed using sensor data and features are extracted from product image data to obtain production parameters for the current time period. The sensor data is smoothed using a multi-dimensional adaptive smoothing algorithm, and the product image data is extracted using a multi-scale adaptive feature extraction algorithm. The goal is to minimize errors in the production process, and then solve for the optimal production parameters. The objective of minimizing errors in the production process is expressed by the first objective function as follows: Where s represents a set of production parameters to be optimized; , These are the weights in the optimization process; s is the data from the i-th sensor under the production parameters. It is the target value of the sensor data; It is image feature data under production parameters. is the target value of the image feature data, and n and m are the number of sensor data and image feature data, respectively. Based on the optimal production parameters, the production process is dynamically adjusted, and the difference between the current time period production parameters and the optimal production parameters is monitored and compared in real time. The model predictive control algorithm is used for optimization to obtain the control input for the next time period, which is used to adjust the operating parameters of the production equipment in the production line. The real-time monitoring compares the differences between the current production parameters and the optimal production parameters, and optimizes them using a model predictive control algorithm, expressed by the second objective function: in, , All are weights of production parameter deviations. It controls the weights of the inputs. and For the bias of sensor data and the bias of image feature data, The variable is the change in the control input, and T is the next time period to be predicted. The control input is optimized by minimizing the deviation and the change in the control input.

6. A computer program product as described in claim 5, comprising a computer program, characterized in that, The sensor data, including temperature, humidity, pressure, and mass data, is collected by sensors installed on the production line; the product image data is captured by cameras arranged along the production line at predetermined stages, showing the product's appearance and assembly status.

7. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions, which, when executed by a processor, implement the intelligent optimization method for an electricity metering equipment production line as described in any one of claims 1-2.

8. An electronic device, characterized in that, include: The device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to perform an intelligent optimization method for an electricity metering equipment production line as described in any one of claims 1-2.

Images