Intelligent control method, system and device for anchorage device machining equipment

Abstract

The invention provides an intelligent control method, system and device for anchorage device machining equipment. The method comprises the steps that a sensor monitors key index data of the anchorage device machining equipment in real time; key index data collected by the sensor is sent to the intelligent control model for analysis through the remote monitoring platform; based on the historical key index data, an intelligent control model of the machining process is established; the intelligent control model carries out real-time processing process analysis and optimization on the key index data; the technological parameters are imported into anchorage device machining equipment, the anchorage device machining equipment executes automatic clamping and automatic machining according to the technological parameters, and automatic clamping and automatic machining are achieved; and meanwhile, by combining a sensor and image recognition, automatic detection and real-time feedback of the product quality in the machining process are achieved, the actual machining result is compared, and deviation or defects in the machining process are recognized and corrected. The system comprises a data acquisition module, a data processing module and a parameter adjustment module. Accurate control and real-time monitoring of the machining process are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent equipment control, and in particular to an intelligent control method, system and device for anchor processing equipment. Background Art

[0002] A steel anchor is a heavy metal device used to fix a ship or other floating structure. It is usually made of steel, has a high density and strength, and can firmly fix a ship or buoy underwater with its heavy weight. The main uses of steel anchors include: ship mooring and anchoring, used to fix ships so that they can be moored or anchored on the water surface to prevent drifting or being affected by wind and waves; marine engineering and construction, in marine engineering, such as the construction of offshore platforms, submarine pipelines, etc., steel anchors are needed to fix and support structures; underwater equipment fixation, used to fix underwater equipment, submarine detectors and buoys to ensure that they are stable and reliable when working underwater; diving operation support, in diving operations, steel anchors can be used as safe anchoring points for divers to provide work stability and safety. In order to ensure that it has sufficient weight and structural strength, steel anchors need to be processed so that they can be effectively used in various water conditions. In short, the processing and use of steel anchors are intended to provide stability and safety for ships and marine engineering, ensuring that they can operate and use effectively in waters. However, the existing steel anchor processing equipment has a low degree of automation, low production efficiency, and cannot achieve autonomous adjustment of processing parameters, resulting in poor accuracy of the processed steel anchor.

[0003] Prior art 1, application number: CN202111572446.8 discloses a multi-hole anchor intelligent processing and sorting equipment and positioning method based on machine vision, including a grasping device, a processing device, a conveying device, a sorting device and an electrical control system. Under the control of the electrical control system, the workpiece to be processed is grasped from the silo and placed at a specified position on the conveyor belt. The telescopic cylinder pushes the workpiece to be processed to the processing disk, the processing disk rotates 900, the photoelectric sensor detects the workpiece to be processed, the clamping cylinder clamps the workpiece to be processed, the visual camera accurately locates the position of the hole to be processed, the drill is processed, the processing disk rotates 900 again, the transfer device places the processed workpiece in the sorting area, and during the movement of the conveyor belt, the qualified product sorting ultrasonic sensor detects whether the workpiece is qualified, the telescopic cylinder pushes the qualified workpiece to the qualified product sorting area, and if it is unqualified, the unqualified product sorting telescopic cylinder pushes it to the unqualified product sorting area. Although it has the effects of easy operation, accurate positioning and rapid sorting. However, its intelligent control structure is relatively small, and the automation control level of the equipment cannot be effectively improved, which affects the processing efficiency to a certain extent.

[0004] Prior art 2, application number: CN 202110613673.4 Replica prestressed porous anchor processing equipment and process, including a frame, on which an upper die and a lower die are arranged in relative positions, and it is characterized in that when the limit plate is transmitted to the lower die, the two corresponding limit plates are clamped with the base arranged under the conveyor belt, so that the two corresponding limit plates form a circular ring, and the control module controls the conveyor belt to stop moving, so that the conveyor belt is wrapped in the cavity formed by the lower die, the base and the two corresponding limit plates, and a barrel for placing the blank is formed between the top of the lower die and the two corresponding limit plates, so that the roundness punch on the lower die forms a roundness hole, and then the blank is extruded by the upper die, and the tapered punch contacts and extrude the end face corresponding to the roundness punch to form a tapered hole that passes through the roundness hole, thereby obtaining the anchor plate, and finally, the conveyor belt is started, so that the limit plate is separated from the base, and the two corresponding limit plates clamp the anchor plate to separate from the lower die, and the anchor plate falls into the conveyor belt to enter the next process. Although the prestressed anchor can be obtained by extrusion, and the prestressed anchor blank can undergo significant dynamic recrystallization, greatly improving the grain refinement effect of the blank, the prestressed porous anchor processing equipment and process, however, the structure is mainly mechanical, requiring manual operation, which reduces production efficiency and cannot achieve autonomous adjustment of processing parameters.

[0005] Prior art three, application number: CN 202211443351.0 discloses an anchor clip sawing device, including: a mounting frame; a conveying device, a blocking device, a platform device, a collecting box, a cutting device and a controller are fixedly installed on the mounting frame; the conveying device is used to convey the anchor clip to the platform device, the blocking device is used to intercept the anchor clip on the conveying device, so that the anchor clip on the conveying device enters the platform device one by one, the cutting device is used to cut the anchor clip on the platform device, and the collecting box is used to collect the anchor clip after cutting; the conveying device, the blocking device, the platform device and the cutting device are all electrically connected to the controller. Although the conveying device and the platform device do not need to manually place the anchor clip on the lower end of the band saw machine, which ensures the safety of the operator, the anchor clip only needs to be placed on the conveying device; compared with manually placing the anchor clip on the lower end of the band saw machine, it not only increases safety, but also improves efficiency. However, its function is relatively single, and only involves partial improvements in anchor processing, and it is difficult to improve the production efficiency of the overall equipment.

[0006] At present, the existing technologies 1, 2 and 3 have the problem that the steel anchor processing equipment has a low degree of automation and cannot realize autonomous adjustment of processing parameters, which reduces the production and processing efficiency and makes the accuracy of the processed steel anchor poor. Therefore, the present invention provides an intelligent control method, system and device for anchor processing equipment to realize accurate control and real-time monitoring of the processing process. Through the intelligent control system, the equipment can realize full-process automation operations such as automatic clamping, automatic processing and automatic detection; reduce manual intervention, improve production efficiency, and can automatically adjust processing parameters according to real-time feedback to ensure stability and accuracy during the processing. Summary of the invention

[0007] In order to solve the above technical problems, the present invention provides an intelligent control method for anchor processing equipment, comprising the following steps:

[0008] Install sensors on the anchor processing equipment to monitor the processing status, process parameters and key indicator data of product quality of the anchor processing equipment in real time; send the key indicator data collected by the sensors to the intelligent control model for analysis through the remote monitoring platform;

[0009] Based on historical key indicator data, an intelligent control model of the processing process is established; the intelligent control model is used to perform real-time processing analysis and optimization of key indicator data, and to control and optimize process parameters;

[0010] The process parameters are imported into the anchor processing equipment, and the anchor processing equipment performs automatic clamping and automatic processing according to the process parameters to realize automatic clamping and automatic processing; at the same time, combined with sensors and image recognition, the product quality in the processing process is automatically detected and real-time feedback is provided, and the actual processing results are compared to identify and correct deviations or defects in the processing.

[0011] Optionally, the key indicator data collection process includes the following steps:

[0012] Obtain the content of the key indicator data, determine the target position of the key indicator data according to the content, mark the target position according to the preset order of the key indicator data, and establish a corresponding relationship list between the target position and the content of the key indicator data; the target position is a processing area or a heating area of ​​the equipment;

[0013] According to the correspondence list, determine the location of establishing the IoT communication node, where the IoT communication node is the location where the sensor is placed;

[0014] Connect each IoT communication node and aggregate it to the IoT bus, then transmit key indicator data to the remote monitoring platform, which is connected to the input end of the intelligent control model through the Internet.

[0015] Optionally, the process of controlling and optimizing the process parameters comprises the following steps:

[0016] Obtain the key indicator data including processing status, process parameters and product quality collected by historical sensors, and perform cleaning and normalization preprocessing operations on the key indicator data; extract time series features from the preprocessed keyword indicator data, and select target features that reach the impact threshold on the predicted target;

[0017] The intelligent control model is trained using target features, and the parameters of the intelligent control model are adjusted using an adaptive learning rate method;

[0018] During the processing, real-time data is continuously acquired from sensors and updated to the constructed intelligent control model. Using the real-time updated key indicator data, the intelligent control model performs real-time analysis and prediction to evaluate the status and trend of the current processing process. Based on the analysis results of the intelligent control model, the process parameters are adjusted in real time to optimize the efficiency and quality of the processing process.

[0019] Optionally, the process of selecting target features that reach a threshold of influence on the prediction target includes the following steps:

[0020] Arrange the preprocessed key indicator data in chronological order, divide the time series data into multiple sliding windows, each window contains a preset number of continuous data points; extract features of the key indicator data in each sliding window according to the number and time interval of data points in each sliding window;

[0021] Use the Pearson correlation coefficient to calculate the correlation coefficient between each feature in the preprocessed key indicator data and the predicted target; according to the calculated correlation coefficient value, sort the features from high to low according to their correlation with the target;

[0022] A correlation coefficient threshold is selected in the sorted feature list as the influence threshold. Based on the correlation coefficient threshold, features with correlation coefficients exceeding the threshold are selected as features that influence the target. The extracted features are combined into feature vectors, and each sliding window corresponds to a feature vector.

[0023] Optionally, the construction process of the intelligent control model includes the following steps:

[0024] Select the neural network model as the intelligent control model, divide the preprocessed key indicator data set into a training set and a validation set in a ratio of 7:3; normalize the features in the training set and the validation set;

[0025] Use the key indicator data and corresponding target features in the training set to train the neural network model using the back propagation algorithm; use the adaptive learning rate method to adjust the parameters of the intelligent control model, and automatically adjust the learning rate according to the performance and convergence of the intelligent control model;

[0026] Use the key indicator data in the validation set and the trained intelligent control model to evaluate and verify the model. The evaluation indicators include prediction error, accuracy and recall rate.

[0027] The optional, automatic learning rate adjustment process consists of the following steps:

[0028] Randomly initialize the weights and biases of the neural network model; input the key indicator data in the training set into the neural network model, and obtain the output of the neural network model through calculation;

[0029] Compare the output of the neural network model with the target features in the training set, calculate the value of the loss function, and measure the prediction error of the neural network model; according to the value of the loss function, use the chain rule to calculate the gradient of the neural network model parameters to the loss function, back-propagate the gradient from the output layer to the input layer, and update the parameters of the neural network model; according to the calculated gradient, update the weights and biases of the neural network model until the preset number of training rounds is reached;

[0030] The learning rate is adaptively adjusted according to the historical gradient of the key indicator data, and the attenuation factor is introduced to perform exponential weighted averaging on the historical gradient of the key indicator data to achieve automatic adjustment of the learning rate changes.

[0031] Optionally, the process of updating the weights and biases of the neural network model includes the following steps:

[0032] For each training sample, input data into the neural network, perform forward propagation calculations, and obtain the output of the neural network model; calculate the gradient of the loss function between the output of the neural network model and the actual label. According to the chain rule, calculate the gradient of each layer from the output layer to the input layer in turn;

[0033] Use the calculated gradients to update the model’s weights and biases for each sample in the training dataset.

[0034] Continue iterating training until the preset number of training rounds or convergence condition is reached.

[0035] Optionally, the specific update rule for updating the weights and biases of the model is: new weight = original weight - learning rate * gradient.

[0036] The present invention provides an intelligent control system for anchor processing equipment, comprising:

[0037] The data acquisition module is responsible for installing sensors on the anchor processing equipment. The sensors monitor the processing status, process parameters and key indicator data of product quality of the anchor processing equipment in real time; the key indicator data collected by the sensors are sent to the intelligent control model for analysis through the remote monitoring platform;

[0038] The data processing module is responsible for establishing an intelligent control model for the processing process based on historical key indicator data; using the intelligent control model to perform real-time processing analysis and optimization of key indicator data, and to control and optimize process parameters;

[0039] The parameter adjustment module is responsible for importing the process parameters into the anchor processing equipment. The anchor processing equipment performs automatic clamping and automatic processing according to the process parameters to realize automatic clamping and automatic processing. At the same time, combined with sensors and image recognition, it automatically detects and provides real-time feedback on product quality during the processing, compares the actual processing results, and identifies and corrects deviations or defects in the processing.

[0040] The present invention provides an intelligent control device for anchor processing equipment, comprising:

[0041] Anchor processing equipment is responsible for laying out sensors, performing automatic clamping and automatic processing according to process parameters, and realizing automatic clamping and automatic processing;

[0042] Sensors are responsible for real-time monitoring of the processing status, process parameters and key indicator data of product quality of anchor processing equipment; and the key indicator data collected by the sensors are sent to the intelligent control model for analysis through the remote monitoring platform;

[0043] The remote monitoring platform is responsible for realizing remote control and monitoring of anchor processing equipment, sensors and intelligent control models; combined with sensors and image recognition, it can automatically detect and provide real-time feedback on product quality during the processing, compare the actual processing results, and identify and correct deviations or defects in the processing;

[0044] The intelligent control model is responsible for real-time processing analysis and optimization of key indicator data, and for controlling and optimizing process parameters.

[0045] The present invention installs sensors and monitors key indicator data in real time, realizes real-time monitoring of key indicators such as the processing status, process parameters and product quality of the anchor processing equipment, obtains real-time processing data, and provides a data basis for subsequent intelligent control and optimization. An intelligent control model of the processing process is established, and by analyzing historical key indicator data, the influence of different process parameters on product quality and equipment status during the processing process is understood; an accurate model is established to provide a basis for subsequent intelligent control and optimization. By importing process parameters, automatic clamping and automatic processing of equipment are realized. At the same time, combined with sensors and image recognition technology, automatic detection and real-time feedback of product quality during the processing process are realized; by comparing the actual processing results, deviations or defects in the processing can be identified and corrected; the automation and intelligence of the processing process are realized, and production efficiency, product quality stability and consistency are improved.

[0046] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0047] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0049] Figure 1 This is a flow chart of the intelligent control method of the anchor processing equipment in Example 1 of the present invention;

[0050] Figure 2 This is a diagram of the collection process of key indicator data in Example 2 of the present invention;

[0051] Figure 3 This is a process diagram for controlling and optimizing process parameters in Example 3 of the present invention;

[0052] Figure 4 A process diagram of selecting a target feature that reaches an impact threshold on a predicted target in Embodiment 4 of the present invention;

[0053] Figure 5 This is a diagram of the construction process of the intelligent control model in Example 5 of the present invention;

[0054] Figure 6 A process diagram of automatically adjusting the learning rate in Embodiment 6 of the present invention;

[0055] Figure 7A process diagram for updating the weights and biases of a neural network model in Embodiment 7 of the present invention;

[0056] Figure 8 A process diagram for realizing automatic adjustment of learning rate changes in Embodiment 8 of the present invention;

[0057] Fig. 9 A process diagram for identifying and correcting deviations or defects in processing in Example 9 of the present invention;

[0058] Fig.10 This is a block diagram of an intelligent control system for an anchor processing device in Embodiment 10 of the present invention;

[0059] Fig.11 This is a block diagram of the intelligent control device of the anchor processing equipment in Example 11 of the present invention. DETAILED DESCRIPTION

[0060] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0061] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the embodiments of the present application. The singular forms of "a", "said" and "the" used in the embodiments of the present application are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0062] When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as described in detail. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and do not have to be used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.

[0063] Example 1: Figure 1 As shown, an embodiment of the present invention provides an intelligent control method for anchor processing equipment, comprising the following steps:

[0064] S100: Install sensors on the anchor processing equipment to monitor key indicator data such as the processing status, process parameters and product quality of the anchor processing equipment in real time; send the key indicator data collected by the sensors to the intelligent control model for analysis through the remote monitoring platform;

[0065] Among them, sensors include but are not limited to temperature sensors, humidity sensors, force sensors, image recognition sensors, etc. These sensors monitor the key parameters of the anchor processing equipment in real time, such as temperature, humidity, force, image, etc.; Processing status: refers to the status of the anchor processing equipment itself, including the equipment's on / off status, operating status, fault status, etc. The status information can be obtained through sensors monitoring the equipment's current, voltage, speed, vibration and other parameters; Process parameters: refers to the parameters that need to be set and adjusted during the anchor processing process to control the processing effect and quality; specific process parameters may include but are not limited to processing speed, cutting tool feed speed, cutting depth, coolant flow and temperature, etc.; Product quality: refers to the quality characteristics of the final product obtained during the processing process. These characteristics may include dimensional accuracy, surface finish, surface roughness, material hardness, etc. Product quality can be detected and evaluated in real time through sensors or image recognition technology;

[0066] S200: Based on historical key indicator data, an intelligent control model of the processing process is established; the intelligent control model is used to perform real-time processing analysis and optimization of key indicator data, and to control and optimize process parameters;

[0067] S300: The process parameters are imported into the anchor processing equipment, and the anchor processing equipment performs automatic clamping and automatic processing according to the process parameters to realize automatic clamping and automatic processing; at the same time, combined with sensors and image recognition, the product quality in the processing process is automatically detected and real-time feedback is provided, and the actual processing results are compared to identify and correct deviations or defects in the processing.

[0068] The working principle and beneficial effects of the above technical scheme are as follows: In this embodiment, sensors are first installed on the anchor processing equipment, and the sensors monitor the key indicator data such as the processing status, process parameters and product quality of the anchor processing equipment in real time; the key indicator data collected by the sensors are sent to the intelligent control model for analysis through the remote monitoring platform; secondly, based on the historical key indicator data, an intelligent control model of the processing process is established; the intelligent control model is used to perform real-time processing process analysis and optimization on the key indicator data, and to control and optimize the process parameters; finally, the process parameters are imported into the anchor processing equipment (the process parameters are input into the controller of the anchor processing equipment, and the controller adjusts the process parameters through the connection relationship with the corresponding structure of the anchor processing equipment), and the anchor processing equipment performs automatic clamping and automatic processing according to the process parameters to realize automatic clamping and automatic processing; at the same time, combined with sensors and image recognition, the product quality in the processing process is automatically detected and real-time feedback is provided, and the actual processing results are compared to identify and correct deviations or defects in the processing. Step S100 of the above scheme installs sensors and monitors key indicator data in real time, realizes real-time monitoring of key indicators such as the processing status, process parameters and product quality of the anchor processing equipment, obtains real-time processing data, and provides a data basis for subsequent intelligent control and optimization. Step S200 establishes an intelligent control model for the processing process, and understands the impact of different process parameters on product quality and equipment status during the processing by analyzing historical key indicator data; establishes an accurate model to provide a basis for subsequent intelligent control and optimization. Step S300 realizes automatic clamping and automatic processing of equipment by importing process parameters. At the same time, combined with sensors and image recognition technology, automatic detection and real-time feedback of product quality during the processing process can be realized; by comparing the actual processing results, deviations or defects in the processing can be identified and corrected; the automation and intelligence of the processing process can be realized, and production efficiency, product quality stability and consistency can be improved.

[0069] To sum up, this embodiment realizes precise control and real-time monitoring of the processing process through real-time monitoring, intelligent analysis and automatic control; through intelligent control models and optimization algorithms, process parameters can be adjusted in real time to improve the stability and precision of the processing process; automatic clamping and automatic processing can reduce human intervention and improve production efficiency; automatic detection and real-time feedback can ensure the stability of product quality; improve production efficiency, product quality and the intelligent level of production management, reduce production costs, and enhance the competitiveness of enterprises; realize precise control and real-time monitoring of the processing process. Through intelligent control, the equipment can realize full-process automation operations such as automatic clamping, automatic processing and automatic detection, reduce human intervention, improve production efficiency, and can automatically adjust processing parameters according to real-time feedback to ensure stability and precision during the processing.

[0070] Example 2: Figure 2As shown, based on Example 1, the key indicator data collection process provided by the embodiment of the present invention includes the following steps:

[0071] S101: Acquire the content of the key indicator data, determine the target position of the key indicator data according to the content, mark the target position according to the preset order of the key indicator data, and establish a corresponding relationship list between the target position and the content of the key indicator data; the target position is a processing area or a heating area of ​​the equipment;

[0072] S102: Determine the location of establishing an IoT communication node according to the correspondence list, where the IoT communication node is the location where the sensor is placed;

[0073] S103: Connect various IoT communication nodes, aggregate them to the IoT bus, and then transmit key indicator data to the remote monitoring platform. The remote monitoring platform is connected to the input end of the intelligent control model through the Internet.

[0074] The working principle and beneficial effects of the above technical solution are as follows: the present embodiment first obtains the content of the key indicator data, determines the target position of the key indicator data according to the content, marks the target position according to the preset sorting of the key indicator data, and establishes a corresponding relationship list between the target position and the content of the key indicator data; the target position is the processing area or heating area of ​​the equipment; secondly, according to the corresponding relationship list, determines the location of establishing the Internet of Things communication node, and the Internet of Things communication node is the location where the sensor is placed; finally, each Internet of Things communication node is connected and aggregated to the Internet of Things bus, and then the key indicator data is transmitted to the remote monitoring platform, and the remote monitoring platform is connected to the input end of the intelligent control model through the Internet. Step S101 of the above scheme establishes a corresponding relationship list between the target position and the content of the key indicator data by obtaining the content of the key indicator data and determining the target position. The list can help clearly understand the key indicator data corresponding to each target position, and provides a basis for subsequent data collection and analysis. Step S102 determines the location where the Internet of Things communication node needs to be established according to the corresponding relationship list; the Internet of Things communication node is the location where the sensor is placed. By placing the sensor in a suitable position, accurate collection of key indicator data can be achieved; ensure accurate key indicator data is obtained, and provide an accurate data basis for subsequent data transmission and analysis. Step S103 connects each IoT communication node and transmits key indicator data to the remote monitoring platform through the IoT bus; the remote monitoring platform is connected to the input end of the intelligent control model through the Internet to realize remote monitoring and data analysis; and realizes real-time monitoring and analysis of key indicator data to provide data support for subsequent intelligent control and optimization.

[0075] To sum up, the key indicator data collection process of this embodiment determines the target position of data collection and the position of sensor placement by establishing a list of corresponding relationships between the target position and the key indicator data content; the collection and transmission of key indicator data are realized through the establishment of Internet of Things communication nodes and Internet of Things buses; real-time monitoring and analysis of key indicator data are realized through the connection and data analysis of the remote monitoring platform; accurate key indicator data are provided to provide a data basis for subsequent intelligent control and optimization, thereby improving production efficiency and the stability of product quality.

[0076] Example 3: Figure 3 As shown, based on Example 1, the process of controlling and optimizing process parameters provided in the embodiment of the present invention includes the following steps:

[0077] S201: Acquire key indicator data including processing status, process parameters and product quality collected by historical sensors, and perform preprocessing operations such as cleaning and normalization on the key indicator data; extract time series features from the preprocessed keyword indicator data, and select target features that reach the impact threshold for the prediction target;

[0078] S202: Using the target features to train the intelligent control model, and adjusting the parameters of the intelligent control model by an adaptive learning rate method;

[0079] S203: During the processing, real-time data is continuously acquired from sensors and updated to the constructed intelligent control model; using the real-time updated key indicator data, the intelligent control model performs real-time analysis and prediction to evaluate the status and trend of the current processing process; according to the analysis results of the intelligent control model, the process parameters are adjusted in real time to optimize the efficiency and quality of the processing process.

[0080] The working principle and beneficial effects of the above technical solution are as follows: the present embodiment first obtains the key indicator data including processing status, process parameters and product quality collected by the historical sensor, and performs pre-processing operations such as cleaning and normalization on the key indicator data; extracts time series features from the pre-processed keyword indicator data, and selects the target features that reach the impact threshold on the prediction target; secondly, the target features are used to train the intelligent control model, and the parameters of the intelligent control model are adjusted by the adaptive learning rate method; finally, in the processing process, real-time data is continuously obtained from the sensor and updated to the constructed intelligent control model; using the real-time updated key indicator data, the intelligent control model performs real-time analysis and prediction, and evaluates the status and trend of the current processing process; according to the analysis results of the intelligent control model, the process parameters are adjusted in real time to optimize the efficiency and quality of the processing process. Step S201 of the above scheme obtains the key indicator data collected by the historical sensor, and performs pre-processing operations such as cleaning and normalization on it, extracts time series features from it, and selects the target features that have an impact on the prediction target; extracts meaningful features from the original key indicator data, and provides an accurate data basis for subsequent model training and prediction. Step S202 uses the target features to train the intelligent control model, and adjusts the model parameters through the adaptive learning rate method; by training the intelligent control model, a mapping relationship between key indicator data and process parameters is established, and through the adjustment of model parameters, the model can accurately predict the key indicators and results in the processing process; through the adaptive learning rate method, the learning rate can be automatically adjusted according to the parameter update situation to improve the model convergence speed and effect. Step S203 During the processing process, sensors are used to continuously obtain real-time data and update it to the constructed intelligent control model; through the real-time updated key indicator data, the intelligent control model can perform real-time analysis and prediction to evaluate the status and trend of the current processing process; according to the analysis results of the model, the process parameters are adjusted in real time to optimize the efficiency and quality of the processing process; to achieve real-time monitoring and control of the processing process, improve production efficiency and product quality stability.

[0081] In summary, this embodiment can realize the control and optimization of process parameters; through preprocessing and feature extraction, meaningful features can be extracted from key indicator data, and a mapping relationship between the model and process parameters can be established; by training the intelligent control model and adjusting the model parameters, accurate prediction of key indicators and results can be achieved; through real-time data updating and model analysis, real-time monitoring and optimization adjustment of the processing process can be achieved; the control and optimization process of process parameters improves production efficiency and product quality stability, reduces production costs and resource consumption, and improves the competitiveness and sustainable development capabilities of the enterprise.

[0082] Example 4: Figure 4As shown, based on Example 3, the process of selecting target features that reach an impact threshold on the predicted target provided by the embodiment of the present invention includes the following steps:

[0083] S2011: Arrange the preprocessed key indicator data in chronological order, divide the time series data into multiple sliding windows, each window contains a preset number of continuous data points; extract features of the key indicator data in each sliding window according to the number and time interval of the data points in each sliding window;

[0084] S2012: Use the Pearson correlation coefficient to calculate the correlation coefficient between each feature in the preprocessed key indicator data and the predicted target; according to the calculated correlation coefficient value, sort the features from high to low according to the correlation with the target;

[0085] S2013: Select a correlation coefficient threshold in the sorted feature list as an influence threshold, and select features whose correlation coefficients exceed the threshold as features that influence the target based on the correlation coefficient threshold; combine the extracted features into a feature vector, and each sliding window corresponds to a feature vector.

[0086] The working principle and beneficial effects of the above technical solution are as follows: the present embodiment first arranges the preprocessed key indicator data in chronological order, divides the time series data into multiple sliding windows, each window contains a preset number of continuous data points; according to the number of data points and the time interval in each sliding window, extracts features from the key indicator data in each sliding window; secondly, uses the Pearson correlation coefficient to calculate the correlation coefficient between each feature in the preprocessed key indicator data and the predicted target; according to the calculated correlation coefficient value, sorts the features from high to low according to the correlation with the target; finally, selects a correlation coefficient threshold in the sorted feature list as the influence threshold, and selects the features with the correlation coefficient exceeding the threshold as the features that influence the target according to the correlation coefficient threshold; combines the extracted features into feature vectors, and each sliding window corresponds to one feature vector. Step S2011 of the above solution converts the time series data into a feature vector representation in the sliding window for subsequent feature analysis and feature selection; converts the time series data into a feature vector of fixed dimension, so that the data is comparable and computable in the feature space, providing a basis for subsequent feature analysis and selection. Step S2012 evaluates the linear correlation between the features and the predicted target, and obtains the correlation coefficient between each feature and the target; understands the degree of association between each feature and the predicted target, and provides a reference for subsequent feature selection; features with higher correlation coefficients may be more likely to have an impact on the target and help improve the prediction accuracy of the model. Step S2013 selects features that reach the impact threshold on the predicted target according to the preset correlation coefficient threshold; screens out features that have a higher impact on the predicted target, reduces the redundancy and noise of the feature dimension, and improves the prediction accuracy and generalization ability of the model. At the same time, selecting features with actual impact can better understand the key factors of the processing process and provide guidance and decision-making basis for adjusting process parameters.

[0087] Example 5: Figure 5 As shown, based on Example 3, the construction process of the intelligent control model provided by the embodiment of the present invention includes the following steps:

[0088] S2021: Select the neural network model as the intelligent control model, divide the preprocessed key indicator data set into a training set and a validation set in a ratio of 7:3; normalize the features in the training set and the validation set;

[0089] S2022: Use the key indicator data and corresponding target features in the training set to train the neural network model using the back propagation algorithm; use the adaptive learning rate method to adjust the parameters of the intelligent control model, and automatically adjust the learning rate according to the performance and convergence of the intelligent control model;

[0090] S2023: Use the key indicator data in the validation set and the trained intelligent control model to evaluate and verify the model. The evaluation indicators include prediction error, accuracy, and recall rate.

[0091] The working principle and beneficial effects of the above technical solution are as follows: In this embodiment, a neural network model is first selected as an intelligent control model, and the preprocessed key indicator data set is divided into a training set and a validation set according to a ratio of 7:3; the features in the training set and the validation set are normalized; secondly, the key indicator data in the training set and the corresponding target features are used to train the neural network model using the back propagation algorithm; the parameters of the intelligent control model are adjusted using the adaptive learning rate method, and the learning rate is automatically adjusted according to the performance and convergence of the intelligent control model; finally, the key indicator data in the validation set and the trained intelligent control model are used to evaluate and verify the model, and the evaluation indicators include prediction error, accuracy and recall rate, etc. Step S2021 of the above solution selects a suitable neural network model as the intelligent control model, and divides the data set into a training set and a validation set; the normalization process makes the feature data within the same scale range, which is conducive to the training and convergence of the model; it provides preparation for subsequent model training and evaluation, ensures that the model is trained and verified on a suitable data set, and improves the performance and generalization ability of the model. Step S2022 uses the back propagation algorithm to train the neural network model and optimize the parameters of the model through the data of the training set; the adaptive learning rate method can automatically adjust the learning rate according to the performance and convergence of the model to better guide the training process of the model; through model training, the intelligent control model can learn the laws and patterns of the processing process from historical data, and gradually optimize the parameters of the model to improve the model's prediction and control capabilities for the processing process. Step S2023 uses the data of the validation set to evaluate the performance and generalization ability of the trained intelligent control model; by calculating the prediction error, accuracy, recall rate and other evaluation indicators, the prediction accuracy of the model and the adaptability of the model can be understood; through the evaluation and verification of the model, the performance of the trained intelligent control model can be checked to determine whether the model achieves the expected control effect, providing reference and basis for the application and real-time adjustment of the model. At the same time, the model can be improved and optimized according to the evaluation results to improve the prediction and control capabilities of the model.

[0092] Example 6: Figure 6 As shown, based on Example 5, the process of automatically adjusting the size of the learning rate provided in the embodiment of the present invention includes the following steps:

[0093] S20221: Randomly initialize the weights and biases of the neural network model; input the key indicator data in the training set into the neural network model, and obtain the output of the neural network model through calculation;

[0094] S20222: Compare the output of the neural network model with the target features in the training set, calculate the value of the loss function, and measure the prediction error of the neural network model; according to the value of the loss function, use the chain rule to calculate the gradient of the neural network model parameters to the loss function, back-propagate the gradient from the output layer to the input layer, and update the parameters of the neural network model; according to the calculated gradient, update the weights and biases of the neural network model until the preset number of training rounds is reached;

[0095] S20223: Adaptively adjust the learning rate according to the historical gradient of the key indicator data, introduce a decay factor, perform exponential weighted averaging on the historical gradient of the key indicator data, and realize automatic adjustment of the learning rate changes.

[0096] The working principle and beneficial effects of the above technical solution are as follows: the embodiment first randomly initializes the weights and biases of the neural network model; inputs the key indicator data in the training set into the neural network model, and obtains the output of the neural network model through calculation; secondly, the output of the neural network model is compared with the target features in the training set, the value of the loss function is calculated, and the prediction error of the neural network model is measured; according to the value of the loss function (cross entropy loss), the gradient of the neural network model parameters to the loss function is calculated using the chain rule, and the gradient is back-propagated from the output layer to the input layer to update the parameters of the neural network model; according to the calculated gradient, the weights and biases of the neural network model are updated until the preset number of training rounds is reached; finally, the learning rate is adaptively adjusted according to the historical gradient of the key indicator data, and the attenuation factor (preset fixed attenuation factor) is introduced to perform exponential weighted averaging on the historical gradient of the key indicator data to achieve automatic adjustment of the change of the learning rate. Step S20221 of the above scheme initializes the parameters of the neural network model, inputs the data in the training set into the model, calculates and obtains the output of the model; provides a starting point for subsequent model training and adjustment, inputs the data into the model, and obtains the initial prediction results of the model. Step S20222 calculates the loss function and gradient, and uses the back propagation algorithm to update the parameters of the neural network model to reduce the prediction error of the model; by updating the model parameters, the neural network model can be gradually optimized to improve the prediction accuracy of key indicator data and the generalization ability of the model. Step S20223 adaptively adjusts the learning rate according to the historical gradient of the key indicator data through the exponential weighted average method; dynamically adjusts the learning rate according to the performance and convergence of the model to better guide the training process of the model; a larger learning rate can speed up the convergence of the model, and a smaller learning rate can avoid overfitting and oscillation, and improve the stability and generalization ability of the model. By adaptively adjusting the learning rate, the training of the model can be made more efficient and accurate.

[0097] Example 7: Figure 7As shown, based on Example 6, the process of updating the weights and biases of the neural network model provided in the embodiment of the present invention includes the following steps:

[0098] S202221: For each training sample, input the input data into the neural network, perform forward propagation calculation, and obtain the output of the neural network model; calculate the gradient of the loss function between the output of the neural network model and the actual label. According to the chain rule, calculate the gradient of each layer from the output layer to the input layer in turn;

[0099] S202222: Use the calculated gradient to update the weights and biases of the model. The specific update rule is: new weight = original weight - learning rate * gradient; update each sample in the training data set;

[0100] S202223: Continue iterative training until the preset number of training rounds or convergence condition is reached.

[0101] The working principle and beneficial effects of the above technical solution are as follows: First, for each training sample, the input data is input into the neural network, and the forward propagation calculation is performed to obtain the output of the neural network model; the gradient of the loss function between the output of the neural network model and the actual label is calculated according to the chain rule, from the output layer to the input layer, and the gradient of each layer is calculated in turn; secondly, the calculated gradient is used to update the weight and bias of the model. The specific update rule is: new weight = original weight - learning rate * gradient; update each sample in the training data set; finally, continue iterative training until the preset number of training rounds or convergence conditions are reached. Step S202221 of the above scheme calculates the output of the neural network model on each training sample and calculates the gradient of the loss function. By calculating the gradient, the degree of influence of the current model parameters on the loss function can be understood; the gradient of each layer is calculated in order to use these gradients to update the parameters of the model in subsequent steps; the output of the model and the gradient of the loss function are calculated on each training sample to understand the performance and error of the model on the current sample; through the calculation of the gradient, the direction and guidance can be provided for the subsequent parameter update to gradually reduce the value of the loss function and improve the performance of the model. Step S202222 uses the calculated gradient to update the weight and bias of the model according to the update rule; each sample is updated with parameters to gradually adjust the parameters of the model so that it can better fit the training data; through the update of parameters, the model can be gradually optimized, the value of the loss function can be reduced, and the degree of fit of the model to the training data can be improved; through the update of each sample, the parameters of the model can be gradually adjusted so that it can better adapt to the characteristics of the training data and improve the prediction accuracy and generalization ability of the model. Step S202223 determines whether to continue the training iteration of the model according to the preset number of training rounds or convergence conditions; if the preset number of training rounds is not reached or the model has not converged, the model training is continued; through multiple iterations of training, the parameters of the model are gradually adjusted to continuously optimize the performance of the model; through iterative training, the model can better learn the characteristics of the training data and improve the generalization ability and prediction accuracy of the model. At the same time, by setting the convergence conditions, the stop timing of the training can be effectively controlled to avoid the occurrence of overfitting.

[0102] Example 8: Figure 8 As shown, based on Example 6, the process of realizing automatic adjustment of the change of learning rate provided by the embodiment of the present invention includes the following steps:

[0103] S202231: Set an initial learning rate value for the initial stage of neural network model training; in each iterative training process, calculate the gradient of the loss function with respect to the parameters based on the current neural network model parameters and training data;

[0104] S202232: Update the weights and biases of the neural network model using the calculated gradients according to stochastic gradient descent with momentum; maintain an exponentially weighted average of the historical gradients of a key indicator data; apply a decay factor to adjust the current learning rate based on the exponentially weighted average of the historical gradients;

[0105] S202233: Repeat the process through multiple iterations until the preset convergence condition is reached.

[0106] The working principle and beneficial effects of the above technical solution are as follows: the present embodiment first sets an initial learning rate value for the initial stage of neural network model training; in each iterative training process, the gradient of the loss function with respect to the parameters is calculated according to the current neural network model parameters and training data; secondly, the calculated gradient is used to update the weights and biases of the neural network model according to the stochastic gradient descent with momentum; the exponential weighted average of the historical gradients of a key indicator data is maintained; an attenuation factor is applied to adjust the current learning rate according to the exponential weighted average of the historical gradients; finally, it is repeatedly executed through multiple iterations until the preset convergence condition is reached. Step S202231 of the above scheme sets the initial learning rate, and calculates the gradient of the loss function with respect to the parameters according to the current model parameters and training data; the setting of the initial learning rate affects the initial speed and direction of model training, and the calculation of the gradient is used for parameter update and learning rate adjustment in subsequent steps; the initial learning rate is set to adapt to the initial stage of model training; by calculating the gradient, the influence of the current model parameters on the loss function can be understood, which provides a basis for subsequent parameter update and learning rate adjustment. Step S202232 uses the calculated gradient to update the weights and biases of the neural network model according to the stochastic gradient descent method with momentum. At the same time, the exponential weighted average of the historical gradients of the key indicator data is maintained for subsequent learning rate adjustment; the parameters of the model are gradually adjusted by updating the parameters and introducing the momentum, so that it can better fit the training data; by maintaining the exponential weighted average of the historical gradient, the current learning rate can be dynamically adjusted to adapt to the changes in the gradient and improve the efficiency and stability of model training. Step S202233 continuously updates the model parameters and adjusts the learning rate through multiple iterations of training until the preset convergence conditions are reached; through repeated iterations, the model can be gradually optimized, the value of the loss function can be reduced, and the degree of fit of the model to the training data can be improved; through multiple iterations of training, the parameters and learning rate of the model are gradually adjusted to continuously improve the performance of the model; by setting the convergence conditions, the timing of stopping the training can be controlled to avoid the occurrence of overfitting. At the same time, through parameter updating and learning rate adjustment, the generalization ability and prediction accuracy of the model can be improved, so that it can better adapt to new data.

[0107] Example 9: Fig. 9As shown, based on Example 1, the process of identifying and correcting deviations or defects in processing provided by the embodiment of the present invention includes the following steps:

[0108] S301: Obtain images of products in the process of processing through sensors or cameras, analyze and compare actual processing results using image matching using computer vision algorithms, compare actual processing results with expected results, and identify deviations or defects in processing;

[0109] S302: Establish an abnormality detection and early warning mechanism using the key indicator data of the intelligent control model and the sensor; by monitoring the abnormal changes of the key indicator data, including exceeding the set threshold or large differences from historical data, identify deviations or defects in the processing and issue early warnings in a timely manner;

[0110] S303: Based on the identified processing deviations or defects, the process parameters are automatically adjusted using an intelligent control model to achieve automatic control and correction of the processing process; based on the identification results, process parameters such as clamping force, tool path and feed speed are automatically adjusted.

[0111] The working principle and beneficial effects of the above technical solution are as follows: This embodiment first obtains product images during processing through sensors or cameras, uses image matching of computer vision algorithms to analyze and compare actual processing results, compares actual processing results with expected results, and identifies deviations or defects in processing; secondly, uses intelligent control models and key indicator data of sensors to establish an abnormality detection and early warning mechanism; by monitoring abnormal changes in key indicator data, including exceeding set thresholds or large differences from historical data, deviations or defects in processing are identified, and early warnings are issued in a timely manner; finally, based on the identified processing deviations or defects, the intelligent control model is used to automatically adjust the process parameters to achieve automatic control and correction of the processing process; according to the recognition results, process parameters such as clamping force, tool path and feed speed are automatically adjusted. In step S301 of the above scheme, image recognition and matching algorithms are used to analyze and compare the product images during the processing process, and the difference between the actual processing results and the expected results is identified; through image matching, deviations or defects in the processing process, such as dimensional errors, surface defects, etc., can be detected; through image recognition and comparison, deviations or defects in the processing can be found in time; it can help operators or systems to quickly find problems and take corresponding corrective measures to avoid unqualified products from entering the next process or being finally delivered to customers, thereby improving product quality and customer satisfaction. In step S302, an abnormality detection and early warning mechanism is established using intelligent control models and sensor data; by monitoring key indicator data, such as temperature, pressure, vibration, etc., abnormal changes in the processing process, situations that exceed the set threshold or differ greatly from historical data can be identified, and early warning signals can be issued; through the abnormality detection and early warning mechanism, deviations or defects in the processing process can be monitored in real time, helping to find production abnormalities in time, reducing the generation of defective products and product waste. At the same time, timely early warnings can remind operators or systems to take corresponding measures to correct them, thereby ensuring the stability of the processing process and product quality. Step S303 uses an intelligent control model to automatically adjust process parameters based on the identified processing deviations or defects; automatically adjust process parameters such as clamping force, tool path, feed speed, etc. according to the identification results to achieve automatic control and correction of the processing process; automatically adjust process parameters to achieve automatic control and correction of the processing process; accurate process parameter adjustment can eliminate deviations or defects in processing and improve product accuracy and quality. At the same time, automatic control and correction can reduce operator intervention and human errors and improve the stability and efficiency of the processing process.

[0112] Example 10: Fig.10 As shown, on the basis of Embodiments 1 to 9, the intelligent control system of the anchor processing equipment provided by the embodiment of the present invention comprises:

[0113] The data acquisition module is responsible for installing sensors on the anchor processing equipment. The sensors monitor the key indicator data such as the processing status, process parameters and product quality of the anchor processing equipment in real time; the key indicator data collected by the sensors are sent to the intelligent control model for analysis through the remote monitoring platform;

[0114] The data processing module is responsible for establishing an intelligent control model for the processing process based on historical key indicator data; using the intelligent control model to perform real-time processing analysis and optimization of key indicator data, and to control and optimize process parameters;

[0115] The parameter adjustment module is responsible for importing the process parameters into the anchor processing equipment. The anchor processing equipment performs automatic clamping and automatic processing according to the process parameters to realize automatic clamping and automatic processing. At the same time, combined with sensors and image recognition, it automatically detects and provides real-time feedback on product quality during the processing, compares the actual processing results, and identifies and corrects deviations or defects in the processing.

[0116] The working principle and beneficial effects of the above technical solution are as follows: the data acquisition module of this embodiment installs sensors on the anchor processing equipment, and the sensors monitor the key indicator data such as the processing status, process parameters and product quality of the anchor processing equipment in real time; the key indicator data collected by the sensor is sent to the intelligent control model for analysis through the remote monitoring platform; the data processing module establishes an intelligent control model of the processing process based on the historical key indicator data; the intelligent control model is used to perform real-time processing process analysis and optimization on the key indicator data, and control and optimize the process parameters; the parameter adjustment module imports the process parameters into the anchor processing equipment, and the anchor processing equipment performs automatic clamping and automatic processing according to the process parameters, realizing automatic clamping and automatic processing; at the same time, combined with sensors and image recognition, the product quality in the processing process is automatically detected and real-time feedback is provided, the actual processing results are compared, and the deviations or defects in the processing are identified and corrected. The data acquisition module of the above scheme monitors and collects key indicator data in real time, provides real-time status and parameters in the processing process; provides real-time data support for the intelligent control model, realizes real-time monitoring and data analysis of the processing process, and provides a basis for subsequent optimization and adjustment. The data processing module establishes an intelligent control model to analyze, optimize and predict key indicator data of the processing process; through data analysis and optimization, it realizes intelligent control of the processing process, improves processing efficiency and product quality, and reduces resource waste and human errors. The parameter adjustment module automatically adjusts the process parameters to achieve automatic clamping and automatic processing; through sensors and image recognition, it detects product quality in real time and identifies deviations or defects in processing; improves production efficiency and product quality stability; automatically corrects processing deviations or defects to reduce the generation of defective products; and provides real-time feedback on product quality information, taking timely measures to improve the processing process.

[0117] Embodiment 11: Based on Embodiment 10, the data acquisition module provided in the embodiment of the present invention comprises:

[0118] The position determination submodule is responsible for obtaining the content of the key indicator data, determining the target position of the key indicator data according to the content, marking the target position according to the preset order of the key indicator data, and establishing a corresponding relationship list between the target position and the content of the key indicator data; the target position is the processing area or heating area of ​​the equipment;

[0119] The relationship correspondence submodule is responsible for determining the location of establishing the IoT communication node based on the corresponding relationship list. The IoT communication node is the location where the sensor is placed;

[0120] The connection establishment submodule is responsible for connecting various IoT communication nodes, aggregating them to the IoT bus, and then transmitting key indicator data to the remote monitoring platform, which is connected to the input end of the intelligent control model through the Internet.

[0121] The working principle and beneficial effects of the above technical solution are as follows: the position determination submodule of this embodiment obtains the content of the key indicator data, determines the target position of the key indicator data according to the content, marks the target position according to the preset sorting of the key indicator data, and establishes a corresponding relationship list between the target position and the content of the key indicator data; the target position is the processing area or heating area of ​​the equipment; the relationship correspondence submodule determines the location of the Internet of Things communication node according to the corresponding relationship list, and the Internet of Things communication node is the location where the sensor is placed; the connection establishment submodule connects each Internet of Things communication node, summarizes it to the Internet of Things bus, and then transmits the key indicator data to the remote monitoring platform, and the remote monitoring platform is connected to the input end of the intelligent control model through the Internet. The position determination submodule of the above scheme determines the target position of the key indicator data, and establishes a corresponding relationship between the target position and the data content, providing accurate location information for subsequent data collection and processing; by accurately determining the target position of the key indicator data, it can ensure that the collected data is associated with the specific processing area or heating area, and improve the reliability and accuracy of the data. The relationship correspondence submodule determines the location of the sensor according to the corresponding relationship list, ensuring that the sensor can accurately collect the key indicator data; by determining the location of the sensor, it can ensure that the sensor can collect the key indicator data at the correct location, and improve the accuracy and reliability of data collection. The connection establishment submodule connects each IoT communication node to the IoT bus to realize the centralized aggregation and transmission of data to the remote monitoring platform; by establishing a connection, the centralized management and transmission of key indicator data are realized, so that the remote monitoring platform can obtain data in a timely manner and interact with the intelligent control model for analysis, thus realizing real-time monitoring and intelligent control.

[0122] In summary, the submodules of the data acquisition module of this embodiment cooperate with each other to ensure the accurate collection and transmission of key indicator data to the remote monitoring platform; provide a basis for subsequent data analysis, optimization and intelligent control; through real-time acquisition and analysis of key indicator data, the process parameters of the anchor processing equipment can be adjusted and optimized in a timely manner to improve production efficiency and product quality. At the same time, the connection of the remote monitoring platform can realize remote monitoring and remote control, and improve the flexibility and efficiency of production management.

[0123] Embodiment 12: Based on Embodiment 10, the data processing module provided in the embodiment of the present invention comprises:

[0124] The feature extraction submodule is responsible for obtaining key indicator data including processing status, process parameters and product quality collected by historical sensors, and performing preprocessing operations such as cleaning and normalization on the key indicator data; extracting time series features from the preprocessed keyword indicator data, and selecting target features that reach the impact threshold on the prediction target;

[0125] The model training submodule is responsible for training the intelligent control model using the target features and adjusting the intelligent control model parameters through the adaptive learning rate method;

[0126] The data update submodule is responsible for continuously acquiring real-time data from sensors during the processing and updating it to the constructed intelligent control model. The intelligent control model uses the real-time updated key indicator data to perform real-time analysis and prediction to evaluate the status and trend of the current processing process. According to the analysis results of the intelligent control model, the process parameters are adjusted in real time to optimize the efficiency and quality of the processing process.

[0127] The working principle and beneficial effects of the above technical scheme are as follows: the feature extraction submodule of this embodiment obtains the key indicator data including processing status, process parameters and product quality collected by historical sensors, and performs preprocessing operations such as cleaning and normalization on the key indicator data; extracts time series features from the preprocessed keyword indicator data, and selects the target features that reach the impact threshold on the predicted target; the model training submodule uses the target features to train the intelligent control model, and adjusts the parameters of the intelligent control model through the adaptive learning rate method; the data update submodule continuously obtains real-time data from the sensor during the processing process and updates it to the constructed intelligent control model; using the real-time updated key indicator data, the intelligent control model performs real-time analysis and prediction, and evaluates the status and trend of the current processing process; according to the analysis results of the intelligent control model, the process parameters are adjusted in real time to optimize the efficiency and quality of the processing process. The feature extraction submodule of the above scheme preprocesses the historical key indicator data to make it meet the model requirements, and extracts time series features from it; through preprocessing and feature extraction, the key indicator data is made more interpretable and usable, providing a better data basis for subsequent model training. The model training submodule uses the target features to train the intelligent control model, so that it can be adaptively adjusted according to the actual situation, improving the accuracy and generalization ability of the model; through model training, an intelligent control model for anchor processing equipment can be established to achieve analysis and optimization of key indicator data, providing a basis for subsequent real-time control and prediction. The data update submodule obtains data from sensors in real time and updates it to the intelligent control model to achieve real-time monitoring and control of the processing process; through real-time updating of key indicator data and model analysis, the status and trend of the current processing process can be timely evaluated, and the process parameters can be adjusted in real time according to the model analysis results, the efficiency and quality of the processing process can be optimized, and the production efficiency and product quality stability can be improved.

[0128] In summary, the submodules of the data processing module of this embodiment cooperate with each other to preprocess, extract features, train models and update in real time key indicator data, and realize intelligent control and optimization of the processing process. Through the work of the data processing module, the stability and efficiency of the processing process can be improved, the generation of defective products can be reduced, and product quality and production efficiency can be improved.

[0129] Embodiment 13: Based on Embodiment 10, the parameter adjustment module provided in the embodiment of the present invention includes:

[0130] The result comparison submodule is responsible for acquiring images of products during processing through sensors or cameras, analyzing and comparing the actual processing results using image matching using computer vision algorithms, comparing the actual processing results with the expected results, and identifying deviations or defects in processing;

[0131] The abnormal monitoring submodule is responsible for establishing an abnormal detection and early warning mechanism by using the key indicator data of the intelligent control model and sensors. It can identify deviations or defects in processing and issue early warnings in a timely manner by monitoring abnormal changes in key indicator data, including those exceeding the set threshold or having large differences with historical data.

[0132] The result recognition submodule is responsible for automatically adjusting the process parameters based on the identified processing deviations or defects using the intelligent control model to achieve automatic control and correction of the processing process; according to the recognition results, it automatically adjusts the process parameters such as clamping force, tool path and feed speed.

[0133] The working principle and beneficial effects of the above technical solution are as follows: the result comparison submodule of this embodiment obtains the product image in the processing process through a sensor or camera, uses the image matching of the computer vision algorithm to analyze and compare the actual processing results, compares the actual processing results with the expected results, and identifies the deviation or defect in the processing; the abnormal monitoring submodule uses the key indicator data of the intelligent control model and the sensor to establish an abnormal detection and early warning mechanism; by monitoring the abnormal changes of the key indicator data, including exceeding the set threshold or a large difference with the historical data, the deviation or defect in the processing is identified, and an early warning is issued in time; the result identification submodule uses the intelligent control model to automatically adjust the process parameters based on the identified processing deviation or defect, and realizes automatic control and correction of the processing process; according to the identification result, the process parameters such as clamping force, tool path and feed speed are automatically adjusted. The result comparison submodule of the above scheme compares the actual processing results with the expected results through image matching analysis, identifies the deviation or defect in the processing; through result comparison, the deviation or defect in the processing can be found in time, helping the operator or system to quickly find problems and take corrective measures to improve product quality and customer satisfaction. The abnormal monitoring submodule identifies abnormal changes in key indicator data during the processing process, determines deviations or defects in the processing, and issues early warnings in a timely manner through abnormal monitoring and early warning mechanisms. Through abnormal monitoring and early warning mechanisms, production abnormalities can be discovered in a timely manner, reducing the production of defective products and product waste. At the same time, timely early warnings can remind operators or systems to take appropriate measures to correct them, thereby ensuring the stability of the processing process and product quality. The result recognition submodule automatically adjusts the process parameters based on the recognition results to achieve automatic control and correction of the processing process; by automatically adjusting the process parameters, automatic control and correction of the processing process can be achieved. Through accurate process parameter adjustment, deviations or defects in the processing can be eliminated, and the accuracy and quality of the product can be improved. At the same time, automatic control and correction can reduce operator intervention and human errors, and improve the stability and efficiency of the processing process.

[0134] In summary, the submodules of the parameter adjustment module of this embodiment cooperate with each other to identify deviations or defects in processing and realize automatic control and correction. Through result comparison, abnormality monitoring and result identification, product quality and production efficiency can be improved, the generation of defective products can be reduced, and customer satisfaction can be improved.

[0135] Example 14: Fig.11 As shown, on the basis of Embodiment 1 to Embodiment 13, the intelligent control device of the anchor processing equipment provided by the embodiment of the present invention comprises:

[0136] Anchor processing equipment is responsible for laying out sensors, performing automatic clamping and automatic processing according to process parameters, and realizing automatic clamping and automatic processing;

[0137] Sensors are responsible for real-time monitoring of key indicator data such as the processing status, process parameters and product quality of anchor processing equipment; and the key indicator data collected by the sensors are sent to the intelligent control model for analysis through the remote monitoring platform;

[0138] The remote monitoring platform is responsible for realizing remote control and monitoring of anchor processing equipment, sensors and intelligent control models; combined with sensors and image recognition, it can automatically detect and provide real-time feedback on product quality during the processing, compare the actual processing results, and identify and correct deviations or defects in the processing;

[0139] The intelligent control model is responsible for real-time processing analysis and optimization of key indicator data, and for controlling and optimizing process parameters.

[0140] The working principle and beneficial effects of the above technical solution are as follows: The specific operation process of this embodiment is as follows: The anchor processing equipment is responsible for arranging sensors, and performs automatic clamping and automatic processing according to the process parameters to ensure the consistency and accuracy of the processing process; the sensor monitors the key indicator data such as the processing status, process parameters and product quality of the anchor processing equipment in real time, and sends the data to the remote monitoring platform for analysis. The sensor can provide real-time data support and provide a basis for subsequent analysis and optimization. The remote monitoring platform realizes remote control and monitoring of the anchor processing equipment, sensors and intelligent control models. Through the connection with the sensors and the intelligent control model, the remote monitoring platform can obtain key indicator data in real time to monitor and analyze the processing process. At the same time, through image recognition technology, automatic detection and real-time feedback of product quality are realized, and deviations or defects in processing are identified and corrected. The intelligent control model performs real-time processing process analysis and optimization on the key indicator data collected by the sensor. By analyzing the key indicator data, the intelligent control model can evaluate the state and trend of the processing process, and then control and optimize the process parameters to improve the processing efficiency and the stability of product quality.

[0141] The automatic clamping and automatic processing of the anchor processing equipment of this embodiment can improve production efficiency and processing accuracy, reduce human errors and labor intensity; the sensor monitors key indicator data in real time, provides real-time data support, and ensures data accuracy and reliability; the remote monitoring platform realizes remote control and monitoring, and can obtain data and status of the processing process in time, and realize remote management and decision-making; image recognition technology realizes automatic detection and real-time feedback of product quality, identifies deviations or defects in processing, corrects problems in time, and improves product quality; the intelligent control model analyzes and optimizes key indicator data in real time, realizes intelligent control of the processing process, and improves processing efficiency and product quality stability.

[0142] In summary, the intelligent control device of the anchor processing equipment of this embodiment realizes real-time monitoring, analysis and optimization of the processing process through the coordinated work of the deployed sensors, remote monitoring platform and intelligent control model, thereby improving the processing efficiency, product quality and flexibility of production management. At the same time, through automatic clamping and automatic processing, human errors are reduced and production efficiency and processing accuracy are improved.

[0143] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technology, the present invention is also intended to include these changes and variations.

Claims

1. An intelligent control method for anchor processing equipment, characterized in that: The following steps are involved: Install sensors on the anchor processing equipment to monitor the processing status, process parameters and key indicator data of product quality of the anchor processing equipment in real time; send the key indicator data collected by the sensors to the intelligent control model for analysis through the remote monitoring platform; Based on historical key indicator data, an intelligent control model of the processing process is established; the intelligent control model is used to perform real-time processing analysis and optimization of key indicator data, and to control and optimize process parameters; The process parameters are imported into the anchor processing equipment, and the anchor processing equipment performs automatic clamping and automatic processing according to the process parameters to realize automatic clamping and automatic processing; at the same time, combined with sensors and image recognition, the product quality in the processing process is automatically detected and real-time feedback is provided, and the actual processing results are compared to identify and correct deviations or defects in the processing.

2. The intelligent control method for anchor processing equipment according to claim 1, characterized in that: The process of collecting key indicator data includes the following steps: Obtain the content of the key indicator data, determine the target location of the key indicator data according to the content, mark the target location according to the preset order of the key indicator data, and establish a corresponding relationship list between the target location and the content of the key indicator data; The target location is the processing area or heating area of ​​the equipment; According to the correspondence list, determine the location of establishing the IoT communication node, where the IoT communication node is the location where the sensor is placed; Connect each IoT communication node and aggregate it to the IoT bus, then transmit key indicator data to the remote monitoring platform, which is connected to the input end of the intelligent control model through the Internet.

3. The intelligent control method for anchor processing equipment according to claim 1, characterized in that: The process of controlling and optimizing process parameters includes the following steps: Obtain the key indicator data including processing status, process parameters and product quality collected by historical sensors, and perform cleaning and normalization preprocessing operations on the key indicator data; extract time series features from the preprocessed keyword indicator data, and select target features that reach the impact threshold on the predicted target; The intelligent control model is trained using target features, and the parameters of the intelligent control model are adjusted using an adaptive learning rate method; During the processing, real-time data is continuously acquired from sensors and updated to the constructed intelligent control model. Using the real-time updated key indicator data, the intelligent control model performs real-time analysis and prediction to evaluate the status and trend of the current processing process. Based on the analysis results of the intelligent control model, the process parameters are adjusted in real time to optimize the efficiency and quality of the processing process.

4. The intelligent control method for anchor processing equipment according to claim 3, characterized in that: The process of selecting target features that reach a threshold of influence on the prediction target includes the following steps: Arrange the preprocessed key indicator data in chronological order, divide the time series data into multiple sliding windows, each window contains a preset number of continuous data points; extract features of the key indicator data in each sliding window according to the number and time interval of data points in each sliding window; Use the Pearson correlation coefficient to calculate the correlation coefficient between each feature in the preprocessed key indicator data and the predicted target; according to the calculated correlation coefficient value, sort the features from high to low according to their correlation with the target; A correlation coefficient threshold is selected in the sorted feature list as the influence threshold. Based on the correlation coefficient threshold, features with correlation coefficients exceeding the threshold are selected as features that influence the target. The extracted features are combined into feature vectors, and each sliding window corresponds to a feature vector.

5. The intelligent control method for anchor processing equipment according to claim 3, characterized in that: The construction process of the intelligent control model includes the following steps: Select the neural network model as the intelligent control model, divide the preprocessed key indicator data set into a training set and a validation set in a ratio of 7:3; normalize the features in the training set and the validation set; Use the key indicator data and corresponding target features in the training set to train the neural network model using the back propagation algorithm; use the adaptive learning rate method to adjust the parameters of the intelligent control model, and automatically adjust the learning rate according to the performance and convergence of the intelligent control model; Use the key indicator data in the validation set and the trained intelligent control model to evaluate and verify the model. The evaluation indicators include prediction error, accuracy and recall rate.

6. The intelligent control method for anchor processing equipment according to claim 5, characterized in that: The process of automatically adjusting the learning rate size includes the following steps: Randomly initialize the weights and biases of the neural network model; input the key indicator data in the training set into the neural network model, and obtain the output of the neural network model through calculation; Compare the output of the neural network model with the target features in the training set, calculate the value of the loss function, and measure the prediction error of the neural network model; according to the value of the loss function, use the chain rule to calculate the gradient of the neural network model parameters to the loss function, back-propagate the gradient from the output layer to the input layer, and update the parameters of the neural network model; according to the calculated gradient, update the weights and biases of the neural network model until the preset number of training rounds is reached; The learning rate is adaptively adjusted according to the historical gradient of the key indicator data, and the attenuation factor is introduced to perform exponential weighted averaging on the historical gradient of the key indicator data to achieve automatic adjustment of the learning rate changes.

7. The intelligent control method for anchor processing equipment according to claim 6, characterized in that: The process of updating the weights and biases of a neural network model consists of the following steps: For each training sample, input data into the neural network, perform forward propagation calculations, and obtain the output of the neural network model; calculate the gradient of the loss function between the output of the neural network model and the actual label. According to the chain rule, calculate the gradient of each layer from the output layer to the input layer in turn; Use the calculated gradients to update the model’s weights and biases for each sample in the training dataset. Continue iterating training until the preset number of training rounds or convergence condition is reached.

8. The intelligent control method for anchor processing equipment according to claim 7, characterized in that: The specific update rule for updating the model's weights and biases is: new weight = original weight - learning rate * gradient.

9. An intelligent control system for anchor processing equipment, characterized in that: Include: The data acquisition module is responsible for installing sensors on the anchor processing equipment. The sensors monitor the processing status, process parameters and key indicator data of product quality of the anchor processing equipment in real time; the key indicator data collected by the sensors are sent to the intelligent control model for analysis through the remote monitoring platform; The data processing module is responsible for establishing an intelligent control model for the processing process based on historical key indicator data; using the intelligent control model to perform real-time processing analysis and optimization of key indicator data, and to control and optimize process parameters; The parameter adjustment module is responsible for importing the process parameters into the anchor processing equipment. The anchor processing equipment performs automatic clamping and automatic processing according to the process parameters to realize automatic clamping and automatic processing. At the same time, combined with sensors and image recognition, it automatically detects and provides real-time feedback on product quality during the processing, compares the actual processing results, and identifies and corrects deviations or defects in the processing.

10. An intelligent control device for anchor processing equipment, characterized in that: Include: Anchor processing equipment is responsible for laying out sensors, performing automatic clamping and automatic processing according to process parameters, and realizing automatic clamping and automatic processing; Sensors are responsible for real-time monitoring of the processing status, process parameters and key indicator data of product quality of anchor processing equipment; and the key indicator data collected by the sensors are sent to the intelligent control model for analysis through the remote monitoring platform; The remote monitoring platform is responsible for realizing remote control and monitoring of anchor processing equipment, sensors and intelligent control models; combined with sensors and image recognition, it can automatically detect and provide real-time feedback on product quality during the processing, compare the actual processing results, and identify and correct deviations or defects in the processing; The intelligent control model is responsible for real-time processing analysis and optimization of key indicator data, and for controlling and optimizing process parameters.

Images