Stamping equipment running state real-time monitoring method and system based on sensor fusion

Abstract

The invention discloses a real-time monitoring method and system for the running state of stamping equipment based on sensor fusion, and relates to the technical field of stamping equipment monitoring, and the method comprises the specific steps of sensor deployment, synchronous data acquisition, data preprocessing and feature fusion, prediction and fault early warning and result display. According to the method, through historical monitoring data, a stamping part size deviation prediction formula and a fault early warning dynamic threshold value formula, the stamping part size deviation is predicted by combining the precision attenuation core sensitive characteristics of the equipment, meanwhile, the early warning threshold value is dynamically adjusted, the precision change trend of the equipment in the future is accurately mastered, and when the predicted value exceeds the dynamic early warning threshold value, the fault early warning is performed. According to the utility model, fault early warning is triggered, rapid response is realized through the audible and visual alarm, and meanwhile, a detailed monitoring curve and a precision change trend chart are displayed on the display screen, so that the innovation point not only improves the timeliness of fault discovery, but also provides a scientific basis for the adjustment of a production plan and the maintenance of equipment by predicting the future precision change trend.

Description

Technical Field

[0001] This invention relates to the field of stamping equipment monitoring technology, specifically to a method and system for real-time monitoring of the operating status of stamping equipment based on sensor fusion. Background Technology

[0002] As a key piece of equipment in modern manufacturing, stamping equipment is widely used in many industries such as automobiles, electronics, and home appliances. Its operating status directly affects product quality and production efficiency. With the advancement of Industry 4.0 and intelligent manufacturing, higher requirements are placed on the real-time monitoring and maintenance of stamping equipment. The traditional periodic maintenance mode can no longer meet the needs of efficient and precise production.

[0003] However, in existing technologies, the monitoring of stamping equipment often relies on a single sensor, resulting in limited data dimensions and making it difficult to comprehensively reflect the equipment status. For example, using only force sensors to monitor load changes ignores the impact of equipment vibration and mold positioning on accuracy; or relying solely on laser displacement sensors cannot capture dynamic load changes during equipment operation. Furthermore, existing methods are inadequate in terms of data synchronization, feature fusion, and predictive early warning, lacking comprehensive analysis of multi-dimensional data and dynamic early warning mechanisms, leading to delayed fault detection and affecting production efficiency and product quality. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method and system for real-time monitoring of the operating status of stamping equipment based on sensor fusion. This invention predicts the dimensional deviation of stamped parts by combining historical monitoring data, a formula for predicting dimensional deviations in stamped parts, and a dynamic threshold formula for fault early warning, along with the core sensitive characteristics of equipment precision decay. Simultaneously, it dynamically adjusts the early warning threshold, achieving accurate control over the future precision change trend of the equipment. When the predicted value exceeds the dynamic early warning threshold, a fault early warning is triggered, and a rapid response is achieved through an audible and visual alarm. At the same time, a display screen shows detailed monitoring curves and precision change trend graphs, providing intuitive and accurate decision support. This innovation not only improves the timeliness of fault detection but also provides a scientific basis for adjusting production plans and maintaining equipment by predicting future precision change trends, effectively ensuring production safety.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On the one hand, a method for real-time monitoring of the operating status of stamping equipment based on sensor fusion, the specific steps of which are as follows: Sensor deployment: Force sensors, laser displacement sensors, and triaxial acceleration sensors are deployed in key parts of the stamping equipment, and an industrial Ethernet data acquisition card is configured to establish communication with each sensor and control the signal transmission delay of each sensor. Data Synchronous Acquisition: After the stamping equipment is running stably, multi-dimensional data covering the entire stamping cycle is collected through various sensors. The data acquisition card signal synchronization mechanism is used to ensure that the data timing is consistent. After the collected data is verified, it is stored in the industrial database. Data preprocessing and feature fusion: Retrieve stored data from the industrial database and preprocess it, then extract load features, positioning features and vibration features, and perform fusion calculation on the three types of features through multi-sensor feature fusion formula to obtain the core sensitive features of equipment accuracy attenuation; Prediction and fault warning: Based on the core sensitive features of equipment accuracy decay and combined with historical monitoring data, the predicted value of size deviation is calculated by the formula for predicting the size deviation of stamped parts, and the dynamic warning threshold is calculated by the formula for dynamic warning threshold. The predicted value of size deviation is compared with the dynamic warning threshold to determine whether a fault warning is triggered and to predict the future trend of accuracy change. Results presentation: Based on fault warning information, predicted dimensional deviation values ​​and deviation growth rates, the system uses an audible and visual alarm to respond to warning signals, displays monitoring curves and accuracy change trend graphs on a screen, and generates maintenance suggestions in conjunction with the stamping equipment maintenance manual and production plan.

[0006] Furthermore, in the sensor deployment, the force sensor is installed at the connection between the slider and the upper die base of the stamping equipment, the laser displacement sensor is installed on both sides of the worktable of the stamping equipment with the laser beam aligned with the die positioning reference surface, and the triaxial acceleration sensors are respectively installed in the middle of the body of the stamping equipment, the lower end of the slider, and the four corners of the worktable. The installation positions of each sensor avoid the interference area of ​​the equipment movement; an industrial Ethernet data acquisition card that supports multi-channel parallel reception is configured to establish a connection with the communication interface matched with each sensor, control the signal transmission delay of each sensor within a uniform range, and determine the temporal correlation of the acquired data.

[0007] Furthermore, in the synchronous data acquisition, the multi-dimensional data includes load fluctuation data, mold positioning deviation data, and vibration signals of various parts of the equipment. Among them, the force sensor collects load fluctuation data, the laser displacement sensor collects mold positioning deviation data, and the triaxial acceleration sensor collects vibration signals of various parts of the equipment.

[0008] Furthermore, in the data preprocessing and feature fusion, the preprocessing operations include noise reduction, outlier removal, and data standardization; load features are extracted from the preprocessed load fluctuation data, mold positioning deviation data, and vibration signals from various parts of the equipment. Location features and vibration characteristics Then, the three types of features are fused and calculated using a multi-sensor feature fusion formula to obtain the core sensitive features of equipment accuracy attenuation. .

[0009] Furthermore, in the data preprocessing and feature fusion, the multi-sensor feature fusion formula is as follows: ,in, This is a core sensitive feature for the attenuation of equipment accuracy. For load characteristics, For location features, Vibration characteristics, , , The weighting coefficients for the features are determined using historical monitoring data.

[0010] Furthermore, in the prediction and fault warning, the formula for predicting the dimensional deviation of the stamped part is: ,in, For the first Predicted dimensional deviation of stamped parts at time [time]. For the first The core sensitive feature of equipment accuracy degradation at any given time, the first The timeframe refers to the real-time moment at which the accuracy prediction is performed. For the first The core sensitive feature of equipment accuracy degradation at any given time, the first This moment is a historical moment. For the first The time decay coefficient at any given moment is determined using historical monitoring data. For the first The time decay factor at time, The length of the sliding window is determined using historical monitoring data; and through Obtain the deviation growth rate.

[0011] Furthermore, in the prediction and fault early warning, the dynamic threshold formula for fault early warning is: ,in, For the first Dynamic fault warning threshold at any time, This serves as the basic threshold for the dimensional deviation of stamped parts. For the first The core sensitive feature of equipment accuracy degradation over time. The historical average of the core sensitive characteristic of equipment accuracy degradation during normal operation is determined through historical monitoring data. The standard deviation of the core sensitive characteristic of equipment accuracy degradation during normal operation is determined through historical monitoring data. The threshold adjustment coefficient is determined using historical monitoring data.

[0012] Furthermore, in the prediction and fault warning, and In contrast, when At that time, a fault warning is triggered, and based on continuous The trend of accuracy change in sequence prediction over future time.

[0013] On the other hand, a real-time monitoring system for the operating status of stamping equipment based on sensor fusion includes: Sensor deployment module: Deploy force sensors, laser displacement sensors, and triaxial acceleration sensors in key parts of the stamping equipment, and configure an industrial Ethernet data acquisition card to establish communication with each sensor and control the signal transmission delay of each sensor; Data synchronization acquisition module: After the stamping equipment is running stably, it collects multi-dimensional data covering the entire stamping cycle through various sensors, uses the signal synchronization mechanism of the acquisition card to ensure data consistency, and stores the collected data in the industrial database after verification. Data preprocessing and feature fusion module: retrieves stored data from the industrial database and preprocesses it, then extracts load features, positioning features and vibration features, and performs fusion calculation on the three types of features through multi-sensor feature fusion formula to obtain the core sensitive features of equipment accuracy attenuation; Prediction and Fault Warning Module: Based on the core sensitive features of equipment accuracy decay and combined with historical monitoring data, the module calculates the predicted value of size deviation using the formula for predicting the size deviation of stamped parts, calculates the dynamic warning threshold using the formula for dynamic warning threshold of fault warning, compares the predicted value of size deviation with the dynamic warning threshold to determine whether a fault warning is triggered and predicts the future trend of accuracy change. Results display module: Based on fault warning information, predicted dimensional deviation values ​​and deviation growth rates, the module uses an audible and visual alarm to respond to warning signals, displays monitoring curves and accuracy change trend graphs on a screen, and generates maintenance suggestions in conjunction with the stamping equipment maintenance manual and production plan.

[0014] Compared with existing technologies, this sensor fusion-based real-time monitoring method and system for the operating status of stamping equipment has the following advantages: I. This invention, by combining historical monitoring data, a formula for predicting stamped part dimensional deviations, and a dynamic threshold formula for fault early warning, with the core sensitive characteristics of equipment precision decay, predicts the dimensional deviations of stamped parts and dynamically adjusts the early warning threshold. This enables precise control over the future precision change trend of the equipment. When the predicted value exceeds the dynamic early warning threshold, a fault early warning is triggered, and a rapid response is achieved through an audible and visual alarm. Simultaneously, a display screen shows detailed monitoring curves and precision change trend graphs, providing intuitive and accurate decision support. This innovation not only improves the timeliness of fault detection but also provides a scientific basis for adjusting production plans and maintaining equipment by predicting future precision change trends, effectively ensuring production safety.

[0015] II. This invention achieves comprehensive and synchronous acquisition of multi-dimensional data, including load fluctuations, mold positioning deviations, and vibration signals from various parts of the equipment, by deploying force sensors, laser displacement sensors, and triaxial accelerometers at key parts of the stamping equipment. The force sensors capture dynamic load changes during the stamping process, the laser displacement sensors measure mold positioning deviations, and the triaxial accelerometers comprehensively reflect the equipment vibration. By using a multi-sensor feature fusion formula, the three types of features are fused and calculated to obtain the core sensitive features of equipment accuracy decay. This innovation not only overcomes the limitations of the limited dimensions of single sensor data but also improves the comprehensiveness and accuracy of equipment condition monitoring through the complementarity and verification of multi-source data, laying a data foundation for dimensional deviation prediction and fault early warning.

[0016] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0018] Figure 1 A flowchart of a method for real-time monitoring of the operating status of stamping equipment based on sensor fusion; Figure 2 This is a framework diagram of a real-time monitoring system for the operating status of stamping equipment based on sensor fusion. Figure 3 This is a framework diagram for prediction, fault warning, and result display in a real-time monitoring method for the operating status of stamping equipment based on sensor fusion. Detailed Implementation

[0019] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0020] Example 1: In the real-time monitoring of the operation status of automotive door inner panel stamping equipment, a site survey was first conducted to determine the installation points of each sensor. A force sensor was fixed at the connection between the slider and the upper die base of the stamping equipment, a position that can capture the load changes between the die and the sheet metal during the stamping process. A laser displacement sensor was mounted on each of the fixed supports on both sides of the worktable. The laser beam angle was repeatedly adjusted to ensure it was projected onto the positioning reference surface of the door inner panel stamping die, ensuring real-time capture of minute displacement deviations of the die. Six triaxial accelerometers were deployed in the middle of the machine body, at the lower end of the slider, and at the four corners of the worktable. All installation positions were pre-emptively avoided in areas of interference with the machine's motion, such as the crankshaft and connecting rod trajectory and the path of the feeding robot arm, to prevent sensor damage due to equipment operation. Simultaneously, an eight-channel industrial Ethernet data acquisition card was configured for the equipment, and the matching communication interfaces of each acquisition card and sensor were connected one by one. During debugging, the signal transmission delay of each sensor was uniformly controlled to be within the same range, laying a solid foundation for subsequent timing alignment of multi-source data. Figure 1 As shown.

[0021] After the stamping equipment completes pre-shift warm-up and no-load trial operation and enters a stable production state, the linkage acquisition program of sensors and data acquisition cards is activated. At this time, the production line produces door inner panels at a rate of twelve pieces per minute. Each sensor synchronously collects data covering the entire stamping cycle, including upper die descent, stamping, upper die return, and part feeding. Among them, the force sensor records the load fluctuation data of the entire stamping process in real time, the laser displacement sensor continuously collects the die positioning deviation data, and the six triaxial acceleration sensors acquire the vibration signals of each installation part of the equipment. Through the signal synchronization mechanism of the acquisition card, the data of all sensors achieves precise and consistent timing. The completeness and validity of the collected raw data are verified. After confirming that there is no missing data or abnormal interference data, all data is transmitted and stored in the factory's industrial database to retain complete data for subsequent analysis.

[0022] The stamping monitoring data of the inner door panel of the stamping equipment was retrieved from the industrial database. Preprocessing was performed on the data, including noise reduction to filter out invalid signals from workshop environmental vibrations and electromagnetic interference, outliers were removed, and data standardization was applied to unify the dimensions and ranges of various data types. After preprocessing, load features were extracted from the load fluctuation data, positioning features from the mold positioning deviation data, and vibration features from the vibration signals of various parts of the equipment. Then, the three types of features were fused using a multi-sensor feature fusion formula, which is as follows: ,in, This is a core sensitive feature for the attenuation of equipment accuracy. For load characteristics, For location features, Vibration characteristics, , , The weighting coefficients for the features are determined through historical monitoring data; ultimately, the core sensitive features that reflect the accuracy degradation of the equipment status are obtained.

[0023] like Figure 2 As shown, based on the core sensitive feature of equipment accuracy attenuation and combined with historical monitoring data of the stamping equipment for producing car door inner panels, the predicted value of the car door inner panel size deviation for each production cycle is calculated using the stamping part size deviation prediction formula. The stamping part size deviation prediction formula is as follows: ,in, For the first Predicted dimensional deviation of stamped parts at time [time]. For the first The core sensitive feature of equipment accuracy degradation at any given time, the first The timeframe refers to the real-time moment at which the accuracy prediction is performed. For the first The core sensitive feature of equipment accuracy degradation at any given time, the first This moment is a historical moment. For the first The time decay coefficient at any given moment is determined using historical monitoring data. For the first The time decay factor at time, The sliding window length is determined using historical monitoring data; and through... The deviation growth rate is calculated synchronously; then, the dynamic warning threshold at the corresponding time is calculated using the dynamic threshold formula for fault warning. The dynamic threshold formula for fault warning is: ,in, For the first Dynamic fault warning threshold at any time, This serves as the basic threshold for the dimensional deviation of stamped parts. For the first The core sensitive feature of equipment accuracy degradation over time. The historical average of the core sensitive characteristic of equipment accuracy degradation during normal operation is determined through historical monitoring data. The standard deviation of the core sensitive characteristic of equipment accuracy degradation during normal operation is determined through historical monitoring data. The threshold adjustment coefficient is determined using historical monitoring data; then the predicted dimensional deviation value is... With dynamic early warning threshold When a comparison is performed, At that time, a fault warning is triggered, and based on continuous The trend of accuracy change in sequence prediction over future time.

[0024] Based on fault warning information, the audible and visual alarms in the workshop issue warning signals to remind on-site maintenance personnel to pay attention to the equipment status in a timely manner. At the same time, the central monitoring display screen in the workshop displays the load fluctuation curve, mold positioning deviation curve, component vibration curve, as well as the predicted value curve and accuracy change trend graph of the door inner panel size deviation of the stamping equipment in real time. This allows maintenance personnel to intuitively grasp the equipment operating status. Based on the original maintenance manual of this model of stamping equipment and combined with the factory's subsequent door inner panel production plan, targeted maintenance suggestions are generated. It is recommended that maintenance personnel perform precise calibration of the mold positioning reference surface during the production gap between the day shift and the night shift of the following day, and at the same time check the tightness of the slider connection parts.

[0025] In summary, for the production scenario of automotive door inner panels using single-point stamping equipment in automobile manufacturing plants, by precisely deploying force sensors, laser displacement sensors, and triaxial acceleration sensors at key parts of the stamping equipment, and configuring a compatible industrial Ethernet data acquisition card to unify signal transmission delay, synchronous acquisition, verification, and storage of multi-dimensional data throughout the stamping cycle are achieved. After data preprocessing and extraction of three types of features, the core sensitive features of equipment accuracy attenuation are obtained through multi-sensor feature fusion formulas. Then, based on the stamping part size deviation prediction formula and the fault early warning dynamic threshold formula, deviation prediction and early warning judgment are completed. Finally, through audible and visual alarms, data visualization, and customized maintenance suggestions, the equipment stability and product accuracy of automotive door inner panel stamping production are ensured.

[0026] Example 2: In the real-time monitoring of the refrigerator outer shell side panel stamping equipment, a comprehensive survey of the equipment's structure and operating trajectory was first conducted to determine the sensor installation scheme. Force sensors were installed at the connection between the equipment's slider and the upper die base to capture load changes during the stamping process. Laser displacement sensors were symmetrically installed in fixed areas on both sides of the equipment's worktable. The sensor positions and angles were repeatedly adjusted to ensure that the laser beam was precisely aligned with the positioning reference surface of the refrigerator outer shell side panel stamping die, ensuring the accuracy of the die positioning data acquisition. Six triaxial accelerometers were installed in the middle of the equipment body, at the lower end of the slider, and at the four corners of the worktable. All installation points avoided motion interference areas such as the equipment's feeding conveyor belt and ejector mechanism to prevent damage to the sensors from collisions with mechanical parts. Subsequently, a six-channel industrial Ethernet data acquisition card was configured, and a stable connection was established between the acquisition card and the adapter communication interfaces of each sensor. During the debugging phase, the signal transmission delay of each sensor was uniformly adjusted to the same interval to ensure that the acquired data had reliable time correlation.

[0027] After the stamping equipment completed its pre-shift no-load test run and all operating parameters reached production standards and entered a stable production state, the sensor and data acquisition card's acquisition program was started. At this time, the production line produced refrigerator outer shell side panels at a rate of fifteen pieces per minute. Each sensor synchronously collected data covering the entire stamping cycle, including upper die descent, stamping, upper die retraction, and sheet material transfer. Among them, the force sensor was responsible for collecting load fluctuation data throughout the stamping process, the laser displacement sensor continuously acquired mold positioning deviation data, and six triaxial acceleration sensors collected vibration signals from various installation parts of the equipment. Through the signal synchronization mechanism of the acquisition card, the data from the three types of sensors were synchronized in time. After a comprehensive verification of the collected data to confirm that there was no missing or erroneous data, all data was stored in the enterprise's industrial database.

[0028] The stamping monitoring data of the refrigerator outer shell side panel of the stamping equipment was retrieved from the industrial database. Data preprocessing was first performed to filter out invalid signals caused by electromagnetic interference and mechanical resonance from the workshop environment. Outliers were then removed, and data standardization was completed to improve overall data quality. After preprocessing, load features were extracted from load fluctuation data, positioning features from mold positioning deviation data, and vibration features from vibration signals from various parts of the equipment. Subsequently, the three types of features were fused using a multi-sensor feature fusion formula, which is as follows: Ultimately, the core sensitive characteristics of the equipment's accuracy attenuation are obtained.

[0029] Based on the core sensitive characteristics of equipment accuracy attenuation, and combined with historical monitoring data of the refrigerator outer shell side panels produced by this equipment, the real-time predicted value of the refrigerator outer shell side panels for each production cycle is calculated using the stamping part size deviation prediction formula. The stamping part size deviation prediction formula is as follows: ; and through The deviation growth rate is obtained synchronously; then, the dynamic warning threshold at the corresponding time is calculated using the fault warning dynamic threshold formula, which is: Then, the predicted dimensional deviation value With dynamic early warning threshold When a comparison is performed, At that time, a fault warning is triggered, and based on continuous The trend of accuracy change in sequence prediction over future time.

[0030] like Figure 3As shown, based on the fault warning information, the audible and visual alarms in the workshop immediately issue a warning, reminding maintenance personnel to handle the situation. On the intelligent monitoring display screen in the workshop, the monitoring curves of load changes, mold positioning, and component vibration of the stamping equipment are simultaneously switched and displayed, as well as the predicted curve of the outer shell side panel size deviation and the accuracy change trend graph. This allows maintenance personnel to quickly determine the direction of equipment abnormality. In conjunction with the maintenance manual of the stamping equipment and the factory's subsequent production plan for refrigerator outer shell side panels, maintenance suggestions are generated. It is recommended that maintenance personnel conduct a comprehensive cleaning and positioning calibration of the mold during the equipment downtime after get off work, and at the same time carry out flaw detection inspection on the connection parts of the equipment body.

[0031] In summary, for the production scenario of refrigerator outer shell side panels in the stamping equipment of home appliance enterprises, the deployment of various types of sensors and communication debugging of the acquisition cards were completed first to ensure the time correlation of data acquisition. Then, multi-dimensional data of the entire stamping cycle were synchronously collected, verified and stored. After preprocessing to extract three types of features—load, positioning and vibration—and fusing them to obtain core sensitive features, the dimensional deviation prediction formula of stamping parts and the dynamic threshold formula for fault early warning were used to realize the prediction of the dimensional deviation of the outer shell side panels and the dynamic threshold comparison. Combined with audible and visual alarms, visualization of monitoring curves and maintenance suggestions that fit the production plan, the risk of equipment accuracy decay was effectively avoided, and the production qualification rate of refrigerator outer shell side panels and equipment operation and maintenance efficiency were improved.

[0032] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for real-time monitoring of the operating status of stamping equipment based on sensor fusion, characterized in that, The specific steps of this method are as follows: Sensor deployment: Force sensors, laser displacement sensors, and triaxial acceleration sensors are deployed in key parts of the stamping equipment, and an industrial Ethernet data acquisition card is configured to establish communication with each sensor and control the signal transmission delay of each sensor. Data Synchronous Acquisition: After the stamping equipment is running stably, multi-dimensional data covering the entire stamping cycle is collected through various sensors. The data acquisition card signal synchronization mechanism is used to ensure that the data timing is consistent. After the collected data is verified, it is stored in the industrial database. Data preprocessing and feature fusion: Retrieve stored data from the industrial database and preprocess it, then extract load features, positioning features and vibration features, and perform fusion calculation on the three types of features through multi-sensor feature fusion formula to obtain the core sensitive features of equipment accuracy attenuation; Prediction and fault warning: Based on the core sensitive features of equipment accuracy decay and combined with historical monitoring data, the predicted value of size deviation is calculated by the formula for predicting the size deviation of stamped parts, and the dynamic warning threshold is calculated by the formula for dynamic warning threshold. The predicted value of size deviation is compared with the dynamic warning threshold to determine whether a fault warning is triggered and to predict the future trend of accuracy change. Results presentation: Based on fault warning information, predicted dimensional deviation values ​​and deviation growth rates, the system uses an audible and visual alarm to respond to warning signals, displays monitoring curves and accuracy change trend graphs on a screen, and generates maintenance suggestions in conjunction with the stamping equipment maintenance manual and production plan.

2. The method for real-time monitoring of the operating status of stamping equipment based on sensor fusion according to claim 1, characterized in that, In the sensor deployment, force sensors are installed at the connection between the slider and the upper die base of the stamping equipment; laser displacement sensors are installed on both sides of the worktable of the stamping equipment with the laser beam aligned with the die positioning reference surface; and triaxial acceleration sensors are installed in the middle of the machine body, the lower end of the slider, and the four corners of the worktable of the stamping equipment. The installation positions of each sensor avoid the interference area of ​​the equipment movement. An industrial Ethernet data acquisition card that supports multi-channel parallel reception is configured to establish a connection with the communication interface matched with each sensor, control the signal transmission delay of each sensor within a uniform range, and determine the temporal correlation of the acquired data.

3. The method for real-time monitoring of the operating status of stamping equipment based on sensor fusion according to claim 1, characterized in that, During the synchronous data acquisition, the multi-dimensional data includes load fluctuation data, mold positioning deviation data, and vibration signals of various parts of the equipment. Among them, the force sensor collects load fluctuation data, the laser displacement sensor collects mold positioning deviation data, and the triaxial acceleration sensor collects vibration signals of various parts of the equipment.

4. The method for real-time monitoring of the operating status of stamping equipment based on sensor fusion according to claim 1, characterized in that, In the data preprocessing and feature fusion process, the preprocessing operations include noise reduction, outlier removal, and data standardization. Load features are extracted from the preprocessed load fluctuation data, mold positioning deviation data, and vibration signals from various parts of the equipment. Location features and vibration characteristics Then, the three types of features are fused and calculated using a multi-sensor feature fusion formula to obtain the core sensitive features of equipment accuracy attenuation. .

5. The method for real-time monitoring of the operating status of stamping equipment based on sensor fusion according to claim 4, characterized in that, In the data preprocessing and feature fusion, the multi-sensor feature fusion formula is as follows: ,in, This is a core sensitive feature for the attenuation of equipment accuracy. For load characteristics, For location features, Vibration characteristics, , , The weighting coefficients for the features are determined using historical monitoring data.

6. The method for real-time monitoring of the operating status of stamping equipment based on sensor fusion according to claim 1, characterized in that, In the prediction and fault warning, the formula for predicting the dimensional deviation of the stamped part is: ,in, For the first Predicted dimensional deviation of stamped parts at time [time]. For the first The core sensitive feature of equipment accuracy degradation at any given time, the first The timeframe refers to the real-time moment at which the accuracy prediction is performed. For the first The core sensitive feature of equipment accuracy degradation at any given time, the first This is a historical moment. For the first The time decay coefficient at any given moment is determined using historical monitoring data. For the first The time decay factor at time, The length of the sliding window is determined using historical monitoring data; and through Obtain the deviation growth rate.

7. The method for real-time monitoring of the operating status of stamping equipment based on sensor fusion according to claim 1, characterized in that, In the prediction and fault early warning, the dynamic threshold formula for fault early warning is: ,in, For the first Dynamic fault warning threshold at any time, This serves as the basic threshold for the dimensional deviation of stamped parts. For the first The core sensitive feature of equipment accuracy degradation over time. The historical average of the core sensitive characteristic of equipment accuracy degradation during normal operation is determined through historical monitoring data. The standard deviation of the core sensitive characteristic of equipment accuracy degradation during normal operation is determined through historical monitoring data. The threshold adjustment coefficient is determined using historical monitoring data.

8. The method for real-time monitoring of the operating status of stamping equipment based on sensor fusion according to claim 7, characterized in that, In the aforementioned prediction and fault warning, and In contrast, when At that time, a fault warning is triggered, and based on continuous The trend of sequence prediction accuracy changes over future time.

9. A real-time monitoring system for the operating status of stamping equipment based on sensor fusion, the system being applicable to the real-time monitoring method for the operating status of stamping equipment based on sensor fusion as described in any one of claims 1-8, characterized in that, The system includes: Sensor deployment module: Deploy force sensors, laser displacement sensors, and triaxial acceleration sensors in key parts of the stamping equipment, and configure an industrial Ethernet data acquisition card to establish communication with each sensor and control the signal transmission delay of each sensor; Data synchronization acquisition module: After the stamping equipment is running stably, it collects multi-dimensional data covering the entire stamping cycle through various sensors, uses the signal synchronization mechanism of the acquisition card to ensure data consistency, and stores the collected data in the industrial database after verification. Data preprocessing and feature fusion module: retrieves stored data from the industrial database and preprocesses it, then extracts load features, positioning features and vibration features, and performs fusion calculation on the three types of features through multi-sensor feature fusion formula to obtain the core sensitive features of equipment accuracy attenuation; Prediction and Fault Warning Module: Based on the core sensitive features of equipment accuracy decay and combined with historical monitoring data, the module calculates the predicted value of size deviation using the formula for predicting the size deviation of stamped parts, calculates the dynamic warning threshold using the formula for dynamic warning threshold of fault warning, compares the predicted value of size deviation with the dynamic warning threshold to determine whether a fault warning is triggered and predicts the future trend of accuracy change. Results display module: Based on fault warning information, predicted dimensional deviation values ​​and deviation growth rates, the module uses an audible and visual alarm to respond to warning signals, displays monitoring curves and accuracy change trend graphs on a screen, and generates maintenance suggestions in conjunction with the stamping equipment maintenance manual and production plan.

Images