Intelligent workshop digital monitoring system for consumer electronics production

Abstract

The invention discloses an intelligent workshop digital monitoring system for consumer electronics production, which relates to the technical field of intelligent manufacturing and comprises a data acquisition module, an image recognition module, an environment monitoring module, an intelligent control terminal and an industrial communication interface module. According to the invention, workshop environment parameters, machine operation parameters and personnel operation data are collected in real time through a multi-sensor array deployed at each station and key process node of consumer electronics production, and the data are analyzed in real time in combination with an edge computing technology; the real-time monitoring and data analysis capability enables abnormal conditions, such as equipment faults and process deviations, in the production process to be quickly identified and processed, so that the downtime is reduced, the equipment utilization rate is improved, meanwhile, the linkage control unit of the intelligent control terminal can establish communication connection with an external MES system and an ERP system, and the production efficiency is improved. Data interaction and cooperative control are realized, the production process is further optimized, and the production efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to a digital monitoring system for an intelligent workshop for the production of consumer electronics. Background Art

[0002] With the rapid development of the consumer electronics industry, consumers' demand for electronic products is increasing, and they are also putting higher requirements on product quality, performance and delivery speed. In order to meet market demand, consumer electronics manufacturers are constantly seeking ways to improve production efficiency, reduce costs and ensure product quality.

[0003] Traditional consumer electronics production workshop monitoring systems often rely on manual inspections and single-parameter monitoring methods, which have many limitations. First, manual inspections are not only inefficient, but also difficult to achieve real-time and comprehensive monitoring of the production process, which can easily lead to untimely detection of abnormal situations, increased downtime and production costs. Second, the single-parameter monitoring method cannot fully reflect the complex state of the production workshop. When multiple parameters are abnormal at the same time, traditional systems find it difficult to accurately determine the root cause of the abnormality, affecting the efficiency and accuracy of fault handling. In addition, traditional systems lack effective integration with external MES systems and ERP systems, resulting in serious data silos and the inability to achieve optimization and collaborative control of production processes. Finally, when responding to dynamic factors such as environmental changes and equipment aging, the early warning mechanism of traditional monitoring systems is often not flexible enough, making it difficult to detect and solve potential problems in advance, affecting production safety and product quality.

[0004] In view of the shortcomings of traditional consumer electronics production workshop monitoring systems, the present invention proposes a digital monitoring system for intelligent consumer electronics production workshops, which is particularly important. Summary of the Invention

[0005] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a digital monitoring system for intelligent workshops in the production of consumer electronics. It can realize comprehensive real-time monitoring and data analysis of the production workshop by integrating advanced components such as multi-sensor arrays, image recognition modules, environmental monitoring modules and intelligent control terminals. The system can not only quickly identify and handle production anomalies such as equipment failures and process deviations, reduce downtime and improve equipment utilization, but also realize data interaction and collaborative control with external systems through linkage control units, and optimize production processes. At the same time, the system uses deep learning algorithms and dynamic early warning mechanisms to enhance production safety and product quality assurance capabilities.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a digital monitoring system for an intelligent workshop for consumer electronics production, the system comprising the following components: a data acquisition module, an image recognition module, an environmental monitoring module, an intelligent control terminal, and an industrial communication interface module; The data acquisition module is composed of a multi-sensor array deployed at each workstation and key process node in the consumer electronics production process, and is used to collect workshop environmental parameters, machine operating parameters, and personnel operation data in real time. The environmental parameters include temperature, humidity, and particulate matter concentration; the machine operating parameters include equipment speed, vibration frequency, current and voltage; and the personnel operation data includes operating actions and operation duration. The image recognition module is connected to the data acquisition module and includes an image acquisition device and an image analysis unit. The image acquisition device is used to obtain images of workstation operations and equipment appearance. The image analysis unit processes the images using a deep learning algorithm to identify human operation violations and equipment appearance anomalies. The environmental monitoring module is connected to the data acquisition module and the intelligent control terminal respectively, and is used to receive and analyze the environmental parameters collected by the data acquisition module. When the environmental parameters exceed the preset threshold, an environmental abnormality signal is generated and sent to the intelligent control terminal; The intelligent control terminal includes a data processing unit, an early warning unit, and a linkage control unit. The data processing unit receives data transmitted by the data acquisition module, image recognition module, and environmental monitoring module, and performs real-time analysis of the data in combination with edge computing technology. The early warning unit generates corresponding early warning prompts for equipment failures, process deviations, operational violations, and environmental anomalies based on the data processing results. The linkage control unit establishes a communication connection with the external MES system and ERP system through the industrial communication interface module to achieve data interaction and collaborative control. The industrial communication interface module uses Ethernet, RS485 or industrial wireless communication protocols to connect the intelligent control terminal to the external MES system, ERP system and workshop production equipment, and supports two-way data transmission.

[0007] Furthermore, the multi-sensor array in the data acquisition module adopts an adaptive sampling frequency adjustment algorithm to dynamically adjust the sampling interval according to the workshop production rhythm and data fluctuations to avoid data redundancy or insufficient sampling. The calculation formula is as follows: ,in is the basic sampling interval of each sensor, temperature and humidity sensor , vibration sensor , current sensor , particulate matter sensor ; is the standard deviation of the current sensor data, reflecting the degree of data fluctuation, and is calculated through a sliding window; The maximum historical standard deviation of the sensor in the consumer electronics production scenario is determined by the sensor data statistics of the same type of production line in the past year (such as temperature and humidity sensor , vibration sensor ); It is the adjustment coefficient, which is set according to the criticality of the production line: core process (SMT patch) , auxiliary processes (material handling) ; When the data fluctuation is small ( ), the sampling interval can be extended to , reduce the amount of data transmission; when the data fluctuation is large ( ), the sampling interval is shortened to , ensuring that abnormal data is not missed; through actual application, this algorithm can reduce the sensor data transmission volume by 30%-40%, while ensuring that the abnormal data capture rate is ≥98%.

[0008] Furthermore, a data transmission reliability guarantee algorithm is used between the data acquisition module and the intelligent control terminal to ensure that data is not lost or tampered through a redundancy check and retransmission mechanism. The calculation formula is as follows: ,in To improve the data retransmission success rate, the selective repeat ARQ protocol is used. When the receiving end detects data loss or error, it only requests the retransmission of the abnormal data packet instead of the entire data frame. After testing, in the scenario where the workshop network packet loss rate is ≤5%, ; In order to improve the data verification success rate, a CRC32+SHA256 double verification mechanism is adopted. CRC32 is used to quickly detect data transmission errors, and SHA256 is used to verify data integrity and prevent data tampering. The verification success rate ; For the data timeout retransmission compensation rate, the timeout threshold T_timeout is set to 200ms (optimized according to the workshop communication distance, T_timeout is 150ms for workshops within 100m, and T_timeout is 200ms for workshops within 500m). When data transmission times out, retransmission compensation is automatically triggered. The t value is dynamically adjusted with the number of timeouts, with t=0.8 for the first timeout and t=0.95 for the second timeout, to ensure that the timed-out data can be retransmitted in time. This algorithm makes the data transmission reliability R≥0.999, which is 0.02 orders of magnitude higher than the traditional single verification mechanism. It can effectively solve the data congestion and loss problem caused by simultaneous transmission of multiple devices in the consumer electronics workshop, and ensure the integrity of the collected data.

[0009] Furthermore, the deep learning algorithm of the image analysis unit in the image recognition module is a multi-feature fusion attention convolutional neural network, which extracts and dynamically weights the device appearance features, human operation features, and environmental interference features in a hierarchical manner. The calculation formula is as follows: ,in For the device appearance abnormality feature function, input The 12-dimensional feature vectors for the edge contour, color deviation, and part position offset of the device appearance image are extracted through three convolutional layers (with kernel sizes of 3×3, 5×5, and 7×7, respectively) and two pooling layers (max pooling with a stride of 2). The output range is [0, 1]. The closer the value is to 1, the higher the probability of device appearance abnormality. For the personnel operation violation feature function, input The 18-dimensional feature vector of the operator's work image, including limb joint angles, tool usage positions, and material placement coordinates, is learned through a bidirectional LSTM network (with 64 hidden layer neurons). The output range is [0, 1], and the closer the value is to 1, the higher the probability of an operation violation. is the environmental interference correction function, input The five-dimensional interference parameters are workshop light intensity, dust obstruction, and equipment reflectivity. The outputs of F1 and F2 are corrected through a fully connected layer (with a Sigmoid activation function) in the range of [0.8, 1.2] to offset the interference of environmental factors on the recognition results. 、 、 is the dynamic weight coefficient, satisfying , whose value is obtained by optimizing the historical data set of consumer electronics production scenarios (including 100,000 sets of normal / abnormal samples) using the gradient descent method, where Initial value 0.45, Initial value 0.4, The initial value is 0.15. After 50 rounds of iterative training, it finally converges to 、 、 , ensuring that when equipment anomalies and operational violations occur at the same time, key anomalies that affect production safety are identified first; the algorithm achieves an anomaly recognition accuracy rate of ≥ 97.5% through the above formula, which is 4.2 percentage points higher than the traditional CNN algorithm, and its adaptability to common lighting changes and dust interference scenarios in consumer electronics workshops is improved by 30%.

[0010] Furthermore, the environmental monitoring module adopts a multi-parameter coordinated adjustment algorithm. When the environmental parameters exceed the threshold, the optimal adjustment scheme is automatically calculated. The calculation formula is as follows: ,in 、 、 are the deviation values ​​of temperature and humidity, particulate matter concentration and workshop heat load, , , ; 、 、 Adjust the sensitivity coefficient of each parameter through experiments: Temperature and humidity adjustment sensitivity (Each adjustment of 100m³ / h ventilation volume results in a temperature change of 0.8°C), particle concentration adjustment sensitivity (For every 50m³ / h ventilation adjustment, the particle concentration decreases by 1.2mg / m³), heat load adjustment sensitivity (For every 1kW increase in cooling capacity, the heat load decreases by 0.6kW); Adjust the efficiency coefficient for the equipment, new ventilation equipment , equipment used for more than 1 year ; Energy consumption coefficient, adjusted according to the workshop electricity price period: Peak time (8:00-22:00) , Valley Time (22:00-8:00) ; This algorithm can shorten the time it takes for environmental parameters to return to within the threshold to 15 minutes, which is three times more efficient than traditional manual adjustment and reduces energy consumption by 18%.

[0011] Furthermore, the data processing unit of the intelligent control terminal adopts a multi-dimensional data coupling analysis algorithm to integrate environmental parameters, equipment parameters and personnel data to accurately determine the root cause of the abnormality. The calculation formula is as follows: ,in is the data deviation of each dimension, environmental dimension =|actual temperature and humidity / particle concentration-preset threshold|, device dimension =|actual vibration frequency / current-rated value|, personnel dimension =|actual operation time / action standardization-standard value|, all After normalization, the range is [0, 1]; The historical correlation coefficients between each dimension and production anomalies are obtained through the statistics of failure data of consumer electronics workshops in the past three years: the correlation coefficients between environmental anomalies and product defects , the correlation coefficient between equipment failure and product defects , the correlation coefficient between personnel violations and product defects ; The weight of each dimension is dynamically adjusted according to the current production line type: Smartphone production line (with many precision components) is set 、 、 ; Tablet computer production line (many assembly processes) 、 、 When the calculated R≥0.6, the dimension is determined to be the main source of the anomaly, and the system automatically triggers the corresponding processing flow. After testing, the algorithm's accuracy in judging the root cause of the anomaly is ≥92%, reducing the misjudgment rate by 45% compared with the traditional single-dimensional judgment method.

[0012] Furthermore, the multi-level warning threshold of the warning unit adopts a dynamic update algorithm, combining the changes in product qualification rate and the degree of equipment aging to adjust the warning trigger conditions in real time. The calculation formula is as follows: ,in is the initial warning threshold, the initial threshold of equipment vibration Th oin =0.5g, initial threshold value of temperature and humidity Th oin =26℃; is the ratio of the current product qualification rate to the historical average qualification rate, The qualified rate of products in the past 7 days. The average qualified rate of products in the past year. When P / P0 < 0.95, it indicates that the current production stability has declined and the warning threshold needs to be lowered to provide early warning. The ratio of equipment usage time to design life. is the actual running time of the device, For the design life of the equipment, When the device enters the aging period, the warning threshold needs to be lowered to cope with potential failures; 、 is the adjustment coefficient, m=-0.15 (lower the threshold when the pass rate decreases), n=-0.1 (lower the threshold when the equipment ages); this algorithm can extend the equipment failure warning advance time to 20-25 minutes, reducing failure losses by 38% compared with fixed threshold warning.

[0013] Furthermore, the edge computing module of the intelligent control terminal adopts a hierarchical data processing algorithm to divide the data into real-time emergency data, quasi-real-time analysis data, and non-real-time storage data, and allocates different computing resources to each of them. The formula is as follows: ,in is the data urgency coefficient, s1=1.0 for real-time data (such as equipment current exceeding the limit), s2=0.5 for quasi-real-time data (such as hourly pass rate analysis), and s3=0.2 for non-real-time data (such as monthly production report); Data utilization rate: real-time data is used for immediate warning, u1=0.95; quasi-real-time data is used for process optimization, u2=0.7; non-real-time data is used for trend analysis, u3=0.4; The data processing priority is set according to the production line requirements: smartphone production line (high real-time requirement) p1=0.6, p2=0.3, p3=0.1; smart wearable device production line (high analysis requirement) p1=0.4, p2=0.45, p3=0.15.

[0014] Compared with the existing technology, this intelligent workshop digital monitoring system for consumer electronics production has the following beneficial effects: 1. This system uses a multi-sensor array deployed at various workstations and key process nodes in consumer electronics production to collect workshop environmental parameters, machine operating parameters, and personnel work data in real time, and combines it with edge computing technology to analyze the data in real time. This real-time monitoring and data analysis capability enables abnormal situations in the production process, such as equipment failures and process deviations, to be quickly identified and handled, thereby reducing downtime and improving equipment utilization. At the same time, the linkage control unit of the intelligent control terminal can establish communication connections with external MES systems and ERP systems to achieve data interaction and collaborative control, further optimizing the production process and improving production efficiency.

[0015] 2. The image recognition module in this system uses deep learning algorithms to process workstation operation images and equipment appearance images. It can accurately identify personnel operation violations and equipment appearance abnormalities, thereby timely preventing potential safety accidents. In addition, the environmental monitoring module can monitor and analyze workshop environmental parameters in real time. When the environmental parameters exceed the preset threshold, it automatically generates environmental abnormality signals and takes corresponding adjustment measures to ensure that the production environment meets process requirements and avoids product quality problems caused by environmental factors. The multi-level warning threshold of the early warning unit adopts a dynamic update algorithm, and adjusts the warning trigger conditions in real time based on the changes in product qualification rate and the degree of equipment aging, so that problems in the production process can be discovered and resolved in advance, further enhancing production safety and product quality.

[0016] Other advantages, objects and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be learned from the practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0018] Figure 1 This is a flowchart of the digital monitoring system operation of a smart workshop for consumer electronics production; Figure 2 This is an overall process flow chart of a digital monitoring system for an intelligent workshop in the production of consumer electronics. DETAILED DESCRIPTION

[0019] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0020] Example 1 After the digital monitoring system of the smart workshop is started, each component module automatically completes initialization. This step lays a stable foundation for the subsequent entire monitoring process and ensures that the system can quickly enter normal working state. In the data acquisition module, the multi-sensor array deployed in key workstations such as the patch workstation, welding workstation, assembly workstation of the smartphone assembly and key process nodes such as material transmission and quality inspection enters the working state and is ready to collect various types of data in real time. Its role is to comprehensively cover the key links of production and provide a full range of data sources for subsequent data analysis; the image acquisition device of the image recognition module (such as a high-definition camera) is respectively aimed at the working area of ​​each workstation and the appearance of key production equipment, and the parameter setting is completed to ensure that the image can be clearly obtained. The role of this operation is to ensure that high-quality images can be obtained in subsequent image recognition, thereby improving Identification accuracy; the environmental monitoring module completes the connection test with the data acquisition module and the intelligent control terminal to confirm that the data transmission channel is unobstructed. The purpose of this connection test is to avoid the subsequent interruption of environmental parameter transmission and ensure that environmental anomalies can be discovered in time; each unit of the intelligent control terminal completes self-inspection, and the edge computing related functions are ready. The purpose of self-inspection is to eliminate the hidden dangers of terminal failure itself, and the readiness of edge computing functions provides a guarantee for subsequent real-time processing of data; the industrial communication interface module establishes a stable communication connection with the MES system (manufacturing execution system) and ERP system (enterprise resource planning system) outside the workshop through the Ethernet protocol to ensure normal subsequent data interaction. The purpose of this connection is to realize data sharing between the system and the external management system, and provide data support for production plan adjustment, cost accounting, etc.

[0021] The multi-sensor array of the data acquisition module follows the adaptive sampling frequency adjustment algorithm, and the calculation formula is as follows: ,in is the basic sampling interval of each sensor; is the standard deviation of the current sensor data; is the maximum historical standard deviation of the sensor in the consumer electronics production scenario; To adjust the coefficient, the sampling interval is dynamically adjusted based on the production rhythm of the smartphone assembly workshop (such as the component placement speed of the SMT station and the welding cycle of the welding station) and real-time data fluctuations. In real time, workshop environmental parameters (such as workshop temperature, humidity, and particulate matter concentration in the air), machine operating parameters (such as the placement accuracy of SMT equipment, the welding temperature of welding equipment, and the operating speed of assembly equipment) and personnel operation data (such as the operating time of operators at each station and the number of times materials are taken) are collected. The purpose of using an adaptive sampling frequency adjustment algorithm is to avoid data redundancy and waste of storage resources, and to prevent the loss of key data due to insufficient sampling, thereby ensuring the validity and economy of the collected data. During the data transmission process to the intelligent control terminal, a data transmission reliability assurance algorithm is adopted. The calculation formula is as follows: ,in is the data retransmission success rate; is the data verification success rate; The algorithm is used to ensure the data timeout retransmission compensation rate. Through data retransmission, data verification and data timeout retransmission compensation mechanisms, it ensures that the collected data is accurately and completely transmitted to the intelligent control terminal to avoid data loss or damage. The role of this algorithm is to ensure the quality of data in the transmission link, provide an accurate data basis for subsequent data processing and analysis, and prevent misjudgment of production anomalies due to data errors.

[0022] The image acquisition device of the image recognition module continuously acquires images of the workstations (such as the operator's operating actions at the welding station and the material assembly process at the assembly station) and equipment appearance images (such as the nozzle appearance of the patch equipment and the welding head appearance of the welding equipment), and transmits these image data to the image analysis unit. The purpose of continuously acquiring images is to capture the changes in human operations and equipment status in the production process in real time, without missing any key moments. The image analysis unit uses a multi-feature fusion attention convolutional neural network algorithm. The calculation formula is as follows: ,in For the device appearance abnormality feature function, input 12-dimensional feature vectors for edge contour, color deviation, and part position offset of the device appearance image; For the personnel operation violation feature function, input 18-dimensional feature vectors of limb joint angles, tool usage positions, and material placement coordinates of the personnel operation image; is the environmental interference correction function, input These are the five-dimensional interference parameters of workshop light intensity, dust shielding, and equipment reflectivity; 、 、 For dynamic weight coefficient, the received image is processed, the algorithm will extract the device appearance characteristics (such as whether the device parts are damaged, whether the position is deviated), personnel operation characteristics (such as whether the operator uses the tool according to the specification, whether the body movement conforms to the standard operation process) and environmental interference characteristics (such as the influence of light change in the workshop on the image, the image blur caused by dust shielding), and dynamic weight distribution, which can effectively distinguish the production key characteristics and environmental interference factors, improve the identification accuracy of personnel operation violation and equipment appearance anomaly, avoid misidentification or missed identification caused by environmental interference, accurately identify personnel operation violation (such as the operator does not wear anti-static bracelet for welding operation according to the regulation) and equipment appearance anomaly (such as the nozzle of the patch device appears wear), and then transmit the identification result to the intelligent control terminal. The function of the identification result transmission is to enable the intelligent control terminal to timely master the violation and abnormal situation in the production process, and to provide basis for subsequent early warning and processing.

[0023] The environmental monitoring module receives the workshop environment parameters transmitted by the data acquisition module in real time, and analyzes these parameters. The functions of real-time receiving and analyzing are to discover environmental changes in time and prevent environmental abnormalities from affecting production quality and personnel safety. When the analysis finds that the environmental parameters of a certain area exceed the preset threshold (such as the temperature near the welding station is too high, and the humidity in the assembly area exceeds the appropriate range), the environmental monitoring module immediately generates an environmental abnormality signal and sends it to the intelligent control terminal. The function of generating and sending the abnormal signal is to enable the management personnel to know the environmental problem at the first time, so as to gain time for timely processing. At the same time, the environmental monitoring module adopts a multi-parameter coordinated regulation algorithm, and the calculation formula is as follows: , wherein , , are the deviation values of temperature, humidity, particulate matter concentration and workshop heat load respectively; , , are the parameter regulation sensitivity coefficients; is the device regulation efficiency coefficient; is the energy consumption coefficient. Combined with the deviation of the current environmental parameters, the regulation sensitivity of each parameter, the device regulation efficiency and energy consumption, etc., the optimal environmental regulation scheme is automatically calculated and transmitted to the intelligent control terminal, which provides a basis for subsequent environmental adjustment. The function of the algorithm is to solve the environmental abnormality problem while considering the device regulation efficiency and energy consumption cost, to realize the economy and effectiveness of environmental regulation, and to avoid resource waste caused by blind regulation.

[0024] After receiving various types of collected data from the data acquisition module, the recognition results of the image recognition module, and the environmental parameter analysis results and abnormal signals from the environmental monitoring module, the data processing unit of the intelligent control terminal uses a multi-dimensional data coupling analysis algorithm to integrate and analyze environmental parameters (such as temperature and humidity anomalies), equipment parameters (such as equipment operating speed and temperature anomalies), and personnel data (such as excessive operating hours and illegal operations by operators). This allows the user to accurately determine the root cause of production anomalies (for example, determining whether product quality problems are caused by abnormal equipment operating parameters or by illegal human operations). The calculation formula is as follows: ,in is the deviation of data in each dimension; is the historical correlation coefficient between each dimension and production anomaly; The algorithm is used to avoid the limitations of single-dimensional data analysis, improve the accuracy of abnormal root cause judgment through multi-dimensional data fusion, and provide the right direction for subsequent targeted processing.

[0025] If the data processing unit analyzes and finds equipment failure (such as out-of-control temperature of welding equipment), process deviation (such as patch process accuracy not meeting requirements), operation violation (such as operator failure to perform standardized operations), or environmental abnormalities, the early warning unit will dynamically update the algorithm based on the analysis results and the multi-level early warning threshold. The calculation formula is as follows: ,in is the initial warning threshold; It is the ratio of the current product qualification rate to the historical average qualification rate; The ratio of equipment usage time to design life. is the actual running time of the device, Design life for equipment; 、 In order to adjust the coefficient, corresponding early warning prompts are generated (such as issuing an alarm through the sound and light alarm device in the workshop, and popping up early warning information on the display interface of the intelligent control terminal), prompting management personnel to deal with it in time. The role of the multi-level early warning threshold dynamic update algorithm is to make the early warning trigger conditions adapt to the actual changes in production, avoid untimely or false warnings due to fixed thresholds, ensure the timeliness and accuracy of early warnings, and thus reduce production losses.

[0026] Based on early warning prompts and environmental adjustment plans, the linkage control unit of the intelligent control terminal sends control commands to relevant production equipment in the workshop through the industrial communication interface module (such as sending adjustment commands to temperature control equipment to reduce the temperature near welding stations and sending pause commands to equipment at risk of failure) to prevent the expansion of failures. The role of linkage control is to achieve rapid response and handling of production anomalies, reduce the scope and extent of anomalies' impact on the production process, and ensure stable production operations. At the same time, the linkage control unit also transmits various types of production process data (such as production progress data, quality inspection data, and equipment operating status data) to external MES and ERP systems through the industrial communication interface module, enabling production data sharing and interaction. The role of data sharing is to enable the MES system to better arrange production plans and the ERP system to more accurately perform management tasks such as cost accounting and material procurement, thereby improving overall production management efficiency. In addition, the edge computing module of the intelligent control terminal uses a hierarchical data processing algorithm to classify the data generated during the production process into real-time emergency data (such as equipment failure warning data), quasi-real-time analysis data (such as production process optimization analysis data), and non-real-time storage data (such as historical production data backup). Different computing resources are allocated to different types of data. The formula is as follows: ,in is the data urgency coefficient; is data utilization; For data processing priority, the role of this algorithm is to ensure that real-time emergency data is processed first, avoid delays in real-time data processing due to unreasonable allocation of computing resources, ensure that emergencies in the production process can be responded to in a timely manner, and at the same time reasonably utilize computing resources to improve the overall efficiency of data processing.

[0027] Example 2 In the consumer electronics tablet screen production workshop, the deployment and preparation of the smart workshop digital monitoring system have been completed. The multi-sensor array of the data acquisition module is deployed at key stations such as the screen glass cutting station, coating station, laminating station, and quality inspection station, as well as key process nodes such as the screen transmission line and storage warehouse. The sensor type is selected according to the needs of each station. For example, a displacement sensor is deployed at the cutting station to monitor the cutting position, and a concentration sensor is deployed at the coating station to monitor the concentration of the coating material. The high-definition image acquisition device of the image recognition module is installed near the quality inspection station and key components of each production equipment to ensure that the screen quality details and the appearance of the equipment can be clearly captured. The monitoring points of the environmental monitoring module are distributed in various areas of the workshop, focusing on monitoring parameters such as temperature, humidity, and cleanliness in areas with higher environmental requirements such as the coating station and laminating station. The intelligent control terminal is installed in the workshop control room to complete the connection configuration with each module. The industrial communication interface module uses the RS485 protocol to establish communication connections with the workshop's MES system, ERP system and various production equipment controllers, and completes communication testing to ensure stable data transmission.

[0028] After the system is officially put into use, the multi-sensor array of the data acquisition module dynamically adjusts the sampling interval based on an adaptive sampling frequency adjustment algorithm, according to the production rhythm of tablet screens, such as glass cutting speed, coating cycle, bonding efficiency, and real-time data fluctuations. It collects real-time workshop environmental parameters (such as temperature, humidity, and cleanliness of the coating station and temperature and humidity of the warehouse), machine operating parameters (such as cutting force and speed of cutting equipment, vacuum level and temperature of coating equipment, pressure and speed of bonding equipment), and personnel operation data (such as the inspection time and number of inspections by quality inspectors, and the number of times operators operate the equipment). During the data transmission process, a data transmission reliability assurance algorithm is adopted to ensure the accurate transmission of collected data to the intelligent control terminal through data retransmission, verification, and timeout retransmission compensation. At the same time, the image acquisition device of the image recognition module continuously obtains images of the tablet screen at the quality inspection station (such as whether the screen has scratches, bubbles, or color difference) and images of the appearance of each production equipment (such as the appearance of the blade of the cutting equipment and the appearance of the cavity of the coating equipment), and transmits the image data in real time to the image analysis unit.

[0029] After receiving the image data, the image analysis unit of the image recognition module uses a multi-feature fusion attention convolutional neural network algorithm to process the screen image and the device appearance image. The algorithm hierarchically extracts screen quality features (such as the length and width of scratches, the size and position of bubbles, and the degree of color difference), device appearance features (such as whether equipment parts are worn, corroded, or loose, and whether there are stains on the equipment surface that affect operation), and environmental interference features (such as the impact of changes in light intensity at the quality inspection station on screen image judgment, and the interference of workshop dust on the clarity of the device appearance image). It then performs dynamic weight allocation to accurately identify quality defects of the tablet screen (such as scratches and bubbles on the screen) and device appearance abnormalities (such as wear on the cutting equipment blade and corrosion in the coating equipment cavity), and transmits the recognition results, including information such as the type and location of quality defects and abnormal parts of the equipment, to the intelligent control terminal.

[0030] The environmental monitoring module receives real-time environmental parameters from the data acquisition module for each area of ​​the workshop and continuously analyzes them. If the cleanliness level of the coating station falls below a preset threshold, or the temperature and humidity at the laminating station exceed the appropriate range, the module immediately generates an environmental anomaly signal and uses a multi-parameter coordinated adjustment algorithm to comprehensively consider factors such as the deviation size, adjustment sensitivity, equipment adjustment efficiency, and energy consumption of each environmental parameter to calculate the optimal environmental adjustment plan. For example, it adjusts the operating power of the clean equipment to improve the cleanliness level of the coating station and adjusts the air-conditioning system parameters to optimize the temperature and humidity at the laminating station. The module then sends the environmental anomaly signal and adjustment plan to the intelligent control terminal. Simultaneously, the data acquisition module continuously collects the operating parameters of each production equipment and transmits them to the intelligent control terminal. The data processing unit combines these parameters and uses a multi-dimensional data coupling analysis algorithm to integrate and analyze equipment operating parameters (such as abnormal cutting force of cutting equipment and unstable vacuum level of coating equipment), environmental parameters (such as the impact of ambient temperature and humidity on equipment operation), and human operation data (such as whether the operator operates the equipment in a standardized manner) to determine whether the equipment is operating normally. If a potential equipment failure risk is found, the type of potential failure and the possible impact range are promptly identified.

[0031] The early warning unit of the intelligent control terminal generates corresponding early warning prompts based on the analysis results of the equipment operation status by the data processing unit, the quality defects and environmental abnormality signals found by the image recognition module, and the dynamic update algorithm of the multi-level early warning threshold (this algorithm refers to the current changes in the qualified rate of tablet computer screen products and the proportion of the production equipment usage time to the design life, and adjusts the early warning trigger conditions in real time). For example, when a large number of screen quality defects are identified and the product qualified rate decreases, a quality warning is issued; when a potential failure risk is found in the equipment, and the equipment usage time is close to the design life and the degree of aging is high, an equipment maintenance warning is issued; when abnormal environmental parameters may affect production quality, an environmental warning is issued. The warning information is conveyed to workshop managers and maintenance personnel through the display screen in the control room, SMS notifications, etc.

[0032] The linkage control unit of the intelligent control terminal sends control instructions to related equipment through the industrial communication interface module according to the early warning prompts and environmental adjustment plans. For example, it sends instructions to the quality inspection equipment to increase the detection frequency to ensure that more screens with quality defects are detected; sends maintenance request instructions to the equipment maintenance system to notify maintenance personnel to promptly repair equipment with failure risks; sends adjustment instructions to the environmental adjustment equipment to adjust the workshop environmental parameters according to the optimal adjustment plan. At the same time, the linkage control unit transmits the quality inspection data (such as the number of qualified screens, defect types and quantities), equipment operation status data (such as equipment operation parameters, fault maintenance records), environmental parameter adjustment data, etc. in the production process to the external MES system and ERP system through the industrial communication interface module. The MES system adjusts the production according to these data. Planning and optimizing production processes, such as adjusting the production task allocation of each workstation to avoid the expansion of quality problems; the ERP system conducts cost analysis based on data, such as calculating the cost losses caused by quality defects and arranging the funds and material procurement required for equipment maintenance. In addition, the edge computing module of the intelligent control terminal adopts a hierarchical data processing algorithm to classify and process production data, and prioritizes the allocation of computing resources for rapid processing of real-time emergency data such as equipment failure warning data and emergency quality non-conforming data to ensure timely response; for quasi-real-time analysis data such as production process optimization analysis and equipment operation efficiency analysis, computing resources are reasonably allocated while ensuring real-time performance; for non-real-time storage data such as historical quality data and equipment historical operation data, appropriate computing resources are allocated for sorting and storage to provide data support for subsequent production analysis and decision-making.

[0033] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A digital monitoring system for intelligent workshops in the production of consumer electronics, characterized by: The system includes the following components: data acquisition module, image recognition module, environmental monitoring module, intelligent control terminal and industrial communication interface module; The data acquisition module is composed of a multi-sensor array deployed at each workstation and key process node in the consumer electronics production process, and is used to collect workshop environmental parameters, machine operating parameters, and personnel operation data in real time; The image recognition module is connected to the data acquisition module and includes an image acquisition device and an image analysis unit. The image acquisition device is used to obtain images of workstation operations and equipment appearance. The image analysis unit processes the images using a deep learning algorithm to identify human operation violations and equipment appearance anomalies. The environmental monitoring module is connected to the data acquisition module and the intelligent control terminal respectively, and is used to receive and analyze the environmental parameters collected by the data acquisition module. When the environmental parameters exceed the preset threshold, an environmental abnormality signal is generated and sent to the intelligent control terminal; The intelligent control terminal includes a data processing unit, an early warning unit, and a linkage control unit. The data processing unit receives data transmitted by the data acquisition module, image recognition module, and environmental monitoring module, and performs real-time analysis of the data in combination with edge computing technology. The early warning unit generates corresponding early warning prompts for equipment failures, process deviations, operational violations, and environmental anomalies based on the data processing results. The linkage control unit establishes a communication connection with the external MES system and ERP system through the industrial communication interface module. The industrial communication interface module uses Ethernet, RS485 or industrial wireless communication protocols to connect the intelligent control terminal to the external MES system, ERP system and workshop production equipment, and supports two-way data transmission.

2. The intelligent workshop digital monitoring system for consumer electronics production according to claim 1 is characterized in that: The multi-sensor array in the data acquisition module adopts an adaptive sampling frequency adjustment algorithm to dynamically adjust the sampling interval according to the workshop production rhythm and data fluctuations to avoid data redundancy or insufficient sampling. The calculation formula is as follows: ,in is the basic sampling interval of each sensor; is the standard deviation of the current sensor data; is the maximum historical standard deviation of the sensor in the consumer electronics production scenario; is the adjustment coefficient.

3. The intelligent workshop digital monitoring system for consumer electronics production according to claim 1 is characterized in that: The data transmission reliability guarantee algorithm is adopted between the data acquisition module and the intelligent control terminal. The calculation formula is as follows: ,in is the data retransmission success rate; is the data verification success rate; The data timeout retransmission compensation rate.

4. The intelligent workshop digital monitoring system for consumer electronics production according to claim 1 is characterized in that: The deep learning algorithm of the image analysis unit in the image recognition module is a multi-feature fusion attention convolutional neural network. This algorithm extracts and dynamically weights the device appearance features, human operation features, and environmental interference features in a hierarchical manner. The calculation formula is as follows: ,in For the device appearance abnormality feature function, input 12-dimensional feature vectors for edge contour, color deviation, and part position offset of the device appearance image; For the personnel operation violation feature function, input 18-dimensional feature vectors of limb joint angles, tool usage positions, and material placement coordinates of the personnel operation image; is the environmental interference correction function, input These are the five-dimensional interference parameters of workshop light intensity, dust shielding, and equipment reflectivity; 、 、 is the dynamic weight coefficient.

5. The intelligent workshop digital monitoring system for consumer electronics production according to claim 1 is characterized in that: The environmental monitoring module uses a multi-parameter coordinated adjustment algorithm. When the environmental parameters exceed the threshold, it automatically calculates the optimal adjustment solution. The calculation formula is as follows: ,in 、 、 are the deviation values ​​of temperature and humidity, particulate matter concentration and workshop heat load respectively; 、 、 Adjust the sensitivity coefficients for each parameter; Adjust the efficiency coefficient for the equipment; is the energy consumption coefficient.

6. The intelligent workshop digital monitoring system for consumer electronics production according to claim 1 is characterized in that: The data processing unit of the intelligent control terminal uses a multi-dimensional data coupling analysis algorithm to integrate environmental parameters, equipment parameters and personnel data to accurately determine the root cause of the abnormality. The calculation formula is as follows: ,in is the deviation of data in each dimension; is the historical correlation coefficient between each dimension and production anomaly; is the weight of each dimension.

7. The digital monitoring system for intelligent workshops in consumer electronics production according to claim 1, characterized in that: The multi-level warning threshold of the early warning unit adopts a dynamic update algorithm, combining the changes in product qualification rate and the degree of equipment aging to adjust the early warning trigger conditions in real time. The calculation formula is as follows: ,in is the initial warning threshold; It is the ratio of the current product qualification rate to the historical average qualification rate; The ratio of equipment usage time to design life, is the actual running time of the device, Design life for equipment; 、 is the adjustment coefficient.

8. The intelligent workshop digital monitoring system for consumer electronics production according to claim 1 is characterized in that: The edge computing module of the intelligent control terminal adopts a hierarchical data processing algorithm to divide the data into real-time emergency data, quasi-real-time analysis data, and non-real-time storage data, and allocates different computing resources to each of them. The formula is as follows: ,in is the data urgency coefficient; is data utilization; Data processing priority.

Images