Method and system for gradient drying of butyronitrile gloves after impregnation

Abstract

The invention relates to a method and system for gradient drying of butyronitrile gloves after impregnation in the field of intelligent manufacturing, and the method comprises the steps: obtaining surface and internal moisture content data of an adhesive film through a sensor array, generating a real-time moisture distribution matrix, and determining a moisture gradient change trend; if the moisture gradient change trend exceeds a preset threshold value, calculating a local moisture evaporation rate according to the distribution matrix to obtain a dynamic temperature and humidity regulation and control parameter; obtaining an adhesive film surface image through a high-resolution imaging system, detecting micro-crack and white point defect areas, and determining defect distribution characteristics; adjusting power distribution of the composite heating system according to the lowest energy consumption operation parameters, and generating an efficient drying operation mode; updating data of the dynamic regulation and control parameters and the defect distribution characteristics are obtained through a feedback control algorithm, and final drying process parameters are determined; and generating a production control instruction from the final drying process parameters, and outputting the production control instruction to a drying equipment controller to finish uniform diffusion of moisture of the adhesive film and elimination of surface defects.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing, specifically to the field of production technology of nitrile gloves, and in particular to a method and system for gradient drying nitrile gloves after dipping. Background Art

[0002] Nitrile gloves are indispensable protective equipment in the medical, industrial and other fields, and their production process directly affects product quality and production efficiency. The drying process after dipping and molding is a key link in determining the performance of gloves. It is necessary to ensure that the moisture in the film evaporates evenly while avoiding surface defects and internal stress accumulation. Traditional drying methods mostly use constant temperature ovens, and the temperature is usually maintained at 70 to 90 degrees Celsius. This single mode causes rapid crusting on the surface of the film, hinders internal moisture migration, and easily causes microcracks and white spot defects. Although the existing segmented drying technology attempts to optimize moisture evaporation through different temperature zones, the temperature switching is discontinuous, there is a lack of precise control of humidity, and the energy consumption is high, making it difficult to meet the needs of efficient production.

[0003] The core challenges facing current drying processes focus on achieving uniform moisture diffusion, reducing energy consumption, and ensuring film surface quality. Constant-temperature drying, due to a mismatch between surface and internal moisture evaporation rates, can easily lead to microcracks and white spots. While staged drying offers some improvements, the discontinuous nature of temperature and humidity control can lead to localized overdrying or water accumulation on the film, impacting its appearance and mechanical properties. Furthermore, high energy consumption limits production line speed increases and cost optimization. These technical challenges stem from the lack of a dynamic, precise temperature and humidity control mechanism, as well as real-time monitoring and feedback of the film's moisture distribution.

[0004] Therefore, how to achieve uniform diffusion of moisture in the film, reduce energy consumption and eliminate surface defects through continuous temperature and humidity gradient control, combined with efficient heating methods and real-time monitoring technology, has become a key issue in the research of gradient drying process after nitrile gloves are impregnated. Summary of the Invention

[0005] The present invention provides a method for gradient drying of nitrile gloves after dipping, comprising the following steps:

[0006] The sensor array acquires the moisture content data on the film surface and inside, generates a real-time moisture distribution matrix, and determines the moisture gradient change trend;

[0007] If the moisture gradient change trend exceeds the preset threshold, the local moisture evaporation rate is calculated according to the distribution matrix to obtain the dynamic temperature and humidity control parameters;

[0008] A composite heating method combining infrared radiation and hot air circulation is used to adjust the heating intensity and wind speed according to dynamic control parameters to generate a continuous temperature and humidity gradient field;

[0009] Use a high-resolution imaging system to obtain film surface images, detect microcracks and white spot defect areas, and determine defect distribution characteristics;

[0010] If the defect distribution characteristics show that microcracks or white spots exceed the preset standards, the local temperature and humidity gradient field is adjusted according to the location of the defect area to optimize the water evaporation path;

[0011] Extract the real-time moisture diffusion coefficient from the film moisture distribution matrix, combine it with the defect distribution characteristics, calculate the energy consumption optimization model, and obtain the minimum energy consumption operating parameters;

[0012] Adjust the power distribution of the composite heating system according to the minimum energy consumption operating parameters to generate an efficient drying operation mode;

[0013] The feedback control algorithm is used to obtain updated data on dynamic control parameters and defect distribution characteristics, optimize the temperature and humidity gradient field, and determine the final drying process parameters;

[0014] Production control instructions are generated from the final drying process parameters and output to the drying equipment controller to achieve uniform diffusion of moisture in the film and eliminate surface defects.

[0015] The present invention provides a system for gradient drying of nitrile gloves after dipping, which mainly includes:

[0016] The moisture data acquisition module is used to obtain the moisture content data on the surface and inside of the film through the sensor array, generate a real-time moisture distribution matrix, and determine the moisture gradient change trend;

[0017] The gradient analysis module is used to calculate the local water evaporation rate based on the distribution matrix if the moisture gradient change trend exceeds the preset threshold, and obtain the dynamic temperature and humidity control parameters;

[0018] The composite heating control module is used to adopt a composite heating method combining infrared radiation with hot air circulation, adjust the heating intensity and wind speed according to dynamic control parameters, and generate a continuous temperature and humidity gradient field;

[0019] Defect detection module, used to obtain film surface images through a high-resolution imaging system, detect microcracks and white spot defect areas, and determine defect distribution characteristics;

[0020] A gradient field optimization module is used to adjust the local temperature and humidity gradient field according to the location of the defect area to optimize the water evaporation path if the defect distribution characteristics show microcracks or white spots exceeding the preset standards;

[0021] Energy consumption optimization module, which is used to extract the real-time moisture diffusion coefficient from the film moisture distribution matrix, combine it with the defect distribution characteristics, calculate the energy consumption optimization model, and obtain the minimum energy consumption operating parameters;

[0022] The power distribution module is used to adjust the power distribution of the composite heating system according to the minimum energy consumption operating parameters to generate an efficient drying operation mode;

[0023] The process control module is used to obtain updated data on dynamic control parameters and defect distribution characteristics through feedback control algorithms, optimize the temperature and humidity gradient field, and determine the final drying process parameters;

[0024] The process control module is used to generate production control instructions from the final drying process parameters and output them to the drying equipment controller to achieve uniform diffusion of moisture in the film and eliminate surface defects.

[0025] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0026] The present invention discloses a method and system for gradient drying of nitrile gloves after dipping. The moisture distribution data of the film is obtained through a sensor array, a real-time moisture distribution matrix is generated, and a high-resolution imaging system is used to detect surface defects. According to the moisture gradient change trend and the defect distribution characteristics, the local moisture evaporation rate is calculated to obtain dynamic temperature and humidity control parameters. A composite heating method of infrared radiation and hot air circulation is adopted to generate a continuous temperature and humidity gradient field according to the control parameters to optimize the moisture evaporation path. Combined with the moisture diffusion coefficient and the defect distribution characteristics, an energy consumption optimization model is calculated to obtain the minimum energy consumption operating parameters. The temperature and humidity gradient field is continuously optimized through a feedback control algorithm, and the drying process parameters are finally determined to achieve uniform diffusion of moisture in the film and elimination of surface defects, while reducing energy consumption. The present invention can effectively improve the drying quality and production efficiency of the film. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 The present invention is a flow chart of a method for gradient drying of nitrile gloves after dipping.

[0028] Figure 2 The figure is a schematic diagram of a method for gradient drying of nitrile gloves after dipping according to the present invention.

[0029] Figure 3 This is another schematic diagram of the method for gradient drying of nitrile gloves after dipping according to the present invention.

[0030] Figure 4 The figure is a schematic diagram of the framework of a system for gradient drying of nitrile gloves after dipping according to the present invention. DETAILED DESCRIPTION

[0031] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] like Figure 1-4 In this embodiment, a method for gradient drying of nitrile gloves after dipping may specifically include:

[0033] S101. Acquire moisture content data on the film surface and inside through a sensor array, generate a real-time moisture distribution matrix, and determine the moisture gradient change trend.

[0034] The sensor array collects moisture data on the film surface and interior to generate an original moisture data set. If the original moisture data set contains noise, a Gaussian filter algorithm is used to process the original moisture data set to obtain a denoised moisture data set. Based on the denoised moisture data set, a real-time moisture distribution matrix is generated to determine the moisture distribution pattern. The moisture difference between adjacent cells is calculated using the moisture distribution matrix to obtain a moisture gradient matrix. If there are outliers in the moisture gradient matrix, a median filter algorithm is used to process the moisture gradient matrix to obtain a smooth gradient matrix. Based on the smooth gradient matrix, the rate of change of the gradient over time is calculated to determine the moisture gradient change trend. Based on the moisture gradient change trend, a dynamic moisture distribution prediction model is generated to obtain a predicted moisture distribution matrix.

[0035] Specifically, a sensor array is used to acquire moisture content data on the film surface and interior. Multiple high-precision moisture sensors are placed on and within the film, with a spacing of 5 mm and a sampling frequency of 10 times per second, ensuring real-time and accurate data acquisition. The sensor data is transmitted via a wireless transmission module to a data processing center, where a Kalman filter algorithm is used to denoise the raw data. Filter parameters are set to a process noise covariance of 0.01 and a measurement noise covariance of 0.1 to improve data reliability. The denoised data is then interpolated to generate a real-time moisture distribution matrix. Bilinear interpolation is used with a 1 mm step size to ensure matrix continuity and smoothness. Once the moisture distribution matrix is generated, a gradient calculation algorithm is used to determine the moisture gradient trend. The gradient calculation uses the central difference method with a 2 mm step size. The gradient value at each point is calculated and a gradient distribution map is generated. The gradient distribution map is visualized using image processing techniques. Color mapping is used to map the gradient values to different colors, ranging from blue (low gradient) to red (high gradient), facilitating intuitive observation of moisture gradient changes. Finally, the moisture gradient change trend was predicted through a machine learning algorithm. The support vector machine (SVM) model was adopted. The training data set contained 1,000 samples and the test data set contained 200 samples. The model prediction accuracy reached more than 95%, providing a scientific basis for film moisture control.

[0036] S102: If the moisture gradient change trend exceeds a preset threshold, the local moisture evaporation rate is calculated according to the distribution matrix to obtain dynamic temperature and humidity control parameters.

[0037] Acquire moisture gradient data from environmental monitoring data, and the moisture gradient data is collected through a sensor array. If the change trend of the moisture gradient data exceeds a preset threshold, obtain real-time moisture distribution data through the sensor array to obtain a moisture distribution matrix. According to the moisture distribution matrix, use the finite difference method to calculate the local moisture evaporation rate to obtain the evaporation characteristic value of the local area. If the evaporation characteristic value exceeds the preset range, predict the temperature and humidity change trend through a regression analysis algorithm to obtain prediction parameters. According to the prediction parameters, use a dynamic programming algorithm to adjust the temperature and humidity control parameters to obtain optimized control instructions. Update the operating status of the environmental control equipment through the control instructions to obtain updated moisture gradient data. If the updated moisture gradient data still exceeds the preset threshold, optimize the moisture distribution matrix through iterative calculation to obtain a new evaporation rate. According to the new evaporation rate, readjust the temperature and humidity control parameters to determine the final dynamic control scheme.

[0038] Specifically, when the monitoring system detects that the soil moisture gradient in a certain area reaches a preset threshold of 0.15 grams per cubic centimeter per hour, the dynamic control mechanism is triggered. The system calls a pre-established two-dimensional moisture distribution matrix M (5×5 grid, 0.2-meter resolution), where each cell stores the current moisture content data, such as [0.18, 0.16, 0.12...]. The evaporation rate is calculated using an improved Penman-Monteith algorithm, introducing a wind speed correction factor k = 1.2 and a solar radiation parameter R = 450 W / m. 2 , aerodynamic resistance ra=25s / m, according to the formula ET=(0.408Δ(RG)+γ(900 / (T+273))u(es-ea)) / (Δ+γ(1+0.34u)), the evaporation rate range of each grid is 0.8-1.3mm / h. The calculation results are input into the temperature and humidity control model, and the parameters are adjusted by the PID controller. The proportional coefficient P is set to 0.6, the integral time Ti=120s, the differential time Td=30s, and the control instructions of the fan speed range [800,1200]rpm and the atomizer opening [15%, 35%] are output. At the same time, the moisture diffusion coefficient D=0.25cm in the historical database is updated. 2 / s, used for iterative calculation of the prediction model for the next cycle. The entire process is completed at the edge computing node, with latency controlled within 200ms to ensure real-time response.

[0039] S103, using a composite heating method combining infrared radiation with hot air circulation, adjusting the heating intensity and wind speed according to dynamic control parameters to generate a continuous temperature and humidity gradient field.

[0040] Ambient temperature and humidity data is acquired through sensors. Using a preset threshold range, the system determines whether the current environment meets the requirements for generating a continuous gradient field, thereby obtaining the initial environmental state. If the initial environmental state is below the preset threshold, heating is initiated via the infrared radiation module, and wind speed is adjusted in conjunction with the hot air circulation module to determine the initial energy allocation for the combined heating system. Based on real-time feedback data, a dynamic adjustment algorithm is used to update control parameters, resulting in an optimized heating intensity and wind speed control scheme. The updated temperature and humidity field distribution data is acquired through the environmental monitoring module to determine whether a continuous gradient field has been formed, thereby obtaining the current gradient field characteristics. If the gradient field characteristics do not meet the preset continuity criteria, the energy allocation model is used to recalculate the ratio of infrared radiation to hot air circulation and determine a new heating scheme. Based on the new heating scheme, the combined heating mechanism adjusts the heating intensity and wind speed, resulting in an updated temperature and humidity field distribution. An iterative optimization algorithm is used to fine-tune the control parameters and energy allocation to determine whether the final temperature and humidity field meets the continuous gradient requirements, resulting in a stable gradient field output.

[0041] Specifically, in the infrared radiation and hot air circulation composite heating system, the infrared emitter array is first radiated at a wavelength of 850nm, and the power density is set to 3.5kW / m 2 , combined with an axial fan to create convection at a wind speed of 6 m / s. The system features a built-in PID controller with a 200ms sampling period. It dynamically adjusts the PWM duty cycle based on real-time collected temperature and humidity data (e.g., when the temperature deviation ΔT = ±2°C). The algorithm uses the incremental PID formula Δu(k) = Kp[e(k)-e(k-1)]+Ki·e(k)+Kd[e(k)-2e(k-1)+e(k-2)], with a proportional coefficient Kp of 0.8 and an integration time Ti of 120s. When the material surface humidity gradient exceeds 15% RH / m, the hot air temperature is adjusted using a fuzzy control algorithm. The membership function uses a triangular distribution. The input variables are the humidity deviation and rate of change, and the output is the corrected heating power value. Defuzzification uses the center of gravity method. Temperature uniformity was optimized through ANSYS simulation, achieving a ±0.5°C gradient control within a 1.2m×0.8m workspace. The thermal imager sampling data was Kalman filtered and compared with the set curve. When the root mean square error exceeded 1.2°C, an adaptive weighting algorithm was triggered to adjust the group start and stop strategy of the infrared radiators. The air duct design was based on CFD simulation, with a Reynolds number of Re = 2.3×10 4 The k-ε turbulence model was used to ensure that the wind speed fluctuation was less than ±0.3 m / s. The SIMPLE algorithm was used for the coupled calculation of temperature field and velocity field, and the convergence residual was set to 10 -6 The entire system communicates with the host computer via the OPC-UA protocol, with a data update frequency of 10 Hz. Historical data is stored in a time series database, and the compression algorithm uses Delta-of-Delta encoding, with a compression ratio of 8:1.

[0042] S104. Obtain an image of the film surface through a high-resolution imaging system, detect microcracks and white spot defect areas, and determine defect distribution characteristics.

[0043] A first image of the film surface is acquired, generated by a high-resolution imaging system. If the first image contains noise, a median filter algorithm is used to denoise the first image to obtain a second image. Based on the grayscale value distribution of the second image, an Otsu threshold segmentation algorithm is used to generate a third image to determine the candidate defect area. If the grayscale value of the candidate defect area in the third image exceeds a preset threshold, morphological processing is used to optimize the third image to generate a fourth image to determine the defect area. Based on the defect area in the fourth image, the geometric features of the defect area are calculated to generate a fifth image to determine the distribution of microcracks and white spot defects. Based on the defect distribution in the fifth image, the spatial location features of the defects are statistically analyzed to generate a sixth image to obtain defect distribution characteristics.

[0044] Specifically, a high-resolution imaging system based on a line scan camera was constructed, employing a 50-megapixel CMOS sensor with a 10x optical zoom lens to capture film surface images at a resolution of 5 microns per pixel. The imaging system was equipped with a ring-shaped LED light source, set to a color temperature of 6500K, and controlled at 1500 lux to ensure uniform illumination. After image acquisition, a Gaussian filter (σ = 1.2) was first used for noise reduction. Then, an improved Canny edge detection algorithm was used with high and low thresholds of 0.3 and 0.1, respectively, combined with 8-neighborhood connected domain analysis to extract potential defect regions. For microcrack detection, a multi-scale Retinex algorithm was applied to enhance contrast. In the HSV color space, the saturation channel was binarized with a threshold of 0.25. Noise was removed by combining a morphological opening operation (3×3 circular structuring element). For white spot defects, a threshold of 85 was set in the L channel of the LAB color space, and region growing was performed (seed point similarity threshold ΔE < 5). During the defect feature quantification phase, the geometric features of each connected domain were calculated, including area (minimum 10 pixels), aspect ratio (range 1.5-5.0), and circularity (0.3-0.7). Automatic defect type identification was achieved using a support vector machine classifier (RBF kernel function, C = 1.0, γ = 0.5). Finally, a spatial statistical method was used to calculate the defect distribution density. Kernel density estimation was performed using a 50×50 pixel grid unit (bandwidth parameter h = 30). A defect heat map was generated, and the coordinate deviation of the largest clustered area (±0.1 mm positioning accuracy) was output.

[0045] S105. If the defect distribution characteristics show that microcracks or white spots exceed the preset standard, the local temperature and humidity gradient field is adjusted according to the location of the defect area to optimize the water evaporation path.

[0046] The sensor acquires surface defect distribution data on the material, generates a defect distribution characteristic image, and determines the location of microcracks and white spots. If the microcracks or white spots in the defect distribution characteristic image exceed a preset threshold, an image segmentation algorithm is used to extract the boundary of the defect area and obtain the location coordinates of the defect area. Based on the location coordinates of the defect area, the local temperature and humidity gradient field distribution is calculated, and gradient field adjustment parameters are generated. The gradient field adjustment parameters are used to control the temperature and humidity control equipment, adjust the local temperature and humidity distribution, and generate optimized water evaporation path data. A support vector machine algorithm is used to determine whether the water evaporation path data meets the preset optimization criteria. If the water evaporation path data does not meet the preset optimization criteria, the gradient field adjustment parameters are adjusted based on the deviation value, and the local temperature and humidity distribution is regenerated. By iteratively optimizing the temperature and humidity distribution data, the optimized evaporation path of the final defect area is generated, and the water evaporation efficiency is determined.

[0047] Specifically, when it is detected that the density of microcracks on the surface of the material exceeds the preset threshold (such as 5 per square millimeter) or the diameter of the white spot is greater than 50 microns, the system will start the temperature and humidity field control algorithm based on finite element analysis. First, a high-precision infrared thermal imager (resolution 0.1°C) is used to scan the defective area, establish a three-dimensional heat conduction model, and use the ANSYS Fluent solver to calculate the optimal temperature gradient curve. A local heating zone of 60±2°C is set in the crack-dense area, and the adjacent areas are maintained at 45°C to form a gradient difference of 15°C. At the same time, the moisture migration path is calculated through the porous medium evaporation model, and the real-time data collected by the humidity sensor (sampling frequency 10Hz) is input into the LSTM neural network to predict the evaporation rate. When the predicted value is lower than 0.2g / (m 2 ·s), the nozzle array is automatically adjusted to form a low-temperature drying zone of 40% RH 20mm above the defect area, while maintaining a high-humidity zone of 60% RH at the edge of the material. The entire regulation process adopts PID closed-loop control, with a temperature regulation accuracy of ±0.5°C and a humidity fluctuation range of ±3% RH. The parameters are dynamically optimized by real-time monitoring of the crack closure rate (μm / s) and the white spot area shrinkage rate (% / min). When the crack width is reduced to less than 10μm and the white spot area is reduced by 80%, it is judged to be qualified. The system updates the defect feature extraction results based on convolutional neural networks (CNN) every 5 minutes, and couples them with the material stress field simulation data for analysis to ensure that temperature and humidity adjustments do not induce new stress concentrations.

[0048] S106. Extract the real-time moisture diffusion coefficient from the film moisture distribution matrix, combine it with the defect distribution characteristics, calculate the energy consumption optimization model, and obtain the minimum energy consumption operating parameters.

[0049] A real-time moisture distribution matrix is obtained from the film surface sensor, and the spatial frequency characteristics of the moisture distribution matrix are extracted using a two-dimensional Fourier transform to obtain spatial characteristic data of the moisture distribution. Based on the spatial characteristic data and the film material properties, the real-time moisture diffusion coefficient is calculated using Fick's diffusion law, where the formula is \(J = D\frac{\partialC}{\partialx}\), where \(J\) represents the diffusion flux, \(D\) represents the diffusion coefficient, \(C\) represents the moisture concentration, and \(x\) represents the diffusion distance. If the real-time diffusion coefficient exceeds a preset threshold range, the defect location distribution is obtained from the defect detection equipment and analyzed using the K-means clustering algorithm to obtain a defect distribution feature vector. According to the real-time diffusion coefficient and the defect distribution characteristic vector, an energy consumption optimization model is constructed and solved by a linear programming algorithm, where the objective function is \(E=\sum(P_iT_i)\), \(E\) represents the total energy consumption, \(P_i\) represents the power of the i-th operating parameter, and \(T_i\) represents the operating time, and the minimum energy consumption parameter is obtained. According to the minimum energy consumption parameter, the operating parameters of the film production equipment are adjusted, and the equipment power and operating time are updated using a closed-loop control system to obtain an optimized operating state. The real-time moisture distribution matrix is obtained from the optimized operating state, and the parameters of the energy consumption optimization model are iteratively optimized using a gradient descent algorithm to obtain continuously optimized operating parameters. If the rate of change of the continuously optimized operating parameters and the parameters of the previous round is lower than the preset threshold, the final minimum energy consumption operating parameters are determined and output to the production control system to obtain a stable operating state.

[0050] Specifically, when extracting the real-time moisture diffusion coefficient from the film moisture distribution matrix, a numerical calculation model based on the finite difference method can be used. For example, in a 10×10 grid matrix, assuming that the moisture concentration difference between adjacent nodes is 0.5 g / m 3 , the grid spacing is 1mm, the time step is set to 1s, and the discretization formula D=Δx is obtained by Fick's second law. 2 / (4Δt)·ln(C1 / C2) to obtain the diffusion coefficient D of 2.3×10 -9 m 2 / s. Combined with the defect distribution characteristics, if infrared detection shows that the defect area accounts for 15% and the diffusion coefficient of the defect area increases to 3.1×10 -9 m 2 / s, it is necessary to establish a spatial weight function w(x,y)=1+0.5·I_defect(x,y) for correction, where I_defect is the defect indication function. In the energy consumption optimization model construction phase, the drying temperature T (unit K) and wind speed v (unit m / s) are used as decision variables, and the heat conduction equation is used. Establish energy consumption function, where k = 0.026W / mK is the thermal conductivity of air, h = 15W / m 2 K is the convection coefficient. A genetic algorithm is used for multi-objective optimization. The population size is set to 100, the crossover probability is 0.8, and the mutation probability is 0.05. After 200 generations of iteration, the Pareto frontier is obtained. Finally, T = 353K and v = 2.5m / s are selected as the parameter combination with the lowest energy consumption. At this time, the energy consumption per unit area is 1.2kW·h / m 2 , a 22% reduction compared to the initial parameters. This process couples computational fluid dynamics simulation with a machine learning proxy model to map defect characteristics to an energy sensitivity matrix, ensuring that the optimization results meet the actual production line constraints.

[0051] S107. Adjust the power distribution of the composite heating system according to the minimum energy consumption operation parameter to generate a high-efficiency drying operation mode.

[0052] Real-time operating data of the composite heating system, including power distribution, energy consumption, and drying efficiency parameters, is acquired to determine an initial operating state. A preset energy consumption optimization algorithm is used to analyze the relationship between power distribution and energy consumption in the initial operating state, determining a power distribution scheme that achieves the lowest energy consumption. A determination is made as to whether the power distribution scheme meets a preset drying efficiency threshold. If so, operating parameters are adjusted based on the power distribution scheme to generate a candidate operating mode. If not, the power distribution scheme is iteratively optimized to obtain a candidate operating mode that meets the preset drying efficiency threshold. Based on the candidate operating mode, the heating unit power of the composite heating system is adjusted using a heating control algorithm to obtain real-time adjusted operating parameters. Drying efficiency data is extracted from the real-time adjusted operating parameters to determine whether a preset high-efficiency drying target is achieved. If so, the candidate operating mode is fixed as the current operating mode. If not, the process returns to the power distribution scheme optimization step to regenerate a candidate operating mode. Based on the current operating mode, parameter configuration for a high-efficiency drying operating mode is generated to obtain a final mode generation result. An operation log is extracted from the final mode generation result and stored in a preset database to obtain historical operating mode data.

[0053] Specifically, in the composite heating system, the least squares method is used to fit the optimal power allocation model by analyzing the current ambient temperature, humidity, and material properties. Assuming an ambient temperature of 25°C, humidity of 60%, an initial material moisture content of 30%, and a target moisture content of 10%, the system first collects data in real time and uses a Kalman filter algorithm to denoise the data to ensure accuracy. Next, based on the first law of thermodynamics, an energy balance equation is established to calculate the total system energy consumption. Assuming a total system power of 10 kW, a genetic algorithm is used to optimize power allocation, resulting in a power of 6 kW for Heater 1 and 4 kW for Heater 2, achieving the lowest system energy consumption. Furthermore, a PID controller is used to precisely control the heaters, ensuring temperature fluctuations within a ±1°C range. Simulations verify that this mode improves drying efficiency by 15% and reduces energy consumption by 10%. Finally, the optimized parameters are stored in the system database for subsequent reference, achieving a highly efficient drying operation mode.

[0054] S108. Obtain updated data of dynamic control parameters and defect distribution characteristics through a feedback control algorithm, optimize the temperature and humidity gradient field, and determine the final drying process parameters.

[0055] Environmental monitoring data and defect distribution characteristics are acquired through sensors. The environmental monitoring data includes temperature and humidity data, and the defect distribution characteristics include defect location and type. A convolutional neural network is used to analyze the environmental monitoring data and defect distribution characteristics to determine defect detection accuracy. If the defect detection accuracy falls below a preset threshold, dynamic control parameters are adjusted using a feedback control algorithm to obtain optimized control parameters. Based on the optimized control parameters, a gradient descent algorithm is used to update the temperature and humidity gradient field to determine the gradient field distribution, which represents the spatial variation of temperature and humidity. By analyzing the gradient field distribution and the defect distribution characteristics, it is determined whether the process stability meets the preset standard, and a stability assessment result is obtained. If the stability assessment result does not meet the standard, the drying process parameters are adjusted using a parameter optimization strategy to determine the final process parameters. Based on the final process parameters, real-time control capability data is acquired, and it is determined whether the real-time control capability data meets dynamic requirements to obtain a control capability assessment result. Based on the control capability assessment result, the data update frequency is updated to determine the real-time control capability.

[0056] Specifically, the feedback control algorithm is used to obtain updated data on dynamic control parameters and defect distribution characteristics. First, sensors are used to collect temperature and humidity data in the drying environment in real time, such as a temperature range of 30°C to 60°C and a humidity range of 20% to 80%. The PID control algorithm is used to dynamically adjust the temperature and humidity, setting the target temperature to 45°C and the target humidity to 40%. The output power of the heating and humidification equipment is adjusted in real time through the proportional coefficient Kp = 1.2, the integral time Ti = 10 minutes, and the differential time Td = 2 minutes to ensure that the temperature and humidity are stable near the target values. At the same time, image processing technology is used to analyze the defect distribution characteristics during the drying process. For example, a convolutional neural network (CNN) is used to detect defects on the surface images of the dried products, identify defects such as cracks and bubbles, and calculate their area proportions. For example, the crack area accounts for 0.5% and the bubble area accounts for 0.3%. Based on these data, a genetic algorithm was used to optimize the temperature and humidity gradient field. The initial population size was set to 50, the crossover probability was 0.8, and the mutation probability was 0.1. Through iterative optimization, the optimal temperature and humidity gradient was determined to be a gradual increase in temperature from 40°C to 50°C, and a gradual decrease in humidity from 50% to 30%. Finally, the drying process parameters were determined to be a drying time of 120 minutes, and the temperature and humidity gradients were as described above to ensure product quality and drying efficiency.

[0057] S109: Generate production control instructions based on the final drying process parameters and output them to the drying equipment controller to achieve uniform diffusion of moisture in the film and eliminate surface defects.

[0058] The drying process parameters are obtained from a database, and the drying process parameters include temperature, humidity, and time distribution. A process parameter set is obtained by analyzing the drying process parameters. If the process parameter set meets a preset threshold, a production control instruction is generated based on the process parameter set, and an instruction sequence is determined. The instruction sequence is transmitted to the drying equipment controller through an instruction output mechanism to obtain the controller response status. Based on the controller response status, the film moisture diffusion rate is adjusted to determine the uniform moisture distribution result. If the uniform moisture distribution result does not meet the preset standard, defect detection technology is used to analyze surface defects to obtain defect distribution data. The process parameters are optimized based on the defect distribution data, and new production control instructions are generated and transmitted to the drying equipment controller. Based on the film quality monitoring data, the moisture diffusion and defect elimination effects are analyzed to determine the final process parameters.

[0059] Specifically, during the process of generating production control instructions for the final drying process parameters, sensors first collect real-time data on the drying oven's temperature, humidity, and film surface moisture distribution. For example, the temperature is set at 85±2°C and the humidity is controlled at 15%±3%. Simultaneously, an infrared imager monitors the film's surface moisture gradient. An adjustment mechanism is triggered when a local moisture content deviation exceeds 5%. Based on this collected data, a PID control algorithm is used to calculate the adjustment value for the hot air flow rate within the drying oven. The algorithm sets the proportional coefficient Kp to 0.8, the integral time Ti to 120 seconds, and the differential time Td to 30 seconds. This real-time correction of the hot air flow rate ensures uniform moisture distribution across the film. To eliminate surface defects, a machine vision system inspects the film's surface. If bubbles or cracks are detected, the system automatically adjusts the drying oven's pressure fluctuation frequency to 0.5Hz, with an amplitude within ±50Pa. The temperature gradient is also adjusted to a zoned control mode with a front zone of 90°C and a rear zone of 75°C. All control parameters are transmitted to the drying equipment controller via the OPC UA protocol. Based on these commands, the controller adjusts the heating element power and fan speed in real time, dynamically adjusting the heating power from 8kW to 6.5kW and reducing the fan speed from 1200rpm to 1000rpm to ensure precise execution of process parameters. Throughout the entire process, data is collected 10 times per second, with a control command response delay of less than 200 milliseconds. Through closed-loop feedback, the film moisture content is ultimately stabilized within the target range of 2.0% ± 0.3%.

[0060] The present invention provides a system for gradient drying of nitrile gloves after dipping, which mainly includes:

[0061] The moisture data acquisition module is used to obtain the moisture content data on the surface and inside of the film through the sensor array, generate a real-time moisture distribution matrix, and determine the moisture gradient change trend;

[0062] The gradient analysis module is used to calculate the local water evaporation rate based on the distribution matrix if the moisture gradient change trend exceeds the preset threshold, and obtain the dynamic temperature and humidity control parameters;

[0063] The composite heating control module is used to adopt a composite heating method combining infrared radiation with hot air circulation, adjust the heating intensity and wind speed according to dynamic control parameters, and generate a continuous temperature and humidity gradient field;

[0064] Defect detection module, used to obtain film surface images through a high-resolution imaging system, detect microcracks and white spot defect areas, and determine defect distribution characteristics;

[0065] A gradient field optimization module is used to adjust the local temperature and humidity gradient field according to the location of the defect area to optimize the water evaporation path if the defect distribution characteristics show microcracks or white spots exceeding the preset standards;

[0066] Energy consumption optimization module, which is used to extract the real-time moisture diffusion coefficient from the film moisture distribution matrix, combine it with the defect distribution characteristics, calculate the energy consumption optimization model, and obtain the minimum energy consumption operating parameters;

[0067] The power distribution module is used to adjust the power distribution of the composite heating system according to the minimum energy consumption operating parameters to generate an efficient drying operation mode;

[0068] The process control module is used to obtain updated data on dynamic control parameters and defect distribution characteristics through feedback control algorithms, optimize the temperature and humidity gradient field, and determine the final drying process parameters;

[0069] The process control module is used to generate production control instructions from the final drying process parameters and output them to the drying equipment controller to achieve uniform diffusion of moisture in the film and eliminate surface defects.

[0070] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any appropriate manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations. In addition, the various different embodiments of the present invention can also be arbitrarily combined, as long as they do not violate the concept of the present invention, they should also be regarded as the content disclosed by the present invention.

Claims

1. A method for gradient drying of nitrile gloves after dipping, characterized in that: The method comprises the following steps: S101, acquiring moisture content data on the surface and inside of the film through a sensor array, generating a real-time moisture distribution matrix, and determining a moisture gradient change trend; S102: If the moisture gradient change trend exceeds a preset threshold, the local moisture evaporation rate is calculated according to the distribution matrix to obtain dynamic temperature and humidity control parameters; S103, using a composite heating method combining infrared radiation with hot air circulation, adjusting the heating intensity and wind speed according to dynamic control parameters to generate a continuous temperature and humidity gradient field; S104. Acquire an image of the film surface through a high-resolution imaging system, detect microcracks and white spot defect areas, and determine defect distribution characteristics; S105. If the defect distribution characteristics show that microcracks or white spots exceed the preset standard, the local temperature and humidity gradient field is adjusted according to the location of the defect area to optimize the water evaporation path; S106, extracting the real-time moisture diffusion coefficient from the film moisture distribution matrix, combining it with the defect distribution characteristics, calculating the energy consumption optimization model, and obtaining the minimum energy consumption operating parameters; S107, adjusting the power distribution of the composite heating system according to the minimum energy consumption operating parameters to generate a high-efficiency drying operation mode; S108. Obtain updated data of dynamic control parameters and defect distribution characteristics through a feedback control algorithm, optimize the temperature and humidity gradient field, and determine the final drying process parameters; S109: Generate production control instructions based on the final drying process parameters and output them to the drying equipment controller to achieve uniform diffusion of moisture in the film and eliminate surface defects.

2. The method for gradient drying of nitrile gloves after dipping according to claim 1, characterized in that: The S101 includes: The sensor array collects moisture data on the surface and inside of the film to generate an original moisture data set; If the original moisture data set contains noise, a Gaussian filter algorithm is used to process the original moisture data set to obtain a denoised moisture data set; generating a real-time moisture distribution matrix based on the denoised moisture data set to determine the moisture distribution pattern; Calculate the moisture difference between adjacent cells through the moisture distribution matrix to obtain a moisture gradient matrix; If there are abnormal values in the moisture gradient matrix, a median filter algorithm is used to process the moisture gradient matrix to obtain a smooth gradient matrix; Calculating the rate of change of the gradient over time based on the smoothed gradient matrix to determine the trend of the moisture gradient change; A dynamic moisture distribution prediction model is generated based on the moisture gradient change trend to obtain a predicted moisture distribution matrix.

3. The method for gradient drying of nitrile gloves after dipping according to claim 1, characterized in that: The S102 includes: Acquiring moisture gradient data from environmental monitoring data, wherein the moisture gradient data is collected by a sensor array; If the change trend of the moisture gradient data exceeds a preset threshold, real-time moisture distribution data is acquired through the sensor array to obtain a moisture distribution matrix; According to the moisture distribution matrix, the local moisture evaporation rate is calculated using the finite difference method to obtain the evaporation characteristic value of the local area; If the evaporation characteristic value exceeds the preset range, the temperature and humidity change trend is predicted by the regression analysis algorithm to obtain the prediction parameters; According to the predicted parameters, a dynamic programming algorithm is used to adjust the temperature and humidity control parameters to obtain optimized control instructions; Update the operating status of the environmental control device through the control instruction to obtain updated moisture gradient data; If the updated moisture gradient data still exceeds the preset threshold, optimizing the moisture distribution matrix through iterative calculation to obtain a new evaporation rate; According to the new evaporation rate, the temperature and humidity control parameters are readjusted to determine the final dynamic control plan.

4. The method for gradient drying of nitrile gloves after dipping according to any one of claims 1 to 3, characterized in that: The S103 includes: The ambient temperature and humidity data are obtained through sensors, and a preset threshold range is used to determine whether the current environment meets the requirements for generating a continuous gradient field to obtain the initial environmental state; If the initial environmental state is lower than the preset threshold, heating is started through the infrared radiation module, and the wind speed is adjusted in combination with the hot air circulation module to determine the initial energy distribution of the composite heating; Based on real-time feedback data, a dynamic adjustment algorithm is used to update the control parameters to obtain the optimized heating intensity and wind speed control scheme; Obtain updated temperature and humidity field distribution data through the environmental monitoring module, determine whether a continuous gradient field is formed, and obtain the current gradient field characteristics; If the gradient field characteristics do not meet the preset continuity standard, the ratio of infrared radiation to hot air circulation is recalculated through the energy distribution model to determine a new heating plan; According to the new heating scheme, a composite heating mechanism is used to adjust the heating intensity and wind speed to obtain an updated temperature and humidity field distribution; The control parameters and energy distribution are fine-tuned through iterative optimization algorithms to determine whether the final temperature and humidity field meets the continuous gradient requirements and obtain a stable gradient field output.

5. The method for gradient drying of nitrile gloves after dipping according to any one of claims 1 to 3, characterized in that: The S104 includes: Acquiring a first image of the surface of the adhesive film, wherein the first image is generated by a high-resolution imaging system; If the first image has noise, performing denoising on the first image using a median filtering algorithm to obtain a second image; Based on the grayscale value distribution of the second image, an Otsu threshold segmentation algorithm is used to generate a third image to determine the defect candidate area; If the grayscale value of the defect candidate area in the third image exceeds a preset threshold, the third image is optimized using morphological processing to generate a fourth image to determine the defect area; Calculating geometric features of the defective area in the fourth image to generate a fifth image and determine the distribution of microcracks and white spot defects; According to the defect distribution of the fifth image, spatial position features of the defects are counted to generate a sixth image to obtain defect distribution features.

6. The method for gradient drying of nitrile gloves after dipping according to any one of claims 1 to 3, characterized in that: The S105 includes: The sensor acquires the surface defect distribution data of the material, generates a defect distribution feature image, and determines the location of microcracks and white spots; If the number of microcracks or white spots in the defect distribution feature image exceeds a preset threshold, an image segmentation algorithm is used to extract the defect area boundary to obtain the position coordinates of the defect area; Calculating the local temperature and humidity gradient field distribution according to the position coordinates of the defect area and generating gradient field adjustment parameters; By adjusting the parameters of the gradient field, the temperature and humidity regulating equipment is controlled to adjust the local temperature and humidity distribution, thereby generating optimized water evaporation path data; Using a support vector machine algorithm to determine whether the water evaporation path data meets a preset optimization standard; If the water evaporation path data does not meet the preset optimization criteria, adjusting the gradient field adjustment parameters according to the deviation value to regenerate the local temperature and humidity distribution; By iteratively optimizing the temperature and humidity distribution data, the optimized evaporation path of the final defect area is generated to determine the water evaporation efficiency.

7. The method for gradient drying of nitrile gloves after dipping according to any one of claims 1 to 3, characterized in that: The S107 includes: Acquiring real-time operating data of the composite heating system, the real-time operating data including power distribution, energy consumption level, and drying efficiency parameters, and obtaining an initial operating state; Analyzing the relationship between power allocation and energy consumption level in the initial operating state using a preset energy consumption optimization algorithm to determine a power allocation scheme with minimum energy consumption; determining whether the power allocation scheme satisfies a preset drying efficiency threshold, and if so, adjusting operating parameters based on the power allocation scheme to generate a candidate operating mode; If not, iteratively optimizing the power allocation scheme to obtain a candidate operating mode that meets the preset drying efficiency threshold; Adjusting the power of the heating unit of the composite heating system using a heating control algorithm according to the candidate operating mode to obtain real-time adjusted operating parameters; Extracting drying efficiency data from the real-time adjusted operating parameters to determine whether a preset high-efficiency drying target is achieved; if so, fixing the candidate operating mode as the current operating mode; If not, returning to the optimization step of the power allocation scheme to regenerate the candidate operating mode; Generating a parameter configuration of a high-efficiency drying operation mode according to the current operation mode to obtain a final mode generation result; The operation log is extracted from the final mode generation result and stored in a preset database to obtain historical data of the operation mode.

8. The method for gradient drying of nitrile gloves after dipping according to any one of claims 1 to 3, characterized in that: The S108 includes: Acquire environmental monitoring data and defect distribution characteristics through sensors, wherein the environmental monitoring data includes temperature and humidity data, and the defect distribution characteristics include defect location and type; A convolutional neural network is used to analyze the environmental monitoring data and defect distribution characteristics to obtain defect detection accuracy; If the defect detection accuracy is lower than a preset threshold, the dynamic control parameters are adjusted through a feedback control algorithm to obtain optimized control parameters; According to the optimized control parameters, a gradient descent algorithm is used to update the temperature and humidity gradient field to determine the gradient field distribution, wherein the gradient field distribution represents the variation characteristics of temperature and humidity in space; By analyzing the gradient field distribution and the defect distribution characteristics, determining whether the process stability meets the preset standard, and obtaining a stability evaluation result; If the stability assessment result does not meet the standards, the drying process parameters are adjusted through parameter optimization strategy to determine the final process parameters; Acquiring real-time control capability data based on the final process parameters, determining whether the real-time control capability data meets dynamic requirements, and obtaining a control capability evaluation result; According to the control capability evaluation result, the data update frequency is updated to determine the real-time control capability.

9. A system for gradient drying of nitrile gloves after dipping, characterized in that: The system is used to implement the method for gradient drying of nitrile gloves after dipping according to any one of claims 1 to 8, and the system comprises: The moisture data acquisition module is used to obtain the moisture content data on the surface and inside of the film through the sensor array, generate a real-time moisture distribution matrix, and determine the moisture gradient change trend; The gradient analysis module is used to calculate the local water evaporation rate based on the distribution matrix if the moisture gradient change trend exceeds the preset threshold, and obtain the dynamic temperature and humidity control parameters; The composite heating control module is used to adopt a composite heating method combining infrared radiation with hot air circulation, adjust the heating intensity and wind speed according to dynamic control parameters, and generate a continuous temperature and humidity gradient field; Defect detection module, used to obtain film surface images through a high-resolution imaging system, detect microcracks and white spot defect areas, and determine defect distribution characteristics; A gradient field optimization module is used to adjust the local temperature and humidity gradient field according to the location of the defect area to optimize the water evaporation path if the defect distribution characteristics show microcracks or white spots exceeding the preset standards; Energy consumption optimization module, which is used to extract the real-time moisture diffusion coefficient from the film moisture distribution matrix, combine it with the defect distribution characteristics, calculate the energy consumption optimization model, and obtain the minimum energy consumption operating parameters; The power distribution module is used to adjust the power distribution of the composite heating system according to the minimum energy consumption operating parameters to generate an efficient drying operation mode; The process control module is used to obtain updated data on dynamic control parameters and defect distribution characteristics through feedback control algorithms, optimize the temperature and humidity gradient field, and determine the final drying process parameters; The process control module is used to generate production control instructions from the final drying process parameters and output them to the drying equipment controller to achieve uniform diffusion of moisture in the film and eliminate surface defects.

Images