Production control method for antistatic and conductive geomembrane
By collecting temperature data and performing EMD denoising, constructing the response coefficient and adjustment coefficient of the Gaussian membership function, and combining it with fuzzy PID for adaptive temperature control, the problem of unstable temperature in the production of antistatic and conductive geomembranes was solved, and more stable production quality was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG GEOSINO NEW MATERIAL CO LTD
- Filing Date
- 2023-10-08
- Publication Date
- 2026-05-19
AI Technical Summary
In the production process of antistatic and conductive geomembranes, the existing technology suffers from unstable temperature control, which is greatly affected by external factors, leading to fluctuations in production quality. The fuzzy PID algorithm fails, making it difficult to achieve effective temperature control.
By collecting temperature data, performing EMD algorithm denoising, obtaining the fluctuation coefficient and difference value of the reconstructed data, constructing the response coefficient and adjustment coefficient of the Gaussian membership function, and combining it with fuzzy PID for temperature control, adaptive adjustment is achieved.
It improves the robustness of temperature control, adapts to more production environments, ensures the stability of antistatic and conductive geomembrane production quality, and reduces the sensitivity of the temperature control system.
Smart Images
Figure CN117341177B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of electrical control system technology, and specifically to a production control method for antistatic and conductive geomembranes. Background Technology
[0002] In the production process of antistatic and conductive geomembranes, the raw materials need to be heated and then extruded through an extruder. However, because antistatic and conductive geomembranes are extremely sensitive to extrusion temperature, the extrusion temperature should not fluctuate too much, otherwise it will affect the production quality of the antistatic and conductive geomembranes.
[0003] Currently, fuzzy PID algorithms are commonly used for temperature control during the extrusion of antistatic and conductive geomembranes. However, due to the differences in the input of raw materials for antistatic and conductive geomembranes, it is necessary to adjust and compensate for the extrusion temperature of the antistatic and conductive geomembranes in real time during the temperature control process to ensure a more stable temperature during the production of antistatic and conductive geomembranes.
[0004] Constructing a fuzzy function can reduce the sensitivity to temperature control during the extrusion of antistatic and conductive geomembranes. However, due to the different temperatures during the production of antistatic and conductive geomembranes, and the significant influence of external factors such as weather on the extrusion temperature, the actual temperature data fluctuates greatly, causing the fuzzy function to fail and affecting the production quality of antistatic and conductive geomembranes. Summary of the Invention
[0005] To address the above problems, this invention provides a production control method for antistatic and conductive geomembranes, comprising the following steps:
[0006] Collect temperature data during the geomembrane production process; obtain data segments based on the temperature data, denoise the data segments, and obtain reconstructed data; obtain the difference values of each data point in the reconstructed data, and obtain the first fluctuation coefficient of the reconstructed data based on the difference values of each data point in the reconstructed data;
[0007] Obtain the second fluctuation coefficient for each data point in the reconstructed data; obtain the fluctuation difference of the reconstructed data based on the difference in the second fluctuation coefficients between adjacent data points; obtain the time interval to be stabilized based on the second fluctuation coefficients of all data points in the reconstructed data.
[0008] Based on the fluctuation differences of the reconstructed data and the time interval to be stabilized, obtain the response coefficient of the Gaussian membership function; based on the first fluctuation coefficient of the reconstructed data and the response coefficient of the Gaussian membership function, obtain the adjustment coefficient of the Gaussian membership function.
[0009] The variance of the Gaussian membership function is obtained based on the adjustment coefficient of the Gaussian membership function, and the temperature is controlled during the geomembrane production process based on the variance of the Gaussian membership function.
[0010] Preferably, the specific steps for denoising the data segment and obtaining the reconstructed data are as follows:
[0011] The data segment is decomposed using the EMD algorithm to obtain multiple IMF components and a residual; the number of peak points in each IMF component is counted, and a line chart of the number of peak points is constructed with the IMF component number as the horizontal axis and the number of peak points corresponding to each IMF component as the vertical axis.
[0012] Connect the first and last data points in the peak point count line graph to form a straight line, which serves as the fitting line. Substitute the number of each IMF component into the equation of the fitting line to obtain the fitted value of the peak point count for each IMF component. Calculate the absolute value of the difference between the actual peak point count and the fitted value of the peak point count for each IMF component. Take the data point in the peak point count line graph corresponding to the IMF component with the largest absolute value of the difference as the inflection point.
[0013] Remove the IMF components corresponding to all data points to the left of the inflection point, and sum the residuals and the remaining IMF components after removal to complete the data reconstruction and obtain the reconstructed data.
[0014] Preferably, the specific steps for obtaining the difference values of each data point in the reconstructed data are as follows:
[0015] Obtain the absolute value of the difference between the temperature value corresponding to each data point in the reconstructed data and the preset reference temperature, and record it as the difference value of each data point in the reconstructed data.
[0016] Preferably, the specific steps for obtaining the first fluctuation coefficient of the reconstructed data based on the difference values of each data point in the reconstructed data are as follows:
[0017] Obtain the information entropy of the difference values corresponding to all data points in the reconstructed data, obtain the mean of the difference values corresponding to all data points in the reconstructed data, and use the product of the mean of the difference values corresponding to all data points in the reconstructed data and the information entropy as the fluctuation coefficient of the reconstructed data.
[0018] Preferably, the specific steps for obtaining the second fluctuation coefficient of each data point in the reconstructed data are as follows:
[0019] Obtain the data segment corresponding to each data point in the reconstructed data, and obtain the first fluctuation coefficient of the data segment corresponding to each data point, which is used as the second fluctuation coefficient of each data point in the reconstructed data.
[0020] Preferably, the specific steps for obtaining the fluctuation difference of the reconstructed data based on the difference in the second fluctuation coefficient between adjacent data points in the reconstructed data are as follows:
[0021] Obtain the absolute value of the difference between the second fluctuation coefficients of all adjacent data points in the reconstructed data, and take the mean of all the absolute values of the differences as the fluctuation difference of the reconstructed data.
[0022] Preferably, the specific steps for obtaining the time interval to be stabilized based on the second fluctuation coefficient of all data points in the reconstructed data are as follows:
[0023] The second fluctuation coefficient of all data points in the reconstructed data is fitted with a polynomial using the least squares method to obtain the fitting function. For each data point in the reconstructed data, the time corresponding to each data point is successively substituted into the fitting function to obtain the predicted fluctuation coefficient corresponding to each data point. The data point corresponding to the predicted fluctuation coefficient of 0 closest to the current time is obtained as the stable data point. The time difference between the start time and the time corresponding to the stable data point is taken as the time interval to be stabilized.
[0024] Preferably, the specific steps for obtaining the response coefficients of the Gaussian membership function based on the fluctuation differences of the reconstructed data and the time interval to be stabilized are as follows:
[0025] D = E × F;
[0026] Where D is the response coefficient of the Gaussian membership function, E is the fluctuation difference of the reconstructed data, and F is the time interval to be stabilized.
[0027] Preferably, the specific steps for obtaining the adjustment coefficient of the Gaussian membership function based on the first fluctuation coefficient of the reconstructed data and the response coefficient of the Gaussian membership function are as follows:
[0028]
[0029] Where G represents the adjustment coefficient of the Gaussian membership function, C is the first fluctuation coefficient of the reconstructed data, D is the response coefficient of the Gaussian membership function, and h and k are hyperparameters.
[0030] Preferably, the specific steps for obtaining the variance value of the Gaussian membership function based on the adjustment coefficient of the Gaussian membership function, and for controlling the temperature during the geomembrane production process based on the variance value of the Gaussian membership function, are as follows:
[0031] The variance of the Gaussian membership function is obtained by calculating the adjustment coefficient of the Gaussian membership function, thus obtaining the Gaussian membership function. The error value and error rate of change are obtained based on the latest temperature data. The error value and error rate of change are used as inputs to the fuzzy PID controller. The fuzzy PID controller is then used to control the production temperature of antistatic and conductive geomembranes based on the Gaussian membership function.
[0032] The beneficial effects of the technical solution of this invention are as follows: This invention collects temperature data during the geomembrane production process, denoises the data segments in the temperature data to obtain reconstructed data, obtains the first fluctuation coefficient of the reconstructed data based on the difference values of each data point in the reconstructed data, obtains the fluctuation difference of the reconstructed data based on the difference of the second fluctuation coefficient between adjacent data points in the reconstructed data, obtains the time interval to be stabilized based on the second fluctuation coefficient of all data points in the reconstructed data, obtains the response coefficient of the Gaussian membership function based on the fluctuation difference of the reconstructed data and the time interval to be stabilized, obtains the adjustment coefficient of the Gaussian membership function in combination with the first fluctuation coefficient of the reconstructed data, and then obtains the variance value of the Gaussian membership function for temperature control during the geomembrane production process. This invention determines the temperature fluctuation in the current control process by constructing a fluctuation coefficient and performing fluctuation analysis. Combined with the response effect of the intelligent temperature control system, it achieves adaptive adjustment of the variance value of the Gaussian membership function, thereby making the temperature control during the production of antistatic and conductive geomembranes more robust and adaptable to more production environments. Attached Figure Description
[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0034] Figure 1 This is a flowchart of the production control method for an antistatic and conductive geomembrane according to the present invention.
[0035] Figure 2 A schematic diagram of IMF components;
[0036] Figure 3 A diagram illustrating the original and reconstructed data;
[0037] Figure 4 The temperature-time curve corresponding to a small variance value of the Gaussian membership function;
[0038] Figure 5 The temperature-time curve corresponding to a large variance value of the Gaussian membership function;
[0039] Figure 6 A schematic diagram of the Gaussian membership function of the P parameter in the fuzzy PID corresponding to the error;
[0040] Figure 7 This is the temperature-time curve corresponding to the production control method of an antistatic and conductive geomembrane according to the present invention. Detailed Implementation
[0041] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a production control method for antistatic and conductive geomembranes proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.
[0043] The following description, in conjunction with the accompanying drawings, details the specific scheme of the production control method for antistatic and conductive geomembranes provided by this invention.
[0044] Please see Figure 1 The diagram illustrates a flowchart of a production control method for an antistatic and conductive geomembrane according to an embodiment of the present invention. The method includes the following steps:
[0045] S001. Collect temperature data during the production process of antistatic and conductive geomembranes.
[0046] Temperature sensors are installed on the extruder during the production of antistatic and conductive geomembranes. The real-time temperature data at the extruder barrel is collected by the temperature sensors and used as the temperature data during the production of antistatic and conductive geomembranes.
[0047] Thus, temperature data were obtained during the production process of antistatic and conductive geomembranes.
[0048] S002. Based on the temperature data during the production process of antistatic and conductive geomembranes, obtain the fluctuation coefficient of the temperature data during the production process of antistatic and conductive geomembranes.
[0049] It should be noted that after obtaining the temperature data, it can be input into the temperature fuzzy control system for the antistatic and conductive geomembrane production process for temperature control. Currently, temperature fuzzy control systems typically employ fuzzy PID systems. In a fuzzy PID system, the error and rate of change of the current temperature data need to be calculated as inputs. Fuzzy functions and corresponding fuzzy rule sets are then constructed for the error and rate of change, respectively. The fuzzy values corresponding to the error and error variables are defuzzified to obtain the PID control coefficients, which are then used for temperature control during the antistatic and conductive geomembrane production process.
[0050] It should be further explained that currently, fixed fuzzy functions are often used when setting them. However, the temperature in the production process of antistatic and conductive geomembranes exhibits certain fluctuations, making the temperature control system overly sensitive even when using a Gaussian fuzzy function. Therefore, adaptive adjustment of the fuzzy function can reduce the sensitivity of the temperature control system to temperature changes, thereby improving its robustness. This embodiment performs fluctuation analysis on the temperature data during the production process of antistatic and conductive geomembranes to obtain fluctuation coefficients. These coefficients are then analyzed to obtain adjustment coefficients for the variance of the Gaussian membership function. Based on these adjustment coefficients, intelligent control of the Gaussian membership function in the temperature control system is achieved.
[0051] In this embodiment, a preset reference temperature T0 is used. This embodiment takes T0 = 180℃ as an example for description. The specific temperature is not limited and can be adjusted by the implementer according to the specific implementation scenario.
[0052] The difference between the current temperature and the reference temperature is obtained as the error value e(t), where t represents the current time. The error rate of change is obtained based on the sampling interval Δt of the temperature sensor. Where e(t) is the temperature error value at the latest moment, and e(t-1) is the temperature error value at the previous moment. It should be noted that the temperature sensor sampling interval Δt can be directly obtained from the temperature sensor device parameters.
[0053] Establish the initial fuzzy control rules and initial membership functions, specifically as follows:
[0054] This embodiment sets 7 fuzzy subsets, defined as {negative large, negative medium, negative small, zero, positive small, positive medium, positive large}. The universe of discourse contains 13 elements, ranging from -6 to 6. It should be noted that this embodiment only uses 7 fuzzy subsets and 13 elements in the universe of discourse as an example; the specific implementation is not limited. The number of fuzzy subsets is not limited, and the implementer can set the fuzzy subsets and universe of discourse according to the specific implementation. This embodiment selects a Gaussian membership function as the initial membership function, with a range of [0,1]. It should be noted that fuzzy subsets and the universe of discourse are well-known techniques in fuzzy PID control and will not be elaborated further here. The error e(t) and the error rate of change ec are the input quantities of the temperature fuzzy control system in the production process of antistatic and conductive geomembranes.
[0055] It should be noted that while the variance of historical error data can be obtained as the variance of the Gaussian membership function, the variance of historical error data may not be applicable to the current variance value due to changes in the working environment temperature. Therefore, it is necessary to optimize and adjust the variance of historical error data to adaptively adjust the Gaussian membership function variance. When adaptively adjusting the Gaussian membership function variance, it is necessary to first determine whether the current temperature value is stable. Since the temperature during feeding is not stable, and the temperature sensor carries some noise data when collecting temperature values, there will inevitably be some noise fluctuations during temperature data acquisition. Therefore, it is necessary to denoise the temperature data and then perform fluctuation analysis on the denoised temperature data.
[0056] In this embodiment, the n most recent temperature data points from the historical temperature data collected by the temperature sensor are used to form a data segment. This data segment is used as the basis for adaptive adjustment of the Gaussian membership function variance value. This embodiment uses n=300 as an example, but the specific value is not limited. In other embodiments, the implementer can set the value according to the specific implementation situation. It should be noted that although the input quantity in the intelligent temperature control system is the temperature data at a single moment, this embodiment uses a data segment for data fluctuation analysis, rather than analyzing the temperature data at a single moment, because it is difficult to measure the specific degree of noise influence on the temperature data at a single moment.
[0057] The data segment was decomposed using the EMD algorithm to obtain multiple IMF components and a residual. Figure 2 This is a schematic diagram of the IMF components obtained by decomposing the data segment. The number of peak points in each IMF component is counted; the more peak points an IMF component has, the more noise it contains. A line graph of the peak point count is constructed with the IMF component number as the horizontal axis and the number of peak points corresponding to each IMF component as the vertical axis.
[0058] Connect the first and last data points in the peak count line graph to form a straight line, which serves as the fitted line. Substitute the index of each IMF component into the equation of the fitted line to obtain the fitted value of the peak count for each IMF component. Calculate the absolute value of the difference between the actual peak count and the fitted value for each IMF component, and take the data point in the peak count line graph corresponding to the IMF component with the largest absolute difference as the inflection point.
[0059] Remove the IMF components corresponding to all data points to the left of the inflection point, and sum the residuals and the remaining IMF components after removal to complete the data reconstruction. The reconstructed data is the result of denoising the data segment. Figure 3 This is a diagram illustrating the original data (i.e., data segments) and the reconstructed data.
[0060] It should be noted that if fluctuations still exist in the denoised data segment, it indicates that the variance of the Gaussian membership function in the current fuzzy function is too small. Therefore, it is necessary to adjust the variance of the Gaussian membership function to reduce response sensitivity, thereby stabilizing temperature control and ultimately ensuring stable production quality of antistatic and conductive geomembranes. Figure 4 The temperature-time curve corresponds to a small variance in the Gaussian membership function, indicating significant temperature data variation and a sensitive response. If the denoised data segment shows little fluctuation, it suggests either optimal or insufficient temperature control response. The fluctuation coefficient alone is insufficient for effective adjustment of the Gaussian membership function variance; it's necessary to analyze the control system's response speed. A larger Gaussian membership function variance results in smoother temperature data after intelligent control system regulation, but a longer response time. Failure to adjust temperature promptly may lead to defects in antistatic and conductive geomembrane production, for example... Figure 5 This is the temperature data curve corresponding to a large variance of the Gaussian membership function. The temperature data is relatively smooth, but the response time is long.
[0061] In this embodiment, the absolute value of the difference between the temperature value corresponding to each data point in the reconstructed data and the reference temperature T0 is obtained and denoted as the difference value of each data point in the reconstructed data. A difference curve is constructed with the time corresponding to each data point in the reconstructed data as the horizontal axis and the difference value of each data point in the reconstructed data as the vertical axis. The difference curve represents the difference between each temperature data point in the denoised data segment and the reference temperature T0.
[0062] It should be noted that a larger fluctuation in the difference curve indicates a larger temperature fluctuation. In this case, the variance of the Gaussian membership function should be increased; otherwise, the temperature value will be unstable, leading to instability in the intelligent temperature control system, unstable production quality of antistatic and conductive geomembranes, and affecting product yield. Conversely, a smaller fluctuation in the difference curve, with the temperature data more closely resembling a straight line, indicates that the current Gaussian membership function effectively addresses the current temperature fluctuation, ensuring the temperature operation of the intelligent temperature control system and stable production of antistatic and conductive geomembranes. In this case, the variance of the Gaussian membership function does not need adjustment. However, it is also possible that an excessively large variance of the Gaussian membership function fails to respond promptly to temperature changes, resulting in a seemingly stable difference curve that is actually detrimental to timely temperature control. In this case, the actual production quality of antistatic and conductive geomembranes is unstable. Therefore, it is necessary to analyze the response speed by combining the fluctuation coefficient of the difference curve to obtain the adjustment coefficient corresponding to the variance of the Gaussian membership function.
[0063] In this embodiment, the difference value corresponding to each data point in the difference curve is obtained. If all difference values are basically consistent and relatively small, the fluctuation degree corresponding to the difference curve is low. The information entropy of the difference values corresponding to all data points in the difference curve is obtained. Information entropy is a well-known technique and will not be described in detail in this embodiment.
[0064] The first fluctuation coefficient of the reconstructed data is obtained based on the mean of the difference values corresponding to all data points in the difference curve and the information entropy:
[0065] C = P × B;
[0066] Where C is the first fluctuation coefficient of the reconstructed data, P is the information entropy of the difference values corresponding to all data points in the difference curve, and B is the mean of the difference values corresponding to all data points in the difference curve. The closer the difference values corresponding to each data point in the difference curve are, the smaller the information entropy P. The smaller the difference between the temperature value in the reconstructed data and the reference temperature T0, the smaller the mean B of the difference values corresponding to all data points in the difference curve. When both the information entropy P and the mean B are small, the first fluctuation coefficient C is small. In this case, the temperature value in the reconstructed data is more stable, the temperature control effect of the intelligent temperature control system is better, and the variance of the Gaussian membership function does not need adjustment. Conversely, the larger the first fluctuation coefficient C is, the worse the temperature control effect of the intelligent temperature control system, and the larger the variance of the Gaussian membership function should be.
[0067] The first fluctuation coefficient of the reconstructed data is used as the variance value of the initial Gaussian membership function.
[0068] Thus, the first fluctuation coefficient was obtained.
[0069] S003. Based on the fluctuation coefficient, the fuzzy function in the temperature control system of the antistatic and conductive geomembrane during extrusion molding is adaptively adjusted.
[0070] It should be noted that after obtaining the fluctuation coefficient corresponding to the denoised data segment (i.e., reconstructed data), when the fluctuation coefficient is large, the variance value of the Gaussian membership function can be directly adjusted to a larger value to stabilize the production quality of antistatic and conductive geomembranes. However, when the fluctuation coefficient is small, it is difficult to determine whether the variance value of the Gaussian membership function is optimal or should be adjusted to a smaller value. In this case, the stability of the production quality of antistatic and conductive geomembranes cannot be judged. When the fluctuation coefficient is small, it means that the variance value of the current Gaussian membership function is large, the data is relatively smooth but the response is slow. At this time, the scope of temperature data analysis can be expanded to analyze the n temperature data points before each data point in the denoised data segment, obtain the fluctuation coefficient corresponding to each data point in the denoised data segment, and obtain the response coefficient of the current Gaussian membership function based on the change of the fluctuation coefficient.
[0071] In this embodiment, the n most recent temperature data points preceding each data point in the reconstructed data are obtained from the historical temperature data, and these are used as the data segment corresponding to each data point in the reconstructed data. The first fluctuation coefficient of the data segment corresponding to each data point in the reconstructed data is obtained using the method in step S002, and this is used as the second fluctuation coefficient of each data point in the reconstructed data.
[0072] Obtain the absolute value of the difference between the second fluctuation coefficients of all adjacent data points in the reconstructed data, and take the mean of all the absolute values of the differences as the fluctuation difference of the reconstructed data.
[0073] The second fluctuation coefficient of all data points in the reconstructed data is fitted using a polynomial method with the least squares approach to obtain the fitting function. For each data point in the reconstructed data, the corresponding time is sequentially substituted into the fitting function to obtain the predicted fluctuation coefficient for each data point. The data point corresponding to the predicted fluctuation coefficient of 0 closest to the current time is selected as the stable data point; the time difference between the start time and the time corresponding to the stable data point is taken as the time interval to be stabilized. The start time refers to the moment when the temperature begins to be controlled using the temperature control system during the extrusion molding of the antistatic and conductive geomembrane.
[0074] Based on the fluctuations in the reconstructed data and the time interval to be stabilized, the response coefficients of the Gaussian membership function are obtained:
[0075] D = E × F;
[0076] Where D is the response coefficient of the Gaussian membership function, E is the fluctuation difference of the reconstructed data. The greater the fluctuation difference, the greater the difference in the second fluctuation coefficient between adjacent data points, indicating that the temperature change is not smooth and the temperature is unstable during the production of antistatic and conductive geomembranes. F is the stabilization time interval. The longer the stabilization time interval, the longer it takes for the temperature data to reach the reference temperature so that the second fluctuation coefficient becomes 0.
[0077] It should be noted that a larger response coefficient of the Gaussian membership function indicates a slower response. In this case, the first fluctuation coefficient of the reconstructed data may be too small, and the variance of the Gaussian membership function needs to be reduced to ensure stable production quality of the antistatic and conductive geomembrane. Conversely, a smaller response coefficient of the Gaussian membership function indicates a faster response. In this case, the first fluctuation coefficient of the reconstructed data may be too large, and the variance of the Gaussian membership function needs to be increased to ensure stable production quality of the antistatic and conductive geomembrane.
[0078] In this embodiment, the adjustment coefficient of the Gaussian membership function is obtained based on the first fluctuation coefficient of the reconstructed data and the response coefficient of the Gaussian membership function:
[0079]
[0080] Where G represents the adjustment coefficient of the Gaussian membership function, C is the first fluctuation coefficient of the reconstructed data, D is the response coefficient of the Gaussian membership function, and h and k are hyperparameters. This embodiment uses h=1 and k=0.2 as an example, but the specific values are not limited. Implementers can set the hyperparameters according to the actual implementation situation. To avoid the first fluctuation coefficient being 0, resulting in an adjusted variance value of 0, when adjusting the variance value of the Gaussian membership function, one is added to the numerator to obtain the adjustment coefficient of the Gaussian membership function. When the first fluctuation coefficient is larger, the response coefficient of the Gaussian membership function is smaller, and the temperature data changes are less smooth. In this case, the variance value of the Gaussian membership function should be larger to ensure stable production quality of antistatic and conductive geomembranes. When the first fluctuation coefficient is smaller, the response coefficient of the Gaussian membership function is larger, and the temperature data changes more smoothly. In this case, the variance value of the Gaussian membership function should be smaller to ensure stable production quality of antistatic and conductive geomembranes.
[0081] By multiplying the adjustment coefficient G of the Gaussian membership function by the variance of the Gaussian membership function, and using the result as the variance of the new Gaussian membership function, the fuzzy function adaptive adjustment in the temperature control system of the antistatic and conductive geomembrane during extrusion molding is realized.
[0082] S004. Based on the adjustment results of the fuzzy function in the temperature control system during the extrusion molding of the antistatic and conductive geomembrane, the production control of the antistatic and conductive geomembrane is carried out.
[0083] After obtaining the variance value of the new Gaussian membership function, obtain the fuzzy functions of the errors and error change rates corresponding to the three parameters P, I, and D in the fuzzy PID, for example... Figure 6 This diagram illustrates the Gaussian membership function of the P parameter in a fuzzy PID controller, corresponding to the error. Taking the current error e(t) and the rate of change of error ec as input, the membership degree of each parameter to the corresponding fuzzy subset is obtained from the Gaussian membership function. Fuzzy inference is performed according to fuzzy rules to obtain the fuzzy subset. Then, the fuzzy subset is defuzzified to obtain the corresponding P, I, and D parameters. The production temperature control of antistatic and conductive geomembranes is achieved through the P, I, and D parameters in the fuzzy PID controller. Figure 7 This embodiment shows the temperature-time curve obtained by adaptively adjusting the variance of the Gaussian membership function. It should be noted that the membership degree of each parameter corresponding to the fuzzy subset is obtained from the Gaussian membership function. Fuzzy inference is then performed according to fuzzy rules to obtain the fuzzy subset. The fuzzy subset is then defuzzified to obtain the corresponding P, I, and D parameters. These are all well-known techniques in fuzzy PID control and will not be elaborated upon here.
[0084] Through the above steps, the production control of antistatic and conductive geomembranes was achieved.
[0085] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A production control method for an antistatic and conductive geomembrane, characterized in that, The method includes the following steps: Temperature data is collected during the geomembrane production process; data segments are obtained based on the temperature data, and noise is removed from the data segments to obtain reconstructed data; the absolute value of the difference between the temperature value corresponding to each data point in the reconstructed data and the preset reference temperature is obtained, which is recorded as the difference value of each data point in the reconstructed data; the information entropy of the difference values corresponding to all data points in the reconstructed data is obtained; the mean of the difference values corresponding to all data points in the reconstructed data is obtained; and the product of the mean of the difference values corresponding to all data points in the reconstructed data and the information entropy is used as the first fluctuation coefficient of the reconstructed data. Obtain the data segment corresponding to each data point in the reconstructed data, and obtain the first fluctuation coefficient of the data segment corresponding to each data point as the second fluctuation coefficient of each data point in the reconstructed data. Obtain the absolute value of the difference between the second fluctuation coefficients of all two adjacent data points in the reconstructed data, and obtain the mean of all the absolute values of the differences as the fluctuation difference of the reconstructed data. Use the least squares method to perform polynomial fitting on the second fluctuation coefficients of all data points in the reconstructed data to obtain the fitting function. For each data point in the reconstructed data, substitute the time corresponding to each data point into the fitting function in turn to obtain the predicted fluctuation coefficient corresponding to each data point. Obtain the data point corresponding to the nearest predicted fluctuation coefficient of 0 to the current time as the stable data point. Use the time difference from the start time to the time corresponding to the stable data point as the time interval to be stabilized. Based on the fluctuations in the reconstructed data and the time interval to be stabilized, the response coefficients of the Gaussian membership function are obtained: Where D is the response coefficient of the Gaussian membership function, E is the fluctuation difference of the reconstructed data, and F is the time interval to be stabilized. The adjustment coefficient of the Gaussian membership function is obtained based on the first fluctuation coefficient of the reconstructed data and the response coefficient of the Gaussian membership function: ;in, This represents the adjustment coefficient of the Gaussian membership function. The first fluctuation coefficient for reconstructing the data is given by D, which is the response coefficient of the Gaussian membership function, and h and k are hyperparameters. The variance of the Gaussian membership function is obtained based on the adjustment coefficient of the Gaussian membership function, and the temperature is controlled during the geomembrane production process based on the variance of the Gaussian membership function.
2. The production control method for an antistatic and conductive geomembrane according to claim 1, characterized in that, The specific steps involved in denoising the data segment and obtaining the reconstructed data are as follows: The data segment is decomposed using the EMD algorithm to obtain multiple IMF components and a residual; the number of peak points in each IMF component is counted, and a line chart of the number of peak points is constructed with the IMF component number as the horizontal axis and the number of peak points corresponding to each IMF component as the vertical axis. Connect the first and last data points in the peak point count line graph to form a straight line, which serves as the fitting line. Substitute the number of each IMF component into the equation of the fitting line to obtain the fitted value of the peak point count for each IMF component. Calculate the absolute value of the difference between the actual peak point count and the fitted value of the peak point count for each IMF component. Take the data point in the peak point count line graph corresponding to the IMF component with the largest absolute value of the difference as the inflection point. Remove the IMF components corresponding to all data points to the left of the inflection point, and sum the residuals and the remaining IMF components after removal to complete the data reconstruction and obtain the reconstructed data.
3. The production control method for an antistatic and conductive geomembrane according to claim 1, characterized in that, The specific steps for obtaining the variance value of the Gaussian membership function based on the adjustment coefficient of the Gaussian membership function, and for controlling the temperature during the geomembrane production process based on the variance value of the Gaussian membership function, are as follows: The variance of the Gaussian membership function is obtained by calculating the adjustment coefficient of the Gaussian membership function, thus obtaining the Gaussian membership function. The error value and error rate of change are obtained based on the latest temperature data. The error value and error rate of change are used as inputs to the fuzzy PID controller. The fuzzy PID controller is then used to control the production temperature of antistatic and conductive geomembranes based on the Gaussian membership function.