Production line monitoring method and device based on industrial internet and storage medium

Abstract

The invention provides a production line monitoring method and device based on the industrial internet and a storage medium, and relates to the technical field of electronic ceramic, and the method comprises the steps: firstly, in the aspect of purity evaluation, constructing a purity evaluation model, setting key parameters of zinc oxide purity, and determining that the zinc oxide purity is unqualified if the key parameters are lower than the zinc oxide purity; and in the aspect of torque evaluation, constructing a torque evaluation model, setting key parameters of a torque standard deviation, and judging that the stirring link is unqualified if the key parameters exceed the torque standard deviation. In the aspect of dimension evaluation, a dimension evaluation model is constructed, key parameters of upper and lower mold closing displacement and horizontal displacement are set respectively, and if the key parameters are exceeded, the forming link is judged to be unqualified; in the aspect of temperature evaluation, a temperature evaluation model is constructed, key parameters are set in each stage of sintering, if the key parameters are exceeded, the sintering link is judged to be unqualified, and finally, a final quality scoring formula is constructed according to each model to evaluate the quality of the zinc oxide resistor disc.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic ceramics, and particularly to a production line monitoring method, device and storage medium based on industrial Internet. Background Art

[0002] In electronic devices and power systems, zinc oxide varistors, as key overvoltage protection components, play a crucial role. With the rapid development of electronic technology, the requirements for the reliability and stability of overvoltage protection in various electrical devices are increasing day by day. The technical field of electronic ceramics is committed to the research and development and production of excellent electronic ceramic components, and zinc oxide varistors are typical representatives among them. With its unique non-linear volt-ampere characteristics, it can present a high-resistance state under normal voltage to ensure the normal operation of the circuit; when encountering overvoltage, the resistance rapidly decreases, introducing the overvoltage current into the ground, thereby effectively protecting the backend equipment from damage.

[0003] However, the current production process of zinc oxide varistors faces many challenges. In terms of raw material control, problems such as fluctuations in zinc oxide purity and uneven mixing of additives frequently occur. Due to the lack of accurate and efficient real-time purity monitoring means, in production, the performance of the final product is often relied on to judge whether the raw materials are qualified through final experimental testing. Once problems are found, a large amount of human, material and time costs invested in the early stage have been wasted. For example, if the purity of zinc oxide does not meet the standard, impurities will interfere with the formation of the internal crystal structure, resulting in a deterioration of the non-linear characteristics of the varistor, but it is often only discovered during the finished product inspection, causing a large amount of raw material waste.

[0004] During the stirring process, it is difficult to ensure torque stability. Although the existing technology can monitor torque, there is no clear standard for setting key parameters, and a large number of experiments are often required to determine the appropriate range, and it is impossible to accurately judge the product quality based on torque fluctuations in a timely manner. If the stirring is uneven, the distribution of the material components is inconsistent, and the performance discreteness of the varistors after sintering is large. Many products need to be reworked or scrapped due to unqualified performance, increasing production costs.

[0005] Controlling dimensional accuracy is also difficult. During the die closing process, due to insufficient displacement monitoring accuracy and unreasonable key parameter settings, it is difficult to ensure the accuracy of the thickness and side length of the varistor. Dimensional deviations of the varistor will affect its compatibility with other components, and dimensional problems are only discovered in the subsequent assembly process, resulting in delays in the overall production process and increased additional costs.

[0006] There are also defects in temperature control during the sintering process. At present, the temperature parameters are mainly adjusted based on the performance experiments of the final product, and the time control at each stage depends on experience, lacking accurate key time range settings. Improper heating, holding or cooling times will cause abnormal internal structures of the materials, such as stress concentration caused by too fast heating and insufficient reaction due to insufficient holding, seriously affecting the product performance, and these problems can only be exposed during the finished product inspection, resulting in a large amount of resource waste.

[0007] In summary, in the existing production process of zinc oxide varistors, quality monitoring overly relies on final experimental judgment and cannot determine whether the product is qualified in a timely manner based on key parameters during the production process, resulting in waste of resources and increased costs. There is a need for a quality monitoring method that can monitor in real time and accurately judge during the production process, reducing the reliance on final experiments, so as to improve production efficiency, reduce costs, and ensure product quality.

[0008] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0009] The purpose of the present invention is to provide a production line monitoring method, device, and storage medium based on the industrial Internet to solve the problems raised in the above background art.

[0010] To achieve the above purpose, the present invention provides the following technical solutions:

[0011] A production line monitoring method based on the industrial Internet, the specific steps include:

[0012] Further, use an XRF spectrometer to measure the fluorescence spectrum of the raw material, obtain the light intensity data at different energies, and through the element counting rate function of the XRF spectrometer, record the counts at Kα (8.63 keV) and Kβ (9.57 keV) as the characteristic peak intensity I Zn of Zn, and at the same time obtain the total characteristic peak intensity I total and the background signal intensity I bkg of the raw material in the production line, and calculate the zinc oxide purity of the raw material in the production line:

[0013]

[0014] In the formula, C Zno represents the zinc oxide purity of the raw material in the production line.

[0015] Further, set the standard zinc oxide purity as C zb , when the zinc oxide purity C Zno < C zb , the purity evaluation model is:

[0016]

[0017] In the formula, S Zno represents the purity evaluation model, and C zb represents the standard zinc oxide purity;

[0018] When the zinc oxide purity C Zno≥C zb When it is, set the output result of the purity evaluation model to 1.

[0019] Furthermore, set the time window to T and the time interval to t c , the torque standard deviation threshold is s max , the number of fluctuations threshold is f max , where T > t c > 0, at a fixed time interval t c , collect the torque data x during the stirring process i , where i = 1, 2,..., n, and n represents the total number of torque data collected within a certain time window. Calculate the average value based on the torque data and the total number of torque data collected within the time window:

[0020]

[0021] In the formula, represents the average value of the torque data within the time window;

[0022] For the j-th time window, its torque standard deviation calculation formula is:

[0023]

[0024] In the formula, s j represents the torque standard deviation within the j-th time window, where j = 1, 2,..., z, and z is the total number of time windows.

[0025] Furthermore, count the number of times that the torque standard deviation s j exceeds the torque standard deviation threshold s max , record the statistical result as the number of fluctuations m, and construct a torque evaluation model based on the number of fluctuations and the number of fluctuations threshold:

[0026]

[0027] In the formula, S force represents the torque evaluation model.

[0028] Furthermore, set the standard thickness L std vert and the standard side length L std hor , during the process of the mold starting to close until it is completely closed, record the real-time change value of the relative position of the upper and lower molds, denoted as the upper and lower mold closing displacement data L vert ; taking the mold center as the reference point, measure its position change in the horizontal circumferential direction to obtain the horizontal direction displacement data L hor , normalize the error of the upper and lower mold closing displacement data:

[0029]

[0030] In the formula, S vert represents the normalized value of the upper and lower die-clamping displacement data error;

[0031] Normalize the displacement data error in the horizontal direction:

[0032]

[0033] In the formula, S hor represents the normalized value of the displacement data error in the horizontal direction;

[0034] Construct a dimensional evaluation model:

[0035] S dimension = α·S vert + β·S hor

[0036] In the formula, S dimension represents the dimensional evaluation model, α and β represent weight coefficients, which are adjusted according to specific process requirements, where α + β = 1.

[0037] Furthermore, during the sintering process, set the standard heating time as T std heat , and record the actual heating duration as T heat ;

[0038] Set the standard heat preservation time as T std hold , and record the actual heat preservation duration as T hold ;

[0039] Set the standard cooling time as T std cool , and record the actual cooling duration as T cool ;

[0040] Evaluate the heating stage:

[0041]

[0042] In the formula, S heat represents the output result of the heating stage evaluation;

[0043] Evaluate the heat preservation stage:

[0044]

[0045] In the formula, S hold represents the output result of the heat preservation stage evaluation;

[0046] Evaluate the cooling stage:

[0047]

[0048] Wherein, S cool represents the output result evaluated in the cooling stage;

[0049] Construct a temperature evaluation model:

[0050] S temp =γ·S heat +δ·S hold +ε·S cool

[0051] Wherein, S temp represents the temperature evaluation model, γ, δ, ε represent weight coefficients, which are adjusted according to specific process requirements, and among them, γ + δ + ε = 1.

[0052] Furthermore, set the key parameter of the zinc oxide purity of the raw materials in the production line as C′ Zno , the key parameter of the torque standard deviation as S′ X ; the key parameters of the upper and lower die closing displacement are {L minvert , L maxvert}}, the key parameters of the horizontal displacement are {L minvert , L maxvert}}; the key parameters of the heating time are {T minheat , T maxheat}}, the key parameters of the heat preservation time are {T minhold , T maxhold}}, the key parameters of the cooling time are {T mincool , T maxcool};

[0053] When the zinc oxide purity C Zno >C′ Zno , alarm that the zinc oxide purity is significantly abnormal; when the torque standard deviation s j >S′ X , alarm that the torque fluctuation is significantly abnormal; when the upper and lower die closing displacement , alarm that the thickness is significantly abnormal; when the horizontal displacement , alarm that the side length is significantly abnormal; when the heating time , alarm that the heating duration is significantly abnormal; when the heat preservation time Alarm that the heat preservation time is significantly abnormal; when the cooling time , alarm that the cooling time is significantly abnormal;

[0054] Calculate the final quality score according to the purity evaluation model, torque evaluation model, size evaluation model and temperature evaluation model:

[0055] Q = ω1·S Zno +ω2·S force +ω3·Sdimension +ω4·S temp

[0056] Wherein, Q represents the final quality score, and ω1 + ω2 + ω3 + ω4 = 1;

[0057] Set the final quality score thresholds Q1 and Q2, where 0 < Q1 < Q2. When Q ≥ Q2, the quality of the product is evaluated as high. When Q2 > Q ≥ Q1, the quality of the product is evaluated as qualified. When Q < Q1, the quality of the product is evaluated as unqualified.

[0058] The present invention further provides a production line monitoring device based on the industrial Internet. The device is used to execute the above-mentioned production line monitoring method based on the industrial Internet, and includes:

[0059] A purity evaluation module, which is used to preset the standard purity of zinc oxide, monitor the zinc oxide purity of the raw materials in the production line, construct a purity evaluation model based on the zinc oxide purity of the production line raw materials and the standard purity of zinc oxide, and record the raw material batch and supplier information;

[0060] A torque fluctuation evaluation module, which is used to set a time window, a time interval, a torque standard deviation threshold, and a fluctuation times threshold. Within each time window, torque data during the stirring process is collected at a fixed time interval, the torque standard deviation within the time window is calculated, and within all time windows, the number of times the torque standard deviation exceeds the torque standard deviation threshold is defined as the number of fluctuations, and a torque evaluation model is constructed based on the number of fluctuations and the fluctuation times threshold;

[0061] A size forming evaluation module, which is used to set a standard thickness and a standard side length. During the process of the mold closing from the start to the fully closed state, the change value of the relative position of the upper and lower molds is recorded as the upper and lower mold closing displacement data. Taking the mold center as the reference point, the position change in the horizontal direction is measured and recorded as the horizontal direction displacement data, and a size evaluation model is constructed based on the standard thickness, the standard side length, the upper and lower mold closing displacement, and the horizontal direction displacement data;

[0062] A sintering temperature evaluation module, which is used to divide the sintering process into a heating stage, an insulation stage, and a cooling stage according to the temperature change situation, record the duration of each stage, set the standard time range for each stage, and construct a temperature evaluation model based on the duration of each stage and the standard time;

[0063] A comprehensive quality score module, which is used to set key parameters, calculate the final quality score by using the weighted summation method, and comprehensively evaluate the product based on the key parameters and the final quality score. The key parameters include zinc oxide purity, torque standard deviation, upper and lower mold closing displacement, horizontal direction displacement, heating time, insulation time, and cooling time.

[0064] The present invention further provides a storage medium storing a computer program, which, when executed by a processor, implements the above-described production line monitoring method based on the industrial Internet.

[0065] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0066] By extracting multi-dimensional data such as the purity, torque, size, and temperature of zinc oxide raw materials, constructing models, and setting key parameter ranges in terms of zinc oxide purity, torque fluctuation, die displacement, and sintering time, the present invention realizes real-time monitoring and evaluation during the production process. When a certain parameter exceeds the set range, an alarm is immediately issued, facilitating timely adjustment by production personnel, preventing the batch appearance of defective products, and reducing the generation of non-conforming products. The present invention also constructs a final quality scoring formula by using the method of weighted summation through the constructed models to realize the quality determination of products and reduce the dependence on final experiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 is a schematic diagram of the overall method flow of the present invention;

[0068] Figure 2 is a schematic diagram of the overall system module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0069] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to specific embodiments.

[0070] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meanings understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second", and similar terms used in the present invention do not denote any order, quantity, or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left", "right", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0071] Embodiment:

[0072] Please refer to Figure 1 , the present invention provides the following technical solution:

[0073] A production line monitoring method based on the industrial Internet, the specific steps including:

[0074] Step 1: Preset the standard purity of zinc oxide, monitor the purity of zinc oxide in the raw materials on the production line, construct a purity evaluation model based on the purity of zinc oxide in the raw materials of the production line and the standard purity of zinc oxide, and record the raw material batch and supplier information;

[0075] Based on industry experience and product performance requirements, the standard purity of zinc oxide is usually 99.5%-99.9% in the production of zinc oxide varistors. When the purity of zinc oxide is lower than 99.5%, impurities may significantly affect the electrical properties of the varistors, such as causing unstable resistance values, poor non-linear coefficients, etc., resulting in the inability to withstand high voltage impacts. 99.7% is a relatively common standard purity. Therefore, the standard purity of zinc oxide is preset as C zb = 99.7%, or it can be set by yourself according to the requirements of product performance;

[0076] When monitoring the purity of zinc oxide in the raw materials on the production line, collect some raw materials from the production line to make uniform and flat thin slices to ensure the accuracy and stability of the measurement. Use an XRF spectrometer to measure the fluorescence spectrum of the raw materials. When the Zn element in the raw materials is excited, it will release fluorescent X-rays with specific energies. The main characteristic peaks are Kα (8.63 keV) and Kβ (9.57 keV). Therefore, the light intensity data at different energies can be obtained through the XRF spectrometer. Through the element counting rate function of the XRF spectrometer, the counts located at Kα (8.63 keV) and Kβ (9.57 keV) are recorded as the characteristic peak intensity I of Zn Zn , and at the same time, obtain the total characteristic peak intensity I through the XRF spectrometer total and the background signal intensity I bkg , and calculate the purity of zinc oxide in the raw materials on the production line:

[0077]

[0078] In the formula, C Zno represents the purity of zinc oxide in the raw materials on the production line. Subtracting the background intensity I bkg is to eliminate the influence of stray radiation generated by the instrument itself and environmental factors on the measurement results, so that the measured intensity value can more accurately reflect the true signal of the element in the sample. This formula ensures that the calculated purity of zinc oxide is between 0 and 1. The closer the value is to 1, the higher the purity of zinc oxide.

[0079] When the purity of zinc oxide C Zno < C zb , the purity evaluation model is:

[0080]

[0081] In the formula, SZno It is expressed as a purity evaluation model, and an evaluation value related to purity is constructed. The closer the result is to 1, the closer the purity of the raw material is to the standard purity of zinc oxide, and the better the quality. On the contrary, the smaller the result, the greater the gap between the purity of the raw material and the standard purity of zinc oxide, and the worse the quality.

[0082] When the purity of zinc oxide is C Zno ≥C zb , the output result of the purity evaluation model is set to 1, indicating that the purity of the raw material meets or exceeds the standard purity of zinc oxide. Record the raw material batch and supplier information, and screen high-quality suppliers based on the evaluation results. When there are quality problems, the root cause of the problem can be determined.

[0083] Step 2: Set the time window, time interval, torque standard deviation threshold, and number of fluctuations threshold. Within each time window, collect the torque data of the stirring process at a fixed time interval, calculate the torque standard deviation within the time window, and define the number of times the torque standard deviation exceeds the torque standard deviation threshold as the number of fluctuations within all time windows. Based on the number of fluctuations and the number of fluctuations threshold, construct a torque evaluation model;

[0084] Install a high-precision torque sensor on the stirring equipment of the production line to ensure that the torque changes during the stirring process can be accurately captured. Set the time window as T, the time interval as t c , the torque standard deviation threshold as s max , the number of fluctuations threshold as f max , where T > t c > 0, collect the torque data x c during the stirring process at a fixed time interval t i , where i = 1, 2,..., n, and n represents the total number of torque data collected within a certain time window;

[0085] The torque data will fluctuate due to various factors, such as changes in material properties and wear of the stirring paddle. The average value can represent the central tendency of the torque data within this time window;

[0086] Calculate the average value based on the torque data and the total number of torque data collected within the time window:

[0087]

[0088] In the formula, represents the average value of the torque data within the time window;

[0089] In the evaluation of the stability of the stirring process, the standard deviation of torque can intuitively reflect the torque fluctuations during the stirring process. A smaller standard deviation of torque indicates that the data is relatively concentrated around the average value, and the stirring process is relatively stable; on the contrary, a larger standard deviation indicates that the torque fluctuates violently, and there may be equipment failures or uneven material states. Therefore, for the j-th time window, the formula for calculating the standard deviation of torque is as follows:

[0090]

[0091] In the formula, s j represents the standard deviation of torque within the j-th time window, where j = 1, 2, …, z, and z is the total number of time windows;

[0092] For the value of the torque standard deviation threshold s max , it can generally be set to 1.5 - 2 times the standard deviation of normal data, or set according to industry experience. Count the number of times the standard deviation of torque s j exceeds the torque standard deviation threshold s max . Denote the statistical result as the number of fluctuations m, and construct a torque evaluation model based on the number of fluctuations and the number of fluctuations threshold:

[0093]

[0094] In the formula, S force represents the torque evaluation model;

[0095] For example, in the monitoring of the stirring process during the production of zinc oxide varistor chips, set the time window T = 15 minutes, which is converted to 900 seconds, and set the time interval t = 3 seconds. Enough data points can be collected within the time window to accurately reflect the torque changes. The total number of torque data points collected within this time window pieces. Through the statistical analysis of a large number of past normal production data, the standard deviation of normal data is about 0.8. Set the torque standard deviation threshold s max = 1.5 × 0.8 = 1.2 according to 1.5 times the standard deviation of normal data. Set the number of fluctuations threshold f max , combined with the requirements of this production process for the stirring stability and past experience, set it to 3 times. A total of z = 10 time windows are monitored, and the number of times the standard deviation of torque s j exceeds the torque standard deviation threshold s max is m = 5 times, and the torque evaluation model is obtained

[0096] Step 3: Set the standard thickness and standard side length. During the process of the mold closing from the start to full closure, record the change value of the relative position of the upper and lower molds as the upper and lower mold closing displacement data. Taking the mold center as the reference point, measure its position change in the horizontal direction and record it as the horizontal direction displacement data. Based on the standard thickness, standard side length, upper and lower mold closing displacement, and horizontal direction displacement data, construct a dimension evaluation model;

[0097] The thickness and side length directly determine the external dimension accuracy of the resistor chip and affect the compatibility of the product with other components. If the dimension deviation is too large, it may be difficult to accurately match with other components in the circuit during the subsequent assembly process. Set the standard thickness L std vert and the standard side length L std hor . During the process of the mold starting to close to full closure, by selecting a high-precision linear displacement sensor, record the real-time change value of the relative position of the upper and lower molds, denoted as the upper and lower mold closing displacement data L vert ; taking the mold center as the reference point, measure its position change in the horizontal circumferential direction through an angle sensor, and obtain the horizontal direction displacement data L hor after calculation. Normalize the error of the upper and lower mold closing displacement data:

[0098]

[0099] In the formula, S vert represents the normalized value of the error of the upper and lower mold closing displacement data. The larger the calculation result, the larger the error of the thickness;

[0100] Normalize the error of the horizontal direction displacement data:

[0101]

[0102] In the formula, S hor represents the normalized value of the error of the horizontal direction displacement data. The larger the calculation result, the larger the error of the thickness. The normalization process can unify data with different dimensions and value ranges into the [0, 1] interval, eliminate the interference caused by differences in the original data units or value ranges, and make the subsequent evaluation more scientific and accurate;

[0103] Construct a dimension evaluation model:

[0104] S dimension =α·S vert +β·S hor

[0105] In the formula, S dimensionDenoted as the dimensional evaluation model, α and β are denoted as weight coefficients. If higher dimensional accuracy requirements are imposed on a certain direction in the production process, a greater weight can be given to that direction, which is adjusted according to specific process requirements. Among them, α + β = 1, α > 0, β > 0. If the thickness deviation is too large, it may cause difficulties in the subsequent assembly of the resistor chip to accurately match other components in the circuit, affecting the performance of the entire circuit. Therefore, α > 0 can reflect the actual impact of the vertical die closing displacement on dimensional accuracy. The horizontal displacement data reflects the position change of the die horizontally and determines the side length accuracy of the resistor chip. The accuracy of the side length is also related to the matching degree of the resistor chip with other components. If the side length deviation exceeds the allowable range, it will also cause difficulties in subsequent assembly and use. Therefore, β > 0 indicates that the horizontal displacement has a non-negligible effect on dimensional accuracy, S dimension The closer the value of S is to 1, the closer the displacements of the die in the vertical die closing and horizontal directions are to the standard values. Conversely, the more deviated from 1, the greater the dimensional deviation, and the product quality may be affected;

[0106] Step 4: According to the temperature change situation, divide the sintering process into heating, holding, and cooling stages, record the duration of each stage, set the standard time range for each stage, and construct a temperature evaluation model based on the duration and standard of each stage;

[0107] When the raw materials are sintered, there are specific requirements for the time of the heating, holding, and cooling stages. For example, too fast heating may cause internal stress concentration and lead to material cracking; insufficient holding time cannot enable the material to fully complete physical and chemical reactions, affecting its final performance. Therefore, during the sintering process, recording the actual heating, holding, and cooling durations can monitor in real time whether the production process is carried out according to the standard process. The deviation between the actual time and the standard time implies problems such as equipment failures and human operation errors;

[0108] Set the standard heating time as T std heat , and record the actual heating duration as T heat ;

[0109] Set the standard holding time as T std hold , and record the actual holding duration as T hold ;

[0110] Set the standard cooling time as T std cool , and record the actual cooling duration as T cool ;

[0111] Evaluate the heating stage:

[0112]

[0113] In the formula, S heatIndicates the output result of the heating stage evaluation. The closer the value is to 1, the closer the actual heating duration T heat is to the heating standard time T std heat ;

[0114] Evaluate the heat preservation stage:

[0115]

[0116] In the formula, S hold Indicates the output result of the heat preservation stage evaluation. The closer the value is to 1, the closer the actual heat preservation duration T hold is to the heating standard time T std hold ;

[0117] Evaluate the cooling stage:

[0118]

[0119] In the formula, S cool Indicates the output result of the cooling stage evaluation. The closer the value is to 1, the closer the actual heat preservation duration T cool is to the heating standard time T std cool ;

[0120] Construct a temperature evaluation model:

[0121] S temp =γ·S heat +δ·S hold +ε·S cool

[0122] In the formula, S tempDenoted as the temperature evaluation model, γ, δ, and ε are denoted as weight coefficients. During the sintering process, the degrees of influence of the heating-up, heat-preservation, and cooling-down stages on product quality are not the same. Therefore, adjustments need to be made according to specific process requirements. Among them, γ + δ + ε = 1. The heat-preservation stage provides the necessary time and stable temperature environment for the chemical reactions in zinc oxide production, enabling various chemical reactions to proceed fully. If the heat-preservation time is insufficient, the decomposition of zinc hydroxide is incomplete, resulting in low purity of zinc oxide in the final product and having a decisive impact on product quality. So the heat-preservation stage plays the most crucial role in ensuring product quality and needs to be given the largest weight δ in the model; the heating-up stage is to make the reactants reach the activation energy required for the reaction and initiate various chemical reactions. If the heating-up rate is too slow or the final temperature fails to meet the requirements, the reaction cannot proceed smoothly or the reaction rate is too slow, affecting production efficiency and product quality. However, compared with the heat-preservation stage, the heating-up stage mainly plays a role in initiating the reaction, while the heat-preservation stage is the key to ensuring the full progress of the reaction. Therefore, the weight γ of the heating-up stage should be less than the weight δ of the heat-preservation stage; the cooling-down stage is mainly to cool the reaction products under relatively stable temperature conditions, further stabilizing the crystal structure of zinc oxide and avoiding problems such as stress concentration caused by rapid cooling, resulting in crystal structure defects or product performance degradation. Although the cooling-down stage has a certain impact on the final performance of the product, it is an auxiliary stage after the reaction is completed, mainly for stabilizing the structure and optimizing the performance of the already formed product, and does not directly affect the progress of chemical reactions and the basic quality characteristics of the product like the heat-preservation stage and the heating-up stage. So its weight ε is slightly smaller. Therefore, the weight coefficients satisfy δ > γ > ε > 0. As S heat , S hold , S cool increases, the result of the temperature evaluation model is higher, which is in line with the actual situation of the influence of each process stage on product quality, indicating that the heat-preservation, heating-up, and cooling-down stages have a positive contribution to the temperature evaluation model.

[0123] For example, set the standard heating-up time as T std heat = 60 minutes, the standard heat-preservation time T std hold = 120 minutes, the standard cooling-down time is T std cool = 90 minutes, δ = 0.5, γ = 0.3, ε = 0.2, the actual heating-up duration T heat = 65 minutes, the actual heat-preservation duration T hold = 115 minutes, and the actual cooling-down duration is recorded as T cool = 95 minutes;

[0124] Evaluate the heating-up stage:

[0125]

[0126] Evaluate the heat-preservation stage:

[0127]

[0128] Evaluate the cooling stage:

[0129]

[0130] The output result of the temperature evaluation model is:

[0131] S temp =γ·S heat +δ·S hold +ε·S cool =0.3×0.92 + 0.5×0.96 + 0.2×0.94 = 0.944.

[0132] Step 5: Set key parameters for key parameters, calculate the final quality score using the weighted summation method, and comprehensively evaluate the product based on the key parameters of the key parameters and the final quality score. The key parameters include zinc oxide purity, torque standard deviation, up and down mold clamping displacement, horizontal displacement, heating time, heat preservation time, and cooling time;

[0133] Set the key parameter of zinc oxide purity of the raw materials in the production line to C′ Zno = 99.5%; Based on historical data, calculate the average value μ s and standard deviation σ s of the torque standard deviation of multiple qualified products. The key parameter of the torque standard deviation is S′ X =[μ s - 3σ s , μ s + 3σ s ; The key parameter of the up and down mold clamping displacement is {L min vert , L max vert}, the key parameter of the horizontal displacement is {L min vert , L max vert}, and the up and down mold clamping displacement and horizontal displacement are set according to the product specifications; The key parameter of the heating time is {T min heat , T max heat}, that is, {100, 300}, the key parameter of the heat preservation time is {T min hold , T max hold}, that is, {60, 180}, and the key parameter of the cooling time is {T min cool , T max cool}, that is, {100, 300};

[0134] In the production of zinc oxide varistors, too many impurities will damage the internal crystal structure, resulting in the inability to withstand high-voltage shocks, seriously affecting the product quality and use safety. When the zinc oxide purity C Zno > C′ ZnoWhen the purity of the warning zinc oxide is significantly abnormal; when the standard deviation of torque s j >S′ X When this occurs, it means that there is a serious abnormality in the stirring process, which may be due to a sudden change in the material properties or severe wear of the stirring paddle, and the warning torque fluctuation is significantly abnormal; when the upper and lower mold closing displacement When this occurs, the warning thickness is significantly abnormal. When the horizontal displacement When this occurs, the warning side length is significantly abnormal, avoiding the inability to complete the subsequent assembly process; when the heating time When this occurs, the warning heating duration is significantly abnormal. When the heat preservation time The warning heat preservation time is significantly abnormal. When the cooling time When this occurs, the warning cooling time is significantly abnormal, avoiding affecting its stable performance;

[0135] Calculate the final quality score according to the purity evaluation model, torque evaluation model, size evaluation model and temperature evaluation model:

[0136] Q = ω1·S Zno +ω2·S force +ω3·S dimension +ω4·S temp

[0137] In the formula, Q represents the final quality score, and ω1 + ω2 + ω3 + ω4 = 1. The higher the purity, the more stable the torque, the smaller the size error and the more suitable the temperature conditions, the S Zno 、S force 、S dimension and S temp The larger the result, the better the product quality and the higher the final quality score. The weights are greater than 0, and the contribution reflected in the final quality score is positive. Thus, it reasonably reflects that the output results of the evaluation model, torque evaluation model, size evaluation model and temperature evaluation model have a positive contribution to the final quality score. Set the values of the weight coefficients according to industry experience, ω1 > ω2 > ω3 > ω4 > 0. For example, the purity of zinc oxide affects the electrical performance and usually accounts for a relatively large weight, so ω1 = 0.35. The torque fluctuation affects the internal density and has a relatively high weight, so ω2 = 0.3. The size error affects the assembly accuracy, so ω3 = 0.2. The temperature affects the microstructure of the material, so ω4 = 0.15;

[0138] Set the final quality score thresholds Q1 = 0.7 and Q2 = 0.85. When Q ≥ Q2, the quality of the product is evaluated as high. When Q2 > Q ≥ Q1, the quality of the product is evaluated as qualified. When Q < Q1, the quality of the product is evaluated as unqualified.

[0139] For example, the purity evaluation model S Zno = 0.9, the torque evaluation model S force= 0.85, Dimension Evaluation Model S dimension = 0.8, Temperature Evaluation Model S temp = 0.75, Final Quality Score:

[0140] Q = ω1·S Zno + ω2·S force + ω3·S dimension + ω4·S temp

[0141] = 0.35×0.9 + 0.3×0.85 + 0.2×0.8 + 0.15×0.75 = 0.8425

[0142] Q2 > Q ≥ Q1, therefore, the quality of the evaluated product is qualified.

[0143] Please refer to Figure 2 , the present invention further provides a production line monitoring device based on the industrial Internet, and the device is used to execute the above-mentioned production line monitoring method based on the industrial Internet, including:

[0144] Purity Evaluation Module, which is used to preset the standard purity of zinc oxide, monitor the zinc oxide purity of raw materials in the production line, construct a purity evaluation model based on the zinc oxide purity of production line raw materials and the standard purity of zinc oxide, and record raw material batches and supplier information;

[0145] Torque Fluctuation Evaluation Module, which is used to set a time window, a time interval, a torque standard deviation threshold, and a fluctuation times threshold. Within each time window, torque data of the stirring process is collected at a fixed time interval, the torque standard deviation within the time window is calculated, and within all time windows, the number of times the torque standard deviation exceeds the torque standard deviation threshold is defined as the number of fluctuations, and a torque evaluation model is constructed based on the number of fluctuations and the fluctuation times threshold;

[0146] Dimension Forming Evaluation Module, which is used to set a standard thickness and a standard side length. During the process of mold closing from the start to full closure, the change value of the relative position of the upper and lower molds is recorded as the upper and lower mold closing displacement data. Taking the mold center as the reference point, the position change in the horizontal direction is measured and recorded as the horizontal direction displacement data, and a dimension evaluation model is constructed based on the standard thickness, the standard side length, the upper and lower mold closing displacement, and the horizontal direction displacement data;

[0147] Sintering Temperature Evaluation Module, which is used to divide the sintering process into heating, heat preservation, and cooling stages according to the temperature change situation, record the duration of each stage, set the standard time range for each stage, and construct a temperature evaluation model based on the duration of each stage and the standard time;

[0148] The comprehensive quality scoring module is used to set key parameters, calculate the final quality score by the method of weighted summation, and comprehensively evaluate the product based on the key parameters and the final quality score. The key parameters include zinc oxide purity, torque standard deviation, upper and lower die clamping displacement, horizontal displacement, heating time, heat preservation time, and cooling time.

[0149] The present invention further provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, the above-mentioned production line monitoring method based on the industrial Internet is realized.

[0150] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0151] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by the combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0152] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0153] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all of them should be covered by the protection scope of this application.

Claims

1. A production line monitoring method based on the industrial Internet, characterized in that, The specific steps include: Step 1: Preset the standard purity of zinc oxide, monitor the purity of zinc oxide in the raw materials on the production line, construct a purity evaluation model based on the purity of zinc oxide in the raw materials of the production line and the standard purity of zinc oxide, and record the raw material batches and supplier information; Step 2: Set the time window, time interval, torque standard deviation threshold, and number of fluctuations threshold. Within each time window, collect the torque data of the stirring process at fixed time intervals, calculate the torque standard deviation within the time window. In all time windows, define the number of times the torque standard deviation exceeds the torque standard deviation threshold as the number of fluctuations, and construct a torque evaluation model based on the number of fluctuations and the number of fluctuations threshold; Step 3: Set the standard thickness and standard side length. During the process of the mold closing from the start to full closure, record the change value of the relative position of the upper and lower molds as the up-and-down mold closing displacement data. Taking the mold center as the reference point, measure its position change in the horizontal direction and record it as the horizontal direction displacement data. Construct a dimension evaluation model based on the standard thickness, standard side length, up-and-down mold closing displacement, and horizontal direction displacement data; Step 4: Divide the sintering process into heating, heat preservation, and cooling stages according to the temperature change situation, and record the duration of each stage. Set the standard time range for each stage, and construct a temperature evaluation model based on the duration of each stage and the standard time; Step 5: Set the key parameters, calculate the final quality score using the weighted summation method, and comprehensively evaluate the product based on the key parameters and the final quality score. The key parameters include the purity of zinc oxide, torque standard deviation, up-and-down mold closing displacement, horizontal direction displacement, heating time, heat preservation time, and cooling time.

2. The production line monitoring method based on industrial Internet according to claim 1, characterized in that: In Step 1, the method for monitoring the purity of zinc oxide in the raw materials on the production line is: Measure the fluorescence spectrum of the raw material using an XRF spectrometer to obtain the light intensity data at different energies. Through the element counting rate function of the XRF spectrometer, record the counts at Kα (8.63 keV) and Kβ (9.57 keV) as the characteristic peak intensity I of Zn. Zn At the same time, obtain the total characteristic peak intensity I through the XRF spectrometer. total And the background signal intensity I. bkg Calculate the purity of zinc oxide in the raw material on the production line: In the formula, C Zno represents the zinc oxide purity of the raw materials in the production line.

3. A production line monitoring method based on industrial Internet according to claim 2, characterized in that: In Step 1, the method for constructing a purity evaluation model based on the purity of zinc oxide in the raw materials of the production line and the standard purity of zinc oxide is: Set the standard purity of zinc oxide to C zb , when the purity of zinc oxide C Zno < C zb , the purity evaluation model is: Wherein, S Zno represents the purity evaluation model, and C zb represents the standard purity of zinc oxide; When the purity of zinc oxide is C Zno ≥ C zb Set the output result of the purity evaluation model to 1.

4. The production line monitoring method based on industrial Internet according to claim 1, wherein: In Step 2, the method for collecting the torque data of the stirring process at fixed time intervals and calculating the torque standard deviation within the time window is: Set the time window as T and the time interval as t c , the torque standard deviation threshold is s max , the number of fluctuations threshold is f max , where T > t c > 0, at a fixed time interval t c , collect the torque data x during the mixing process i , where i = 1, 2,..., n, and n represents the total number of torque data collected within a certain time window. Calculate the average value based on the torque data and the total number of torque data collected within the time window: wherein, represents the average value of torque data within a time window; For the j-th time window, the formula for calculating its torque standard deviation is: where s j represents the standard deviation of torque within the j-th time window, where j = 1, 2, …, z, and z is the total number of time windows.

5. The production line monitoring method based on industrial Internet according to claim 4, characterized in that: In Step 2, the method for constructing a torque evaluation model based on the number of fluctuations and the number of fluctuations threshold is: Statistically analyze the standard deviation s of torque within all time windows j Exceeding the standard deviation threshold s of torque max The number of times is recorded as the fluctuation number m. Based on the fluctuation number and the fluctuation number threshold, a torque evaluation model is constructed: where S force represents the torque evaluation model.

6. The production line monitoring method based on industrial Internet according to claim 1, wherein: In Step 3, the method for constructing a dimension evaluation model based on the standard thickness, standard side length, up-and-down mold closing displacement, and horizontal direction displacement data is: Set the standard thickness L std vert and the standard side length L std hor During the process of the mold starting to close until it is completely closed, record the real-time change values of the relative positions of the upper and lower molds, denoted as the upper and lower mold closing displacement data L vert Taking the mold center as the reference point, measure its position change in the horizontal circumferential direction to obtain the horizontal direction displacement data L hor Normalize the error of the upper and lower mold closing displacement data: Wherein, S vert represents the value normalized by the error of the upper and lower die clamping displacement data; Normalize the error of the horizontal direction displacement data: where S hor represents the value of the normalization of the horizontal displacement data error; Construct a dimension evaluation model: S dimension = α·S vert + β·S hor In the formula, S dimension represents the dimensional evaluation model, and α and β represent the weight coefficients, which are adjusted according to specific process requirements. Among them, α > 0, β > 0, and α + β = 1.

7. The production line monitoring method based on industrial Internet according to claim 1, wherein: In Step 4, the method for constructing a temperature evaluation model based on the duration of each stage and the standard time is: During the sintering process, the set standard heating time is T std heat , and the actual heating duration is recorded as T heat ; Set the standard heat preservation time as T std hold , and record the actual heat preservation duration as T hold ; Set the standard cooling time as T std cool , and record the actual cooling duration as T cool ; Evaluate the heating stage: Where S heat represents the output result evaluated in the heating-up stage; Evaluate the heat preservation stage: Where S hold represents the output result evaluated for the heat preservation stage; Evaluate the cooling stage: Where S cool represents the output result evaluated in the cooling stage; Construct a temperature evaluation model: S temp = γ·S heat + δ·S hold + ε·S cool In the formula, S temp represents the temperature evaluation model, and γ, δ, and ε represent weight coefficients, which are adjusted according to specific process requirements. Among them, γ + δ + ε = 1, and δ > γ > ε > 0.

8. A production line monitoring method based on industrial Internet according to claim 1, characterized in that: In Step 5, the method for comprehensively evaluating the product quality based on the key parameters and the final quality score is: The key parameter for setting the zinc oxide purity of the raw materials in the production line is C' Zno ,; the key parameter for the standard deviation of torque is S' X ; the key parameter for the upper and lower mold closing displacement is {L min vert , L max vert}, and the key parameter for the horizontal displacement is {L min vert , L max vert}; the key parameter for the heating time is {T min heat , T max heat}, the key parameter for the heat preservation time is {T min hold , T max hold}, and the key parameter for the cooling time is {T min cool , T max cool}; When the purity C of zinc oxide Zno > C' Zno a warning is given that the purity of zinc oxide is significantly abnormal; When the standard deviation s of the torque j > S′ X the alarm for obvious abnormal torque fluctuation is triggered; When the vertical mold clamping displacement occurs, the warning thickness is significantly abnormal; when the horizontal displacement occurs, the warning side length is significantly abnormal. When the heating-up time the alarm heating-up duration is significantly abnormal; When the heat preservation time Alarm: The heat preservation time is significantly abnormal; When the cooling time is reached, the alarm cooling time is significantly abnormal; Calculate the final quality score according to the purity evaluation model, torque evaluation model, dimension evaluation model, and temperature evaluation model: Q = ω1·S Zno + ω2·S force + ω3·S dimension + ω4·S temp In the formula, Q represents the final quality score, and ω1 + ω2 + ω3 + ω4 = 1, ω1 > ω2 > ω3 > ω4 > 0; Set the final quality score thresholds Q1 and Q2, where 0 < Q1 < Q2. When Q ≥ Q2, the quality of the product is evaluated as high; when Q2 > Q ≥ Q1, the quality of the product is evaluated as qualified; when Q < Q1, the quality of the product is evaluated as unqualified.

9. A production line monitoring device based on the industrial Internet, characterized in that: The device is used to execute a production line monitoring method based on the industrial Internet according to any one of claims 1-8: A purity evaluation module, which is used to preset the standard purity of zinc oxide, monitor the zinc oxide purity of raw materials in the production line, construct a purity evaluation model based on the zinc oxide purity of raw materials in the production line and the standard purity of zinc oxide, and record the raw material batch and supplier information; A torque fluctuation evaluation module, which is used to set a time window, a time interval, a torque standard deviation threshold, and a fluctuation times threshold. Within each time window, torque data of the stirring process is collected at a fixed time interval, the torque standard deviation within the time window is calculated, and within all time windows, the number of times the torque standard deviation exceeds the torque standard deviation threshold is defined as the number of fluctuations. A torque evaluation model is constructed based on the number of fluctuations and the fluctuation times threshold; A size forming evaluation module, which is used to set a standard thickness and a standard side length. During the process of the mold closing from the start to the fully closed state, the change value of the relative position of the upper and lower molds is recorded as the upper and lower mold closing displacement data. Taking the mold center as the reference point, the position change in the horizontal direction is measured and recorded as the horizontal direction displacement data. A size evaluation model is constructed based on the standard thickness, the standard side length, the upper and lower mold closing displacement, and the horizontal direction displacement data A sintering temperature evaluation module, which is used to divide the sintering process into a heating stage, a heat preservation stage, and a cooling stage according to the temperature change situation, record the duration of each stage, set the standard time range for each stage, and construct a temperature evaluation model based on the duration of each stage and the standard time; A comprehensive quality scoring module, which is used to set key parameters, calculate the final quality score by using the weighted summation method, and comprehensively evaluate the product based on the key parameters of the key parameters and the final quality score. The key parameters include zinc oxide purity, torque standard deviation, upper and lower mold closing displacement, horizontal direction displacement, heating time, heat preservation time, and cooling time.

10. A storage medium, characterized in that: The computer program is stored in the storage medium, and when the computer program is executed by a processor, it implements a production line monitoring method based on the industrial Internet according to any one of claims 1-8.

Images