A management system for forging intelligent production line based on Internet of Things

Abstract

The invention discloses a management system for a forging intelligent production line based on the Internet of Things, which specifically relates to the technical field of forging production monitoring and control, including a heating process monitoring module, a temperature control performance analysis module, a forging process monitoring module, a forging control performance analysis module and a production status analysis module. By collecting key heating process and forging process parameters in forging production, real-time continuous data is obtained and analyzed, the heating process is monitored and controlled based on the temperature control performance index, and the forging process is monitored and controlled based on the forging control performance index, thereby effectively realizing the management of key processes in forging production, solving the problems in the prior art of being unable to timely discover and interrupt abnormal processes in forgings, and lacking effective monitoring and control of forging production processes; the production efficiency abnormality evaluation index of forging intelligent production is understood through the production status analysis module, helping users to timely discover forging production abnormalities.

Description

Technical Field

[0001] The present invention relates to the technical field of forging production monitoring and control, and more specifically, to a management system for a forging intelligent production line based on the Internet of Things. Background Art

[0002] The IoT-based forging intelligent production line management system is an advanced production management system that integrates IoT technology, aiming to achieve comprehensive monitoring, optimization and management of the forging production process. The system connects various equipment on the production line through IoT technology, realizes centralized monitoring and management of equipment, collects equipment operation status, production data and other information in real time, and predicts equipment maintenance needs and potential failures through data analysis, so as to carry out maintenance and repair in advance and avoid production interruptions.

[0003] However, it still has many shortcomings in actual use. For example, the existing intelligent production management system for forgings collects real-time single-point data in production through the Internet of Things, but cannot reflect the characteristics of real-time continuous data in forging production, resulting in inaccurate data analysis results, and cannot promptly discover and interrupt abnormal processes in forgings, resulting in the inability to effectively monitor and control the forging production process. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a management system for an intelligent production line of forgings based on the Internet of Things. By real-time monitoring of the heating and forging processes in forging production, real-time continuous data on temperature and forging parameters can be obtained, so as to more effectively monitor the forging production process and, based on the monitoring results, achieve effective control of the process to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a management system for a forging intelligent production line based on the Internet of Things, comprising:

[0006] A heating process monitoring module is used to collect the real-time temperature of the heating process and transmit the collected real-time temperature to the temperature control performance analysis module;

[0007] The temperature control performance analysis module inputs the collected real-time temperature into the temperature control performance analysis model and outputs the temperature control performance index WK of the heating process;

[0008] The forging process monitoring module is used to collect the rolling force, rolling speed, rolling friction resistance and rolling temperature parameters in the forging process and transmit them to the forging control performance analysis module;

[0009] The forging control performance analysis module inputs the collected rolling force, rolling speed, rolling friction resistance and rolling temperature parameters into the forging control performance analysis model and outputs the forging control performance index DK of the forging process;

[0010] The production process control module receives the temperature control performance index WK and the forging control performance index DK, and takes control measures based on the relationship between the temperature control performance index WK and the forging control performance index DK and the corresponding preset values.

[0011] Preferably, the operation process of the temperature control performance analysis model includes the following steps:

[0012] Step S11: Set the weight coefficient q of each acquisition point based on the distance between each acquisition point and the workpiece. c ;

[0013] Step S12, calculate the temperature uniformity description parameter wj: execute the temperature uniformity analysis instruction, record the temperature collected at each collection point as w c , through the formula Xw c =w c *q c Calculate the corrected temperature Xw of each sampling point c , calculate the variance of the corrected temperature and output the temperature uniformity description parameter wj;

[0014] Step S13, compare the temperature uniformity description parameter wj with the preset value wj0. When the preset value requirement is met, the temperature uniformity is output as normal, and the workpiece heating process is continued. The temperature uniformity description parameter is monitored until the workpiece heating process is completed, and a temperature consistency analysis instruction is generated. When the preset value requirement is not met, the temperature uniformity is output as abnormal, and the user is prompted that the heating furnace is abnormal, and the workpiece heating process is interrupted.

[0015] Step S14: Execute the temperature consistency analysis command: obtain the average value of the corrected temperature at different locations at the same time Continuously monitor and record the corrected temperature average value at each time point to obtain the corrected temperature real-time change curve; compare the corrected temperature real-time change curve with the preset temperature change curve, calculate the square of the temperature difference at each time point, and take the average of the accumulated sum of all time points to obtain the temperature consistency description parameter wy;

[0016] Step S15, compare the temperature consistency description parameter wy with the preset value wy0, if it meets the preset value requirement, the output temperature consistency is normal, continue the workpiece heating process, monitor the temperature consistency description parameter until the workpiece heating process is completed, and generate a temperature control performance analysis instruction; if it does not meet the preset value requirement, the output temperature consistency is abnormal, prompt the user that the heating furnace is abnormal, and interrupt the workpiece heating process;

[0017] Step S16, executing the temperature control performance analysis instruction, when the temperature uniformity description parameter wj and the temperature consistency description parameter wy both meet the corresponding preset values, jointly analyze the temperature uniformity description parameter wj and the temperature consistency description parameter wy to obtain the temperature control performance index WK.

[0018] Preferably, by formula Get the weight coefficient q of each collection point c , where l sc Indicates the distance between the collection point and the heating position of the forging, k w is a proportional coefficient used to adjust the sensitivity of weight changes with distance, k w The larger the value, the more sensitive the weight is to changes in distance; w The smaller the value, the smoother the weight changes with distance, and the actual application scenario needs to be adjusted and optimized.

[0019] Preferably, the number of collection points at the same time is P, q represents the number of the collection points, and the corrected temperature of the qth collection point is recorded as Xw cq , the average corrected temperature at different locations at the same time satisfies the formula By formula The temperature uniformity description parameter wj is calculated.

[0020] Preferably, assume that K time points are monitored, s is used to represent the number of the time point, and FT is used s Indicates the temperature at the sth time point in the corrected temperature real-time change curve, expressed as FY s It represents the preset temperature at the sth time point in the preset temperature change curve, through the following formula,

[0021] The temperature consistency description parameter wy is calculated.

[0022] Preferably, by formula The temperature control performance index WK is calculated, where α1 and α2 represent weight coefficients of each item, 0<α1<1, 0<α2<1, and α1+α2=1.0.

[0023] Preferably, the operation process of the forging control performance analysis model includes the following steps:

[0024] Step 21, execute the forging instruction, monitor the rolling temperature and rolling friction resistance in real time during forging, and when the rolling temperature and rolling friction resistance are abnormal, warn the user, interrupt the forging process, and adjust the rolling temperature and rolling friction resistance until the requirements are met;

[0025] Step 22, obtaining rolling force control parameters, comparing the actual rolling force curve with the preset rolling force curve, calculating the square of the rolling force difference at each time point, accumulating and summing all time points and averaging them to obtain the rolling force control parameters;

[0026] Step 23, obtaining rolling speed control parameters, comparing the actual rolling speed curve with the preset rolling speed curve, calculating the square of the rolling speed difference at each time point, accumulating and summing all time points and averaging them to obtain the rolling speed control parameters;

[0027] Step 24: jointly analyze the rolling force control parameters, rolling speed control parameters, rolling temperature parameters and abnormal description parameters of rolling friction resistance, and calculate the forging control performance index DK.

[0028] Preferably, by formula The forging control performance index DK is calculated, where Zf represents the rolling force control parameter, Zv represents the rolling speed control parameter, Zwd represents the rolling temperature parameter, the average rolling friction resistance is recorded as Mf, and the preset rolling friction resistance is recorded as Mf 预 , the abnormal description parameters of rolling friction resistance are calculated β1, β2, β3, and β4 represent various proportional coefficients, β1+β2+β3+β4=1.0.

[0029] Preferably, the control measures include: when the temperature control performance index WK is not lower than a preset value, an early warning is sent to indicate that the heating process is abnormal, heating is interrupted, and the workpiece in the heating process does not enter the next production step; when the temperature control performance index WK is lower than a preset value, it indicates that the heating process is normal, and after the heating process is completed, the workpiece in the heating process enters the next production step; when the forging control performance index DK is not lower than a preset value, an early warning is sent to indicate that the forging process is abnormal, forging is interrupted, and the workpiece in the forging process does not enter the next production step; when the forging control performance index DK is lower than a preset value, it indicates that the forging process is normal, and after the forging process is completed, the workpiece in the forging process enters the next production step.

[0030] Preferably, the system of the present invention further comprises:

[0031] The production status monitoring module is used to obtain the actual unit workload production efficiency and actual unit workload energy consumption in each production batch, and transmit them to the production status analysis module;

[0032] The production status analysis module inputs the actual unit workload production efficiency and the actual unit workload energy consumption into the production status analysis model, and outputs the production efficiency abnormality evaluation index XNY of the production batch;

[0033] Point cloud data acquisition module, which installs machine vision equipment at the output end of the production batch to collect point cloud data of the finished forgings in real time and transmit the collected point cloud data to the forging appearance defect recognition module;

[0034] Forging appearance defect recognition module: collects point cloud information of forging finished products through machine vision, inputs the point cloud information into the appearance defect recognition model, outputs the defect location and defect type of the forging appearance, and transmits the output defect location and defect type to the forging appearance quality analysis module;

[0035] The forging appearance quality analysis module inputs the appearance defect size and appearance defect type of each batch of forging finished products into the appearance quality assessment model and outputs the appearance quality assessment index WYp of the production batch.

[0036] Preferably, the production status analysis model satisfies the formula Among them, XNY represents the production efficiency abnormality evaluation index, Sxl represents the actual unit workload production efficiency, Sxn represents the actual unit workload energy consumption, Yxl represents the preset unit workload production efficiency, Yxn represents the preset unit workload energy consumption, γ1, γ2, γ3 represent the proportional coefficients of each item respectively, and γ1+γ2+γ3=1.0.

[0037] Preferably, assuming that the production batch contains N forging finished products, i represents the number of the forging finished product, the appearance quality evaluation model satisfies the formula Among them, QM_i represents the appearance defect area of ​​the i-th forging finished product, y_i represents the influencing factor of the appearance defect type of the i-th forging finished product, and Mi_i represents the area of ​​the i-th forging finished product.

[0038] Preferably, the system in the embodiment of the present invention further includes:

[0039] Forging performance risk management module: Based on the appearance quality evaluation index WYp, production efficiency abnormality evaluation index XNY, temperature control performance index WK and forging control performance index DK of each production batch, a joint analysis is performed to obtain the performance abnormality risk coefficient of each batch, and measures are taken based on the relationship between the performance abnormality risk coefficient and the preset value.

[0040] Preferably, the performance abnormality risk coefficient Fxn is obtained by linearly normalizing the production efficiency abnormality evaluation index, the appearance quality evaluation index, the temperature control performance index and the forging control performance index and recording them as: XNY", WYp", WK", DK", and calculating them by the formula

[0041] Technical effects and advantages of the present invention:

[0042] (1) The management system of the forging intelligent production line based on the Internet of Things provided by the present invention obtains real-time continuous data and analyzes it by collecting key heating process and forging process parameters in forging production, monitors and controls the heating process based on the temperature control performance index WK, and monitors and controls the forging process based on the forging control performance index DK, thereby effectively realizing the management of the key processes in forging production and solving the problems in the prior art that the Internet of Things in the forging production process collects real-time single-point data in production, cannot timely discover and interrupt abnormal processes in forgings, and lacks effective monitoring and control of the forging production processes.

[0043] (2) The management system of the forging intelligent production line based on the Internet of Things provided by the present invention, through the production status analysis module, understands the production efficiency anomaly evaluation index of forging intelligent production from the perspectives of production efficiency, energy consumption and the ratio of energy consumption to production efficiency; the production efficiency deviation is one of the key indicators for evaluating whether the forging intelligent production is abnormal; the energy consumption deviation reflects the amount of energy consumed in the production process, and the deviation of the ratio of energy consumption to production efficiency provides a comprehensive perspective to evaluate production anomalies. The ratio of energy consumption to production efficiency reflects the amount of energy consumed per unit of production efficiency. When the production efficiency anomaly evaluation index exceeds a certain preset threshold, the production efficiency anomaly evaluation index can help users promptly discover forging production anomalies.

[0044] (3) The management system of the forging intelligent production line based on the Internet of Things provided by the present invention builds an appearance defect recognition model, predicts the defect position and the actual defect position into a defect position loss parameter calculation model, and outputs the defect position loss parameter; inputs the predicted defect type probability and the actual defect type into the defect type loss parameter calculation model, and outputs the defect type loss parameter; jointly analyzes the defect position loss parameter and the defect type loss parameter to obtain an appearance defect recognition model that adapts to the actual situation, realizes the forging appearance quality inspection based on machine vision, can save resources, and improve the quality of forging appearance inspection; helps users quickly understand the appearance quality of forgings and screen out forgings with abnormal appearance. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a structural block diagram of the forging intelligent production line management system of the present invention.

[0046] Figure 2 It is a schematic diagram of the temperature control performance analysis model flow of the present invention.

[0047] Figure 3 It is a schematic diagram of the forging control performance analysis model flow of the present invention.

[0048] Figure 4 This is a structural block diagram of the forging intelligent production line management system based on forging performance risk control of the present invention. DETAILED DESCRIPTION

[0049] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0050] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present application, its application, or uses.

[0051] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0052] Example 1

[0053] See also Figure 1 The structural block diagram of the forging intelligent production line management system of the present invention provides the following Figure 1 The management system of a forging intelligent production line based on the Internet of Things shown includes:

[0054] A heating process monitoring module is used to collect the real-time temperature of the heating process and transmit the collected real-time temperature to the temperature control performance analysis module;

[0055] The temperature control performance analysis module inputs the collected real-time temperature into the temperature control performance analysis model and outputs the temperature control performance index WK of the heating process;

[0056] The forging process monitoring module is used to collect the rolling force, rolling speed, rolling friction resistance and rolling temperature parameters in the forging process and transmit them to the forging control performance analysis module;

[0057] The forging control performance analysis module inputs the collected rolling force, rolling speed, rolling friction resistance and rolling temperature parameters into the forging control performance analysis model and outputs the forging control performance index DK of the forging process;

[0058] The production process control module receives the temperature control performance index WK and the forging control performance index DK, and takes control measures based on the relationship between the temperature control performance index WK and the forging control performance index DK and the corresponding preset values.

[0059] In the embodiments of the present invention, it is necessary to further explain the invention. Figure 2 The operation process of the temperature control performance analysis model includes the following steps:

[0060] Step S11: Set the weight coefficient q of each acquisition point based on the distance between each acquisition point and the workpiece. c ;

[0061] Description: In the heating process, the workpiece is placed at a fixed position, and a collection point for collecting temperature is set in the heating device. The closer the collection point is to the workpiece, the greater the weight should be, because the temperature at these positions has a greater impact on the workpiece; the weight coefficient is set according to the actual situation, and the reciprocal of the distance from the collection point to the heating position or the reciprocal square of the distance is used as the weight;

[0062] Exemplarily, in the embodiment of the present invention, the reciprocal square of the distance is used as the weight coefficient, and the formula Get the weight coefficient q of each collection point c , where l sc Indicates the distance between the collection point and the heating position of the forging, k w is a proportional coefficient used to adjust the sensitivity of weight changes with distance, k w The larger the value, the more sensitive the weight is to changes in distance; w The smaller the value, the smoother the weight changes with distance, and the actual application scenario needs to be adjusted and optimized.

[0063] Step S12, calculate the temperature uniformity description parameter wj: execute the temperature uniformity analysis instruction, record the temperature collected at each collection point as w c , through the formula Xw c =w c *q c Calculate the corrected temperature Xw of each sampling point c , calculate the variance of the corrected temperature and output the temperature uniformity description parameter wj;

[0064] For example, in the embodiment of the present invention, it is assumed that there are P collection points at the same time, q represents the number of the collection points, and the corrected temperature of the qth collection point is recorded as Xw cq , the average corrected temperature at different locations at the same time satisfies the formula By formula The temperature uniformity description parameter wj is calculated;

[0065] Step S13, compare the temperature uniformity description parameter wj with the preset value wj0. When the preset value requirement is met, the temperature uniformity is output as normal, and the workpiece heating process is continued. The temperature uniformity description parameter is monitored until the workpiece heating process is completed, and a temperature consistency analysis instruction is generated. When the preset value requirement is not met, the temperature uniformity is output as abnormal, and the user is prompted that the heating furnace is abnormal, and the workpiece heating process is interrupted.

[0066] Step S14, execute the temperature consistency analysis command: obtain the average value T of the corrected temperature at different locations at the same time tw , continuously monitor and record the corrected temperature average value at each time point to obtain the corrected temperature real-time change curve; compare the corrected temperature real-time change curve with the preset temperature change curve, calculate the square of the temperature difference at each time point, and take the average of the accumulated sum of all time points to obtain the temperature consistency description parameter wy;

[0067] For example, in the embodiment of the present invention, assume that K time points are monitored, s is used to represent the number of the time point, and FT is used to represent the number of the time point. s Indicates the temperature at the sth time point in the corrected temperature real-time change curve, expressed as FY s It represents the preset temperature at the sth time point in the preset temperature change curve, through the following formula,

[0068] The temperature consistency description parameter wy is calculated;

[0069] Step S15, compare the temperature consistency description parameter wy with the preset value wy0, if it meets the preset value requirement, the output temperature consistency is normal, continue the workpiece heating process, monitor the temperature consistency description parameter until the workpiece heating process is completed, and generate a temperature control performance analysis instruction; if it does not meet the preset value requirement, the output temperature consistency is abnormal, prompt the user that the heating furnace is abnormal, and interrupt the workpiece heating process;

[0070] Step S16, executing the temperature control performance analysis instruction, when the temperature uniformity description parameter wj and the temperature consistency description parameter wy both meet the corresponding preset values, jointly analyze the temperature uniformity description parameter wj and the temperature consistency description parameter wy to obtain the temperature control performance index WK.

[0071] Exemplarily, in the embodiment of the present invention, by formula The temperature control performance index WK is calculated, where α1 and α2 represent weight coefficients of each item, 0<α1<1, 0<α2<1, and α1+α2=1.0.

[0072] In the embodiments of the present invention, it is necessary to further explain the invention. Figure 3 The forging control performance analysis model process diagram is shown in FIG. The operation process of the forging control performance analysis model includes the following steps:

[0073] Step 21, execute the forging instruction, monitor the rolling temperature and rolling friction resistance in real time during forging, and when the rolling temperature and rolling friction resistance are abnormal, warn the user, interrupt the forging process, and adjust the rolling temperature and rolling friction resistance until the requirements are met;

[0074] Step 22, obtaining rolling force control parameters, comparing the actual rolling force curve with the preset rolling force curve, calculating the square of the rolling force difference at each time point, accumulating and summing all time points and averaging them to obtain the rolling force control parameters;

[0075] Step 23, obtaining rolling speed control parameters, comparing the actual rolling speed curve with the preset rolling speed curve, calculating the square of the rolling speed difference at each time point, accumulating and summing all time points and averaging them to obtain the rolling speed control parameters;

[0076] Step 24: jointly analyze the rolling force control parameters, the rolling speed control parameters, the rolling temperature parameters and the abnormal description parameters of the rolling friction resistance, and calculate the forging control performance index DK.

[0077] Exemplarily, in the embodiment of the present invention, by formula The forging control performance index DK is calculated, where Zf represents the rolling force control parameter, Zv represents the rolling speed control parameter, Zwd represents the rolling temperature parameter, the average rolling friction resistance is recorded as Mf, and the preset rolling friction resistance is recorded as Mf 预 , the abnormal description parameters of rolling friction resistance are calculated β1, β2, β3, and β4 represent various proportional coefficients, β1+β2+β3+β4=1.0.

[0078] What needs to be further explained in the embodiments of the present invention is that the control measures include: when the temperature control performance index WK is not lower than a preset value, an alert is given to the user, indicating that the heating process is abnormal, heating is interrupted, and the workpiece in the heating process does not enter the next production step; when the temperature control performance index WK is lower than a preset value, it indicates that the heating process is normal, and after the heating process is completed, the workpiece in the heating process enters the next production step; when the forging control performance index DK is not lower than a preset value, an alert is given to the user, indicating that the forging process is abnormal, forging is interrupted, and the workpiece in the forging process does not enter the next production step; when the forging control performance index DK is lower than a preset value, it indicates that the forging process is normal, and after the forging process is completed, the workpiece in the forging process enters the next production step.

[0079] Summary: The embodiment of the present invention collects parameters in key heating and forging processes in forging production, obtains real-time continuous data and performs analysis, monitors and controls the heating process based on the temperature control performance index WK, and monitors and controls the forging process based on the forging control performance index DK, thereby effectively realizing the management of key processes in forging production and solving the problem in the prior art of lack of real-time monitoring, analysis and intelligent control of the heating and forging processes in forging production.

[0080] Example 2

[0081] Based on the heating process monitoring module, temperature control performance analysis module, forging process monitoring module, forging control performance analysis module and production process control module in Example 1, intelligent production of forgings is performed to obtain several batches of forging finished products.

[0082] See also Figure 4 The structural block diagram of the forging intelligent production line management system based on forging performance risk control is shown in FIG. The difference between the embodiment of the present invention and embodiment 1 is that the embodiment of the present invention provides a management system for forging intelligent production line based on the Internet of Things, further comprising:

[0083] The production status monitoring module is used to obtain the actual unit workload production efficiency and actual unit workload energy consumption in each production batch, and transmit them to the production status analysis module;

[0084] The production status analysis module inputs the actual unit workload production efficiency and the actual unit workload energy consumption into the production status analysis model, and outputs the production efficiency abnormality evaluation index XNY of the production batch;

[0085] Point cloud data acquisition module, which installs machine vision equipment at the output end of the production batch to collect point cloud data of the finished forgings in real time and transmit the collected point cloud data to the forging appearance defect recognition module;

[0086] Forging appearance defect recognition module: collects point cloud information of forging finished products through machine vision, inputs the point cloud information into the appearance defect recognition model, outputs the defect location and defect type of the forging appearance, and transmits the output defect location and defect type to the forging appearance quality analysis module;

[0087] The forging appearance quality analysis module inputs the appearance defect size and appearance defect type of each batch of forging finished products into the appearance quality assessment model and outputs the appearance quality assessment index WYp of the production batch.

[0088] In the embodiment of the present invention, it is necessary to further explain that the production status analysis model satisfies the formula Among them, XNY represents the production efficiency abnormality evaluation index, Sxl represents the actual unit workload production efficiency, Sxn represents the actual unit workload energy consumption, Yxl represents the preset unit workload production efficiency, Yxn represents the preset unit workload energy consumption, γ1, γ2, γ3 represent the proportional coefficients of each item respectively, and γ1+γ2+γ3=1.0.

[0089] What needs to be further explained in the embodiments of the present invention is that the actual unit workload production efficiency is obtained by dividing the output of the production line within a certain period of time by the corresponding input workload, for example, the actual unit workload production efficiency is obtained by dividing the production batch consumption time by the corresponding input workload; the energy consumption per unit workload is obtained by dividing the total energy consumption of the production line within a period of time by the corresponding workload, for example, the energy consumption per unit workload is obtained by dividing the total energy consumption of the production line within the production batch consumption time by the corresponding workload; the preset unit workload production efficiency and the preset unit workload energy consumption are obtained by pre-setting according to factors such as process requirements, equipment performance and material characteristics, or by analyzing parameters under normal forging production conditions.

[0090] In the embodiment of the present invention, it is necessary to further explain that, assuming that the production batch contains N forging finished products, i represents the number of the forging finished product, and the appearance quality evaluation model satisfies the formula Among them, QM_i represents the appearance defect area of ​​the i-th forging finished product, y_i represents the influencing factor of the appearance defect type of the i-th forging finished product, and Mi_i represents the area of ​​the i-th forging finished product.

[0091] It is necessary to further explain the invention in the embodiment of the present invention that the system in the embodiment of the present invention further includes:

[0092] Forging performance risk management module: Based on the appearance quality evaluation index WYp, production efficiency abnormality evaluation index XNY, temperature control performance index WK and forging control performance index DK of each production batch, a joint analysis is performed to obtain the performance abnormality risk coefficient of each batch, and measures are taken based on the relationship between the performance abnormality risk coefficient and the preset value.

[0093] It is necessary to further explain the invention in the embodiment of the present invention that the performance abnormality risk coefficient Fxn is obtained by linearly normalizing the production efficiency abnormality evaluation index, the appearance quality evaluation index, the temperature control performance index and the forging control performance index and recording them as: XNY", WYp", WK", DK", and calculating them by the formula

[0094] What needs to be further explained in the embodiments of the present invention is that measures are taken based on the performance abnormality risk coefficient, including dynamically adjusting the number of performance sampling inspections in the production batch, and setting different performance test sampling ratios according to the performance abnormality risk coefficient Fxn. When the performance abnormality risk coefficient Fxn exceeds the preset value, the sampling ratio is increased to more comprehensively detect its performance; when the performance abnormality risk coefficient Fxn is lower than the preset value, the sampling ratio is reduced to save detection costs; according to actual production conditions and needs, the sampling ratio is adjusted regularly or irregularly. For example, when the performance abnormality risk coefficient Fxn of a batch of forging finished products is low for a period of time, the sampling ratio of the batch should be increased to ensure product quality.

[0095] It is necessary to further explain the invention in the embodiment of the present invention that the process of building the shape defect recognition model includes the following steps:

[0096] Step S11, collecting point cloud information of the forging, manually marking the defect type and defect location in the point cloud information, and dividing the point cloud information into a training set and a verification set;

[0097] Step S12: Build a convolutional neural network and set parameters and activation functions;

[0098] Note that the parameters are learning rate, batch size, and number of iterations; the activation function is used to introduce nonlinear factors so that the neural network can fit complex functional relationships. The activation function is one of ReLU and Sigmoid;

[0099] Step S13: input the point cloud data and corresponding labels in the training set into the convolutional neural network, calculate the loss function by comparing the convolutional neural network output labels with the true labels through forward propagation; update the network parameters through back propagation; repeat this process until the loss function meets the preset value or reaches the maximum number of iterations, and obtain the trained convolutional neural network framework;

[0100] Step S14, using the validation set to verify the performance parameters of the trained convolutional neural network framework;

[0101] Step S15: record the convolutional neural network framework whose performance parameters meet the requirements as an appearance defect recognition model, and output it for application.

[0102] It is necessary to further explain the invention in the embodiment of the present invention that the process of obtaining the loss function includes the following steps:

[0103] Get the defect location loss parameter L w:The predicted defect position and the actual defect position are input into the defect position loss parameter calculation model, and the defect position loss parameter is output; the defect position loss parameter calculation model is: obtaining the intersection and union of the predicted defect position and the actual defect position; using the intersection of the predicted defect position and the actual defect position and the ratio of the predicted defect position to the actual defect position to represent the defect position loss parameter; the defect position loss parameter L w The value range is [0, 1], L w The larger the value, the closer the predicted position is to the actual position;

[0104] Description: The defect position loss parameter is used to represent the deviation between the predicted defect position and the actual defect position; the intersection refers to the part covered by the predicted defect area and the actual defect area, which is determined by comparing the pixels or coordinates of the two areas; the union refers to the entire area covered by the predicted defect area and the actual defect area after being merged; the intersection and union are the intersection area or volume of the predicted area and the actual area.

[0105] Get defect type loss parameter L g : Input the predicted defect type probability and the actual defect type into the defect type loss parameter calculation model, and output the defect type loss parameter; the defect type loss parameter calculation model is a cross entropy loss function;

[0106] Note that the defect type loss parameter calculation model satisfies:

[0107]

[0108] Assume that there are H appearance defects in the forging, and use g to represent the number of the appearance defect, where y_g is the actual defect type probability; if the forging defect belongs to the gth category, then y_g = 1, otherwise y_g = 0; p_g is the predicted defect type probability that the forging defect belongs to the gth category.

[0109] Note: When the predicted probability distribution is closer to the actual probability distribution, the value of the cross entropy loss function is smaller, indicating better prediction performance; conversely, if the predicted probability distribution is significantly different from the actual probability distribution, the value of the cross entropy loss function will increase.

[0110] Combined analysis of the defect location loss parameter and the defect type loss parameter to obtain a loss function. For example, the defect location loss parameter L w and defect type loss parameter L g Input loss function analysis model, output loss function, the loss function analysis model is through formula L loss =q1*L w +q2*L g Calculate the loss function L loss, where q1 represents the defect position influencing factor, q2 represents the defect type influencing factor, 0<q1<100%, 0<q2<100%, and q1+q2=100%.

[0111] Summary: The embodiment of the present invention performs intelligent production of forgings on the basis of embodiment 1, obtains several batches of finished forgings, and understands the production efficiency abnormality evaluation index of intelligent production of forgings from the perspective of production efficiency, energy consumption and the ratio of energy consumption to production efficiency by monitoring and analyzing the production status; helps managers understand the abnormality of forging production; obtains the appearance quality evaluation index WYp by analyzing the appearance of the finished forgings, and helps to screen out finished forgings with normal appearance; dynamically adjusts the number of performance tests of finished forgings based on the appearance quality evaluation index WYp and the production efficiency abnormality evaluation index of each production batch. Makes full use of automation and machine vision technology, improves the appearance inspection of finished forgings, reduces the possibility of human error, and makes the forging process more intelligent and controllable.

[0112] Finally: 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 in the protection scope of the present invention.

Claims

1. A management system for a forging intelligent production line based on the Internet of Things, characterized in that: include: A heating process monitoring module is used to collect the real-time temperature of the heating process and transmit the collected real-time temperature to the temperature control performance analysis module; The temperature control performance analysis module inputs the collected real-time temperature into the temperature control performance analysis model and outputs the temperature control performance index WK of the heating process; The operation process of the temperature control performance analysis model includes the following steps: Step S11: Set the weight coefficient q of each acquisition point based on the distance between each acquisition point and the workpiece. c ; Step S12, calculate the temperature uniformity description parameter wj: execute the temperature uniformity analysis instruction, record the temperature collected at each collection point as w c , through the formula Xw c =w c *q c Calculate the corrected temperature Xw of each sampling point c , calculate the variance of the corrected temperature and output the temperature uniformity description parameter wj; Step S13, compare the temperature uniformity description parameter wj with the preset value wj0. When the preset value requirement is met, the temperature uniformity is output as normal, and the workpiece heating process is continued. The temperature uniformity description parameter is monitored until the workpiece heating process is completed, and a temperature consistency analysis instruction is generated. When the preset value requirement is not met, the temperature uniformity is output as abnormal, and the user is prompted that the heating furnace is abnormal, and the workpiece heating process is interrupted. Step S14: Execute the temperature consistency analysis command: obtain the average value of the corrected temperature at different locations at the same time Continuously monitor and record the corrected temperature average value at each time point to obtain the corrected temperature real-time change curve; compare the corrected temperature real-time change curve with the preset temperature change curve, calculate the square of the temperature difference at each time point, and take the average of the accumulated sum of all time points to obtain the temperature consistency description parameter wy; Step S15, compare the temperature consistency description parameter wy with the preset value wy0, if it meets the preset value requirement, the output temperature consistency is normal, continue the workpiece heating process, monitor the temperature consistency description parameter until the workpiece heating process is completed, and generate a temperature control performance analysis instruction; if it does not meet the preset value requirement, the output temperature consistency is abnormal, prompt the user that the heating furnace is abnormal, and interrupt the workpiece heating process; Step S16, executing the temperature control performance analysis instruction, when the temperature uniformity description parameter wj and the temperature consistency description parameter wy both meet the corresponding preset values, jointly analyzing the temperature uniformity description parameter wj and the temperature consistency description parameter wy to obtain the temperature control performance index WK; The temperature control performance index WK is obtained as follows: By formula Get the weight coefficient q of each collection point c , where l sc Indicates the distance between the collection point and the heating position of the forging, k w is a scaling factor that adjusts the sensitivity of weight changes with distance; Assume that there are P collection points at the same time, q represents the number of the collection points, and the corrected temperature of the qth collection point is recorded as Xw cq , the average corrected temperature at different locations at the same time satisfies the formula By formula The temperature uniformity description parameter wj is calculated; Assume that K time points are monitored, s is used to represent the number of the time point, and FT is used s Indicates the temperature at the sth time point in the corrected temperature real-time change curve, expressed as FY s It represents the preset temperature at the sth time point in the preset temperature change curve, through the formula The temperature consistency description parameter wy is calculated; By formula The temperature control performance index WK is calculated, where α1 and α2 represent the weight coefficients of each item, 0<α1<1, 0<α2<1, and α1+α2=1.0; The forging process monitoring module is used to collect the rolling force, rolling speed, rolling friction resistance and rolling temperature parameters in the forging process and transmit them to the forging control performance analysis module; The forging control performance analysis module inputs the collected rolling force, rolling speed, rolling friction resistance and rolling temperature parameters into the forging control performance analysis model and outputs the forging control performance index DK of the forging process; The operation process of the forging control performance analysis model includes the following steps: Step 21, execute the forging instruction, monitor the rolling temperature and rolling friction resistance in real time during forging, and when the rolling temperature and rolling friction resistance are abnormal, warn the user, interrupt the forging process, and adjust the rolling temperature and rolling friction resistance until the requirements are met; Step 22, obtaining rolling force control parameters, comparing the actual rolling force curve with the preset rolling force curve, calculating the square of the rolling force difference at each time point, accumulating and summing all time points and averaging them to obtain the rolling force control parameters; Step 23, obtaining rolling speed control parameters, comparing the actual rolling speed curve with the preset rolling speed curve, calculating the square of the rolling speed difference at each time point, accumulating and summing all time points and averaging them to obtain the rolling speed control parameters; Step 24, jointly analyzing the rolling force control parameter, the rolling speed control parameter, the rolling temperature parameter and the abnormal description parameter of the rolling friction resistance, and calculating the forging control performance index DK; The forging control performance index DK is obtained by: The forging control performance index DK is calculated, where Zf represents the rolling force control parameter, Zv represents the rolling speed control parameter, Zwd represents the rolling temperature parameter, the average rolling friction resistance is recorded as Mf, and the preset rolling friction resistance is recorded as Mf 预 , the abnormal description parameters of rolling friction resistance are calculated β1, β2, β3, and β4 represent the proportional coefficients of each item, β1+β2+β3+β4=1.0; A production process control module receives the temperature control performance index and the forging control performance index, and takes control measures based on the relationship between the temperature control performance index and the forging control performance index and the corresponding preset values; The production status monitoring module is used to obtain the actual unit workload production efficiency and actual unit workload energy consumption in each production batch, and transmit them to the production status analysis module; The production status analysis module inputs the actual unit workload production efficiency and the actual unit workload energy consumption into the production status analysis model, and outputs the production efficiency abnormality evaluation index XNY of the production batch; The production status analysis model satisfies the formula Among them, XNY represents the production efficiency abnormality evaluation index, Sxl represents the actual unit workload production efficiency, Sxn represents the actual unit workload energy consumption, Yxl represents the preset unit workload production efficiency, Yxn represents the preset unit workload energy consumption, γ1, γ2, γ3 represent the proportional coefficients of each item respectively, and γ1+γ2+γ3=1.

0.

2. According to the management system of the forging intelligent production line based on the Internet of Things in claim 1, it is characterized in that: When the temperature control performance index WK is not lower than the preset value, an alert is given to the user, indicating that the heating process is abnormal, heating is interrupted, and the workpiece in the heating process does not enter the next production step; when the temperature control performance index WK is lower than the preset value, it indicates that the heating process is normal. After the heating process is completed, the workpiece in the heating process enters the next production step; when the forging control performance index DK is not lower than the preset value, an alert is given to the user, indicating that the forging process is abnormal, forging is interrupted, and the workpiece in the forging process does not enter the next production step; when the forging control performance index DK is lower than the preset value, it indicates that the forging process is normal. After the forging process is completed, the workpiece in the forging process enters the next production step.

3. According to the management system of the forging intelligent production line based on the Internet of Things in claim 1, it is characterized in that: The system further comprises: Point cloud data acquisition module: a machine vision device is installed at the output end of the production batch to collect point cloud data of the finished forgings in real time, transmit the collected point cloud data to the appearance defect recognition model, and transmit the output defect location and defect type to the forging appearance quality analysis module; The forging appearance quality analysis module inputs the appearance defect size and appearance defect type of each batch of forging finished products into the appearance quality assessment model, and outputs the appearance quality assessment index WYp of the production batch. Assuming that the production batch contains N forging finished products, i represents the number of the forging finished products, and the appearance quality assessment model satisfies the formula Among them, QM_i represents the appearance defect area of ​​the i-th forging finished product, y_i represents the influencing factor of the appearance defect type of the i-th forging finished product, and Mi_i represents the area of ​​the i-th forging finished product.

4. According to the management system of the forging intelligent production line based on the Internet of Things in claim 3, it is characterized in that: The system further comprises: Forging performance risk management module: Based on the appearance quality evaluation index WYp, production efficiency abnormality evaluation index XNY, temperature control performance index WK and forging control performance index DK of each production batch, a joint analysis is performed to obtain the performance abnormality risk coefficient of each batch, and measures are taken based on the relationship between the performance abnormality risk coefficient and the preset value.

5. According to the management system of the forging intelligent production line based on the Internet of Things according to claim 4, it is characterized in that: The performance abnormality risk coefficient Fxn is obtained by linearly normalizing the production efficiency abnormality evaluation index, the appearance quality evaluation index, the temperature control performance index and the forging control performance index and recording them as: XNY", WYp", WK", DK", and calculating them by the formula

Images