A temperature control system for a stainless steel welded steel pipe heat treatment furnace

Abstract

The invention discloses a temperature control system for a stainless steel welded steel pipe heat treatment furnace, which belongs to the technical field of stainless steel heat treatment, and specifically comprises: a temperature sensor, which is used to measure the temperature in the heat treatment furnace in real time and output a temperature signal; a data acquisition module, which is used to convert the temperature signal into a digital signal; a controller, which is used to receive the digital signal, calculate the temperature difference between the actual temperature in the heat treatment furnace and the set temperature, and calculate the heat difference according to the temperature difference, the controller screens parameters representing the degree of equipment aging, and stores a prediction model of the relationship between the heat deviation and the parameters of the degree of equipment aging, and the controller calculates the heat deviation value of the heat treatment furnace according to the prediction model; the controller adjusts the control signal output to the heating device according to the heat deviation, and adjusts the heating power of the heat treatment furnace; a visual interface, which is used to set the stainless steel process type, heating power, and display the real-time temperature and system working status.

Description

Technical Field

[0001] The invention relates to the technical field of stainless steel heat treatment, and in particular to a temperature control system of a stainless steel welded steel pipe heat treatment furnace. Background Art

[0002] In the production process of stainless steel welded steel pipes, heat treatment is an extremely critical and indispensable core process. After the stainless steel welded steel pipe is welded and formed, its internal structure will become complex and uneven due to the high temperature heat generated during the welding process, and internal stress will also accumulate in large quantities. The precise implementation of heat treatment can cleverly control the temperature strictly, prompting the atoms inside the steel pipe to rearrange and recombine, effectively eliminating welding residual stress, so that the steel pipe can obtain an ideal structure.

[0003] However, in the actual production process, as the heat treatment furnace equipment is used for a long time and at a high intensity day after day and year after year, the aging of the equipment gradually emerges like an unstoppable trend. After a lot of practical observation and data statistical analysis, it is found that the degree of equipment aging and the deviation of the temperature control of the heat treatment furnace show a significant positive correlation, that is, the more serious the equipment aging, the greater the temperature control deviation.

[0004] In actual operation scenarios, operators often set the heating power according to the ideal process parameters, hoping that the heat treatment furnace can quickly reach and stably maintain the target temperature. However, due to the temperature control deviation caused by equipment aging, the actual temperature in the furnace is often significantly different from the preset value. Once a temperature deviation occurs, the operator has to rely on experience to slowly and carefully adjust the heating power. Each adjustment requires a lot of time to observe temperature changes and repeatedly try different power settings, which undoubtedly greatly consumes the operator's energy and time costs.

[0005] Moreover, this frequent and imprecise temperature adjustment method seriously threatens the stability of product quality. Because the heat treatment is carried out in an environment with frequent temperature fluctuations, the internal structure of the stainless steel welded steel pipe is difficult to transform and optimize in the expected way, resulting in large differences in quality and uneven performance of each batch or even the same batch of steel pipes. Summary of the invention

[0006] The purpose of the present invention is to provide a temperature control system for a stainless steel welded steel pipe heat treatment furnace to solve the following technical problems:

[0007] However, due to the temperature control deviation caused by the aging of the equipment, the actual temperature in the furnace is often quite different from the preset value. Each adjustment requires a lot of time to observe the temperature changes, which consumes the energy and time cost of the operators, and leads to large differences in the quality of each batch or even the same batch of steel pipes.

[0008] The purpose of the present invention can be achieved through the following technical solutions:

[0009] A temperature control system for a stainless steel welded steel pipe heat treatment furnace, comprising:

[0010] Temperature sensor, used to measure the temperature in the heat treatment furnace in real time and output temperature signal;

[0011] A data acquisition module, connected to the temperature sensor, converting the temperature signal into a digital signal;

[0012] A controller connected to the data acquisition module receives the digital signal, calculates the temperature difference between the actual temperature and the set temperature in the heat treatment furnace, and calculates the heat difference according to the temperature difference. The controller selects parameters indicating the degree of equipment aging, and stores a prediction model of the relationship between the heat deviation and the parameters of the degree of equipment aging. The controller calculates the heat deviation value of the heat treatment furnace according to the prediction model.

[0013] A heating device connected to the controller, wherein the controller adjusts a control signal output to the heating device according to the heat deviation value to adjust the heating power of the heat treatment furnace;

[0014] A visual interface is connected to the controller and is used to set the stainless steel process type, heating power, and display real-time temperature and system working status.

[0015] As a further solution of the present invention: In the controller, the process of calculating the temperature difference between the actual temperature and the set temperature in the heat treatment furnace, and calculating the heat difference according to the temperature difference is as follows:

[0016] Obtain the temperature difference ΔT between the set temperature value and the actual temperature value of each heat treatment process of the same type, and calculate the heat difference ΔQ based on the temperature difference ΔT. The calculation formula is:

[0017] ;

[0018] Where N represents the number of different types of materials in the heat treatment furnace, c k represents the specific heat capacity of the kth material in the heat treatment furnace, k∈[1,N], m ik Indicates the mass of the kth material during the operation of the i-th heat treatment furnace, ΔT i Represents the temperature difference when the i-th heat treatment furnace is working.

[0019] As a further solution of the present invention: in the controller, the process of obtaining the parameter indicating the degree of aging of the equipment is as follows:

[0020] The equipment index and heat difference ΔQ of the heat treatment furnace are collected each time it works. The number of collections is marked as n, and the number of indexes is marked as m. A linear fit is performed on the relationship between the heat difference and the change of any equipment index, and the fitting degree R of the linear fit is calculated. 2 The calculation formula is:

[0021] ;

[0022] In the formula, represents the heat difference fitting value corresponding to the jth index when the i-th heat treatment furnace is working, i∈[1,n], j∈[1,m], y i is the actual value of the heat difference when the i-th heat treatment furnace is working, It indicates the average value of heat difference each time the heat treatment furnace works, R 2 ∈[-∞,1]; obtain the M indicators whose fitting degree is closest to 1, where M is the set value and marked as the parameter representing the degree of equipment aging.

[0023] As a further solution of the present invention: the process of the prediction model of the relationship between the heat deviation and the aging degree parameter is:

[0024] Normalize the equipment aging parameter data and the heat difference data, and divide the data into a training set and a validation set;

[0025] A prediction model is constructed based on a convolutional neural network and a long short-term memory network. The equipment aging degree parameters are input into the convolutional neural network to extract the equipment aging degree features. The features are nonlinearly transformed through an activation function. The features are downsampled using a maximum pooling layer. The extracted feature sequence is input into the long short-term memory network to extract the dynamic relationship between the equipment aging degree parameters and the heat difference. The feature vector output by the long short-term memory network is input into the fully connected layer. The features are extracted through a nonlinear transformation using an activation function to generate a prediction model.

[0026] The data in the validation set is input into the prediction model to obtain the predicted value of the temperature deviation. The value of the loss function is calculated based on the predicted value and the true value, and the model is updated based on the gradient backpropagation of the loss function until the loss function value converges.

[0027] As a further solution of the present invention: in the controller, different heat treatment processes of the stainless steel pipe correspond to different prediction models, and each prediction model is generated by training according to the equipment aging degree parameters and temperature difference during the corresponding heat treatment process.

[0028] As a further solution of the present invention: the process in which the controller calculates the heat deviation value of the heat treatment furnace according to the prediction model is:

[0029] Get the currently set stainless steel process type and target temperature, select the prediction model corresponding to the current process, and get the equipment aging degree parameters at the current moment, input the equipment aging degree parameters into the corresponding prediction model, and output the corresponding heat deviation value.

[0030] As a further solution of the present invention: the controller adjusts the control signal output to the heating device according to the heat deviation value to adjust the heating power of the heat treatment furnace in the following process:

[0031] The mass of the stainless steel pipe to be treated in the heat treatment furnace, as well as the current temperature in the heat treatment furnace and the target temperature are obtained. According to the mass of the stainless steel pipe, the temperature in the furnace and the target temperature, the heat Q1 required to make the stainless steel pipe to be treated reach the target temperature is calculated, and the duration t0 of this heat treatment process is obtained. The heating power of the heat treatment furnace is set to (Q1+heat deviation value) / t0.

[0032] As a further solution of the present invention: when the temperature setting curve of the current heat treatment process is nonlinear, the heating power of the heat treatment furnace is dynamically adjusted according to the temperature setting curve.

[0033] As a further solution of the present invention: the temperature sensor is based on a thermocouple or an infrared thermometer, and a plurality of temperature sensors are arranged in the heat treatment furnace.

[0034] Beneficial effects of the present invention:

[0035] (1) The present invention measures the temperature in the furnace in real time through a temperature sensor, and the data acquisition module digitizes the signal and transmits it to the controller. The controller can accurately calculate the difference between the actual temperature and the set temperature, and calculate the heat difference based on this, innovatively screen out the parameters representing the degree of equipment aging, and establish a prediction model for the relationship between the heat deviation and the equipment aging degree parameters. Through this model, the controller can calculate the corresponding heat deviation value according to the aging of the equipment, and then accurately adjust the heating power. This design fully considers the aging problem of the equipment after long-term use. Even when the temperature control deviation increases due to equipment aging, the temperature control accuracy of the heat treatment furnace can be guaranteed, which greatly extends the effective service life of the equipment and reduces the cost of frequent equipment replacement or large-scale maintenance due to equipment aging.

[0036] (2) The present invention has independently trained prediction models for different heat treatment processes of stainless steel pipes. The controller can select the corresponding prediction model according to the currently set stainless steel process type, and calculate the heat deviation value in combination with the current equipment aging degree parameter. This ensures that the system can adapt to a variety of different heat treatment process requirements. Regardless of the process requirements, accurate temperature control can be achieved, which greatly improves the versatility and adaptability of the system and meets the diverse process requirements in the production process of stainless steel welded pipes. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The present invention will be further described below in conjunction with the accompanying drawings.

[0038] Figure 1 It is a module schematic diagram of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] See also Figure 1 As shown, the present invention is a temperature control system for a stainless steel welded steel pipe heat treatment furnace, comprising:

[0041] Temperature sensor, the temperature sensor is a key component of the temperature control system of the stainless steel welded steel pipe heat treatment furnace, which is used to measure the temperature in the heat treatment furnace in real time and output the temperature signal. The temperature sensor is based on a thermocouple or an infrared thermometer, and multiple temperature sensors are arranged in the heat treatment furnace to ensure the accuracy and comprehensiveness of the temperature measurement.

[0042] A data acquisition module, connected to the temperature sensor, converting the temperature signal into a digital signal;

[0043] The controller is the core component of the temperature control system of the stainless steel welded steel pipe heat treatment furnace. It is connected to the data acquisition module and receives the digital signal output by the temperature sensor and converted by the data acquisition module. The main function of the controller is to calculate the temperature difference between the actual temperature and the set temperature in the heat treatment furnace, and calculate the heat difference based on the difference, so as to achieve precise control of the heating power of the heat treatment furnace.

[0044] Temperature difference and heat difference calculation

[0045] The controller first obtains the temperature difference ΔT between the set temperature value and the actual temperature value of each heat treatment process of the same type. Based on the temperature difference ΔT, the controller calculates the heat difference ΔQ, and the calculation formula is as follows:

[0046] ;

[0047] Where N represents the number of different types of materials in the heat treatment furnace, c k represents the specific heat capacity of the kth material in the heat treatment furnace, k∈[1,N], m ik Indicates the mass of the kth material during the operation of the i-th heat treatment furnace, ΔT i Represents the temperature difference when the i-th heat treatment furnace is working.

[0048] The controller screens the parameters representing the degree of equipment aging, and the screening process is:

[0049] Data collection:

[0050] The controller first collects the equipment indicators and heat difference ΔQ of the heat treatment furnace each time it works. These equipment indicators may include the resistance of the heating element, the insulation performance of the furnace, the speed of the fan, the life of the equipment and the operating time, which reflect the actual status of the equipment during the working process.

[0051] The number of collections is marked as n, and the number of indicators is marked as m. Through long-term monitoring and recording, the controller can accumulate a large amount of equipment indicator data, providing a basis for subsequent analysis and screening.

[0052] Linear Fit:

[0053] For each device index, the controller performs a linear fit on the relationship between the temperature difference and the index. The purpose of the linear fit is to find out the linear relationship between the device index and the temperature difference, that is, how the temperature difference changes with the change of the device index.

[0054] The result of the linear fit is a linear equation that describes the relationship between the device index and the temperature difference. With this equation, the controller can predict the expected value of the temperature difference under a given device index.

[0055] Goodness of fit calculation:

[0056] For each linear fit, the controller calculates its fit R 2 . Goodness of fit R 2 Is a statistic used to measure the accuracy of the linear fit. Its value ranges from −∞ to 1, and the closer the value is to 1, the more accurate the linear fit is.

[0057] Goodness of fit R 2 The calculation formula is as follows:

[0058] ;

[0059] In the formula, represents the heat difference fitting value corresponding to the jth index when the i-th heat treatment furnace is working, i∈[1,n], j∈[1,m], y i is the actual value of the heat difference when the i-th heat treatment furnace is working, It indicates the average value of heat difference each time the heat treatment furnace works, R 2 ∈[-∞,1]; obtain the M indicators whose fitting degree is closest to 1, where M is the set value and marked as the parameter representing the degree of equipment aging.

[0060] The controller stores a prediction model of the relationship between heat deviation and equipment aging parameters. The construction process is as follows:

[0061] Before building the prediction model, the equipment aging parameter data and heat difference data are first normalized. The purpose of normalization is to convert data of different dimensions and magnitudes to the same scale, usually normalizing the data to the range of [0,1] or [-1,1]. This step helps improve the training efficiency and convergence speed of the model.

[0062] The normalized data is divided into a training set and a validation set. The training set is used to train the model, and the validation set is used to evaluate the performance of the model and adjust the model parameters. Usually, the ratio of the training set to the validation set can be set to 80% and 20%, or 70% and 30%.

[0063] The device aging parameters are input into the convolutional neural network. CNN extracts the features of the device aging through the convolution layer and the pooling layer. The convolution layer is used to extract local features, and the pooling layer is used to downsample, reduce the dimension of the features, and improve the robustness of the model.

[0064] The pooling layer usually uses the maximum pooling (MaxPooling), which is used to retain the maximum value in the feature map and reduce the size of the feature map.

[0065] The extracted feature sequence is input into the long short-term memory network. LSTM is a special recurrent neural network (RNN) that can effectively handle long-term dependency problems in time series data. LSTM controls the flow of information through a gating mechanism, including input gate, forget gate, and output gate.

[0066] The feature vector output by LSTM is input to the fully connected layer. The fully connected layer performs nonlinear transformation through the activation function, extracts features, and generates the final prediction value. The activation function of the fully connected layer usually uses ReLU or softmax, and the specific choice depends on the type of output layer.

[0067] Input the data in the validation set into the prediction model to obtain the predicted value of temperature deviation. By comparing the predicted value with the true value, the value of the loss function is calculated. Commonly used loss functions include mean square error (MSE) and cross entropy loss (Cross-EntropyLoss)

[0068] The parameters of the model are updated based on the gradient back propagation of the loss function. Gradient back propagation updates the parameters by calculating the partial derivative of the loss function with respect to each parameter, so that the loss function value gradually decreases. Commonly used optimization algorithms include stochastic gradient descent (SGD), Adam, and RMSprop.

[0069] Through multiple iterations of training, the loss function value converges. The convergence criterion can be that the change in the loss function value is less than a certain threshold, or that the preset number of iterations is reached. Usually, the training process will set an early stopping mechanism. When the loss function value on the validation set no longer decreases in multiple consecutive iterations, the training is stopped to prevent overfitting.

[0070] In the controller, different heat treatment processes of stainless steel pipes correspond to different prediction models. Each prediction model is generated based on the equipment aging degree parameters and temperature difference training during the corresponding heat treatment process. The specific steps are as follows:

[0071] Process type identification: Get the currently set stainless steel process type and target temperature.

[0072] Model selection: Select the prediction model corresponding to the current process.

[0073] Parameter acquisition: Get the device aging degree parameters at the current moment.

[0074] Heat deviation calculation: Input the equipment aging degree parameters into the corresponding prediction model and output the corresponding heat deviation value.

[0075] In the temperature control system of the stainless steel welded steel pipe heat treatment furnace, the controller has a highly intelligent function and can accurately control the temperature and adjust the heat according to different heat treatment process requirements. Specifically, this process involves the following key steps:

[0076] Process type identification:

[0077] The controller first obtains the currently set stainless steel process type through the visual interface or the preset process parameter library. Different heat treatment processes, such as annealing, normalizing, quenching, tempering, etc., have different requirements for temperature control. For example, the annealing process may require a lower heating temperature and a longer holding time, while the quenching process requires rapid heating to a high temperature and rapid cooling.

[0078] By identifying the current process type, the controller can call the control strategy and parameter settings that match the process to ensure the accuracy and consistency of the heat treatment process.

[0079] Target temperature acquisition:

[0080] Accompanying the process type is the target temperature, which is the precise temperature that needs to be reached during the heat treatment process. The target temperature is set based on the characteristics of the material and the desired final properties. For example, for some stainless steel materials, the target temperature for solution treatment may be between 1050°C and 1150°C, while the target temperature for stress relief annealing may be between 650°C and 700°C.

[0081] Accurate setting of the target temperature is crucial to ensuring the heat treatment effect. The controller allows operators to make adjustments and confirmations according to specific needs by interacting with the visual interface.

[0082] Model library call:

[0083] The controller stores multiple prediction models, each corresponding to a specific heat treatment process. These models are generated based on a large amount of historical data and advanced machine learning algorithms (such as convolutional neural networks and long short-term memory networks) to accurately predict the impact of equipment aging on thermal deviation under different process conditions.

[0084] When the controller recognizes the current process type, it will automatically select the prediction model that matches the process from the model library. This process ensures that the controller can use the most appropriate model to calculate the heat deviation in different heat treatment processes.

[0085] Real-time data collection:

[0086] The controller is connected to various sensors and monitoring equipment of the heat treatment furnace to collect the aging parameters of the equipment in real time. These parameters may include the resistance change of the heating element, the decrease of the insulation performance of the furnace, the decrease of the efficiency of the fan, etc. These factors will affect the heating efficiency and temperature uniformity of the heat treatment furnace.

[0087] Through regular equipment inspections and online monitoring, the controller can obtain the latest equipment aging parameters, which reflect the actual performance of the equipment under the current working status.

[0088] Parameter screening and processing:

[0089] The controller screens and processes the collected equipment aging parameters, removes invalid or abnormal data, and ensures that the parameters input into the prediction model are accurate and reliable. This process involves steps such as data cleaning and normalization to improve the accuracy and stability of the prediction model.

[0090] Model input and calculation:

[0091] The processed equipment aging parameters are input into the selected prediction model. The prediction model extracts the characteristics of equipment aging through convolutional neural networks, and then analyzes the dynamic relationship between these characteristics and heat deviation through long short-term memory networks, and finally outputs the corresponding heat deviation value.

[0092] The calculation of the heat deviation value takes into account the impact of equipment aging on the heat treatment process, and can dynamically adjust the heating power to ensure that the temperature in the heat treatment furnace can accurately reach the target temperature.

[0093] Control signal generation: Based on the calculated heat deviation value, the controller generates a corresponding control signal to adjust the power of the heating device. If the heat deviation value is positive, it means that the actual temperature is lower than the target temperature, and the controller will increase the heating power; conversely, if the heat deviation value is negative, it means that the actual temperature is higher than the target temperature, and the controller will reduce the heating power.

[0094] Through this dynamic adjustment, the controller can compensate for the impact of equipment aging in real time, ensure the temperature control accuracy and stability of the heat treatment process, and thus improve the heat treatment quality of stainless steel welded pipes.

[0095] The heating device is a key executive component in the temperature control system of the stainless steel welded steel pipe heat treatment furnace and is closely connected to the controller. The controller adjusts the control signal output to the heating device according to the calculated heat deviation value, thereby achieving precise control of the heating power of the heat treatment furnace. The specific control process is as follows:

[0096] Get the quality of stainless steel pipe to be processed:

[0097] The controller first obtains the mass m of the stainless steel pipe to be treated in the heat treatment furnace. This parameter can be obtained in real time through a weighing sensor or other measuring equipment to ensure the accuracy and real-time nature of the data.

[0098] Get the current furnace temperature and target temperature:

[0099] The controller obtains the actual temperature in the current heat treatment furnace from the temperature sensor and obtains the target temperature from the process parameter library or the operator input.

[0100] Calculate the calories needed:

[0101] According to the mass m of the stainless steel pipe and the current furnace temperature T current and target temperature T target , the controller calculates the heat Q1 required to make the stainless steel pipe to be treated reach the target temperature. The calculation formula is as follows:

[0102] Q1=mc(Ttarge t−T current );

[0103] Here, c is the specific heat capacity of stainless steel, which is usually 0.5 J / (g·℃).

[0104] Get the heat treatment process duration:

[0105] The controller obtains the duration t0 of the heat treatment process from the process parameter library or the operator input. This parameter is set according to the specific heat treatment process requirements to ensure the integrity and effectiveness of the heat treatment process.

[0106] Calculate the heating power:

[0107] The controller calculates the heating power P of the heat treatment furnace based on the required heat Q1 and the heat deviation value, as well as the heat treatment process duration t0. The calculation formula is as follows:

[0108] P=(Q1+heat deviation value)t0;

[0109] In this way, the controller can dynamically adjust the power of the heating device to ensure that the temperature in the heat treatment furnace can accurately reach the target temperature.

[0110] Identify non-linear temperature setpoint curves:

[0111] The controller recognizes when the temperature setpoint profile of the current heat treatment process is nonlinear. Nonlinear temperature setpoint profiles are often used in complex heat treatment processes such as multi-stage heating, holding and cooling processes that require precise temperature control and dynamic adjustments.

[0112] Dynamic adjustment of heating power:

[0113] The controller adjusts the power of the heating device in real time according to the nonlinear temperature setting curve. The specific steps are as follows:

[0114] Segment processing: The nonlinear temperature setting curve is divided into multiple time periods, and the temperature change pattern in each time period is relatively simple.

[0115] Real-time calculation: In each time period, the controller calculates the required heat deviation value based on the temperature set value and actual temperature value at the current time point.

[0116] Power adjustment: Based on the calculated heat deviation value, the controller dynamically adjusts the power of the heating device to ensure that the temperature in the heat treatment furnace can accurately change according to the set curve.

[0117] Feedback and corrections:

[0118] The controller monitors the temperature changes in the heat treatment furnace in real time, and provides feedback and corrections to the heating power. If there is a deviation between the actual temperature and the set temperature, the controller will adjust the heating power in time to ensure the accuracy and stability of temperature control.

[0119] Through the above process, the heating device, under the precise control of the controller, can dynamically adjust the heating power according to different heat treatment processes and equipment aging degrees, to ensure that the temperature in the heat treatment furnace can accurately reach the target temperature.

[0120] A visual interface is connected to the controller and is used to set the stainless steel process type, heating power, and display real-time temperature and system working status.

[0121] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data for software simulation. The preset parameters and thresholds in the formula are selected by technicians in this field according to actual conditions. The above is only a specific implementation method of the present application, but the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims. An embodiment of the present invention is described in detail above, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equal changes and improvements made according to the scope of application of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A temperature control system for a stainless steel welded steel pipe heat treatment furnace, characterized in that: include: Temperature sensor, used to measure the temperature in the heat treatment furnace in real time and output temperature signal; A data acquisition module, connected to the temperature sensor, converting the temperature signal into a digital signal; A controller connected to the data acquisition module receives the digital signal, calculates the temperature difference between the actual temperature and the set temperature in the heat treatment furnace, and calculates the heat difference according to the temperature difference. The controller selects parameters indicating the degree of equipment aging, and stores a prediction model of the relationship between the heat deviation and the parameters of the degree of equipment aging. The controller calculates the heat deviation value of the heat treatment furnace according to the prediction model. A heating device connected to the controller, wherein the controller adjusts a control signal output to the heating device according to the heat deviation value to adjust the heating power of the heat treatment furnace; A visual interface, connected to the controller, for setting the stainless steel process type, heating power, and displaying the real-time temperature and system working status; The controller adjusts the control signal output to the heating device according to the heat deviation value, and the process of adjusting the heating power of the heat treatment furnace is as follows: The mass of the stainless steel pipe to be treated in the heat treatment furnace, as well as the current temperature in the heat treatment furnace and the target temperature are obtained. According to the mass of the stainless steel pipe, the temperature in the furnace and the target temperature, the heat Q1 required to make the stainless steel pipe to be treated reach the target temperature is calculated, and the duration t0 of this heat treatment process is obtained. The heating power of the heat treatment furnace is set to (Q1+heat deviation value) / t0.

2. The temperature control system of a stainless steel welded steel pipe heat treatment furnace according to claim 1, characterized in that: In the controller, the process of calculating the temperature difference between the actual temperature and the set temperature in the heat treatment furnace and calculating the heat difference according to the temperature difference is as follows: Obtain the temperature difference ΔT between the set temperature value and the actual temperature value of each heat treatment process of the same type, and calculate the heat difference ΔQ based on the temperature difference ΔT. The calculation formula is: ； Where N represents the number of different types of materials in the heat treatment furnace, c k represents the specific heat capacity of the kth material in the heat treatment furnace, k∈[1,N], m ik Indicates the mass of the kth material during the operation of the i-th heat treatment furnace, ΔT i Represents the temperature difference when the i-th heat treatment furnace is working.

3. The temperature control system of a stainless steel welded steel pipe heat treatment furnace according to claim 2, characterized in that: In the controller, the screening process of the parameter indicating the degree of equipment aging is as follows: The equipment index and heat difference ΔQ of the heat treatment furnace are collected each time it works. The number of collections is marked as n, and the number of indexes is marked as m. A linear fit is performed on the relationship between the heat difference and the change of any equipment index, and the fitting degree R of the linear fit is calculated. 2 The calculation formula is: ； In the formula, represents the heat difference fitting value corresponding to the jth index when the i-th heat treatment furnace is working, i∈[1,n], j∈[1,m], y i is the actual value of the heat difference when the i-th heat treatment furnace is working, It indicates the average value of heat difference each time the heat treatment furnace works, R 2 ∈[-∞,1]; obtain the M indicators whose fitting degree is closest to 1, where M is the set value and marked as the parameter representing the degree of equipment aging.

4. The temperature control system of a stainless steel welded steel pipe heat treatment furnace according to claim 3, characterized in that: The construction process of the prediction model of the relationship between thermal deviation and aging degree parameters is as follows: Normalize the equipment aging parameter data and the heat difference data, and divide the data into a training set and a validation set; A prediction model is constructed based on a convolutional neural network and a long short-term memory network. The equipment aging degree parameters are input into the convolutional neural network to extract the equipment aging degree features. The features are nonlinearly transformed through an activation function. The features are downsampled using a maximum pooling layer. The extracted feature sequence is input into the long short-term memory network to extract the dynamic relationship between the equipment aging degree parameters and the heat difference. The feature vector output by the long short-term memory network is input into the fully connected layer. The features are extracted through a nonlinear transformation using an activation function to generate a prediction model. The data in the validation set is input into the prediction model to obtain the predicted value of the temperature deviation. The value of the loss function is calculated based on the predicted value and the true value, and the model is updated based on the gradient backpropagation of the loss function until the loss function value converges.

5. The temperature control system of a stainless steel welded steel pipe heat treatment furnace according to claim 2, characterized in that: In the controller, different heat treatment processes of the stainless steel pipe correspond to different prediction models, and each prediction model is generated by training according to equipment aging degree parameters and temperature difference during the corresponding heat treatment process.

6. The temperature control system of a stainless steel welded steel pipe heat treatment furnace according to claim 5, characterized in that: The process of the controller calculating the heat deviation value of the heat treatment furnace according to the prediction model is as follows: Get the currently set stainless steel process type and target temperature, select the prediction model corresponding to the current process, and get the equipment aging degree parameters at the current moment, input the equipment aging degree parameters into the corresponding prediction model, and output the corresponding heat deviation value.

7. The temperature control system of a stainless steel welded steel pipe heat treatment furnace according to claim 1, characterized in that: When the temperature setting curve of the current heat treatment process is nonlinear, the heating power of the heat treatment furnace is dynamically adjusted according to the temperature setting curve.

8. The temperature control system of a stainless steel welded steel pipe heat treatment furnace according to claim 1, characterized in that: The temperature sensor is based on a thermocouple or an infrared thermometer, and a plurality of temperature sensors are arranged in the heat treatment furnace.

Images