Temperature self-tuning method in hot riveting process based on data acquisition and PID

Abstract

The application discloses a temperature self-tuning method in a hot riveting process based on data acquisition and PID, and belongs to the technical field of temperature control. The method comprises the following steps: acquiring real-time temperature digital quantities of a hot riveting head at multiple continuous sampling moments to form a real-time temperature sequence; calculating a temperature change acceleration and a temperature change jerk at the current sampling moment according to the real-time temperature sequence; determining adaptive adjustment amounts of a proportional coefficient, an integral coefficient and a differential coefficient of a PID controller based on the temperature change acceleration and the temperature change jerk; superimposing the adaptive adjustment amounts and corresponding current coefficient values respectively to obtain updated coefficients; and calculating an output control amount by the PID controller by using the updated coefficients to drive a heating actuator to adjust the temperature of the hot riveting head. The method dynamically adjusts PID parameters by using high-order change trends of the temperature, can adapt to complex fluctuations of thermal loads in the hot riveting process, and improves the response speed and steady-state accuracy of temperature control.

Description

Technical Field

[0001] This invention relates to the field of temperature control technology, specifically to a temperature self-tuning method for a hot riveting process based on data acquisition and PID control. Background Technology

[0002] Hot riveting is a relatively mature existing technology. In the manufacturing of battery modules (CCS), it specifically refers to the process of using heating to melt and reshape plastic rivets, thereby locking aluminum busbars, injection molded brackets, or FPC flexible circuit boards. Essentially, it is a "molecular-level" physical anchoring of insulating and conductive components, and it is the core technology of CCS module assembly.

[0003] During hot riveting, the temperature control accuracy of the riveting head directly affects the riveting quality and product consistency. Conventional solutions use a PID controller for closed-loop adjustment based on the deviation between the set target temperature and the feedback temperature. The proportional, integral, and derivative coefficients are typically fixed after being determined during the commissioning phase. However, the hot riveting head frequently comes into contact with workpieces of varying initial temperatures and heat capacities, resulting in a continuously fluctuating thermal load. A fixed-parameter PID controller struggles to balance rapid response and overshoot suppression under such variable load conditions. When the thermal load suddenly increases, the temperature recovery is slow; when the thermal load suddenly decreases, overshoot is likely. This decline in control quality can lead to inconsistent riveting strength or thermal damage to the workpiece surface. To improve temperature control under variable load conditions, some existing methods have introduced parameter self-tuning strategies. Some solutions use graded switching of the PID coefficients based on the amplitude or rate of change of the temperature deviation. While this simple rule-based approach can improve response to some extent, it only utilizes first-order deviation information and lacks the ability to perceive the acceleration trend and impact disturbances of temperature changes. When the temperature deviation has not yet increased significantly but the rate of change is rapidly escalating, the control system cannot intervene in advance, resulting in a lag in adjustment. Another approach uses online identification of the controlled object model for parameter optimization; however, the thermal contact state changes rapidly and exhibits significant nonlinear characteristics during the hot riveting process. The computational delay and accuracy limitations of model identification make it difficult for such methods to track the dynamic changes of thermal load in real-time in practical applications. Therefore, how to enable the controller to perceive higher-order dynamic trends of temperature changes and continuously fine-tune control parameters accordingly, while maintaining the simplicity of the PID controller structure, has become a problem to be solved. Summary of the Invention

[0004] This invention provides a temperature self-tuning method for hot riveting processes based on data acquisition and PID control. The aim is to generate adaptive adjustment values ​​for PID parameters in real time using the acceleration and jerk of temperature change. By continuously adjusting the proportional coefficient, integral coefficient, and derivative coefficient through superposition and updating, the controller can respond in advance to the accelerating and abrupt trends of temperature change when the thermal load fluctuates, thereby achieving a control effect that balances fast response and temperature stability under hot riveting variable load conditions.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A temperature self-tuning method based on data acquisition and PID control during the hot riveting process includes: The real-time digital temperature of the hot riveting head is acquired at multiple consecutive sampling moments using a temperature sensor and an analog-to-digital converter, forming a real-time temperature sequence. Based on this real-time temperature sequence, the temperature change acceleration and temperature change jerk at the current sampling moment are calculated. The temperature change acceleration reflects the rate of temperature change, while the temperature change jerk reflects the intensity of the acceleration change itself; both provide richer dynamic information for subsequent adaptive parameter adjustment. Based on the temperature change acceleration and jerk, the adaptive adjustment amounts of the proportional, integral, and derivative coefficients of the PID controller are determined. These adaptive adjustment amounts are then superimposed with their respective current coefficient values ​​to obtain updated proportional, integral, and derivative coefficients. Using these updated coefficients, the PID controller calculates and outputs a control quantity to drive the heating actuator to adjust the temperature of the hot riveting head.

[0006] This scheme utilizes temperature change acceleration and jerk to directly correct PID parameters online. During the temperature rise and fall and steady-state maintenance stages of the hot riveting process, it can self-tune control parameters according to the temperature curve shape, thereby effectively suppressing temperature overshoot and oscillation, accelerating dynamic response speed, and improving steady-state control accuracy.

[0007] In one embodiment of the present invention, when calculating the temperature change acceleration and temperature change jerk, the temperature value at the current sampling moment, the temperature value at the previous sampling moment, and the temperature value at the two previous sampling moments are obtained from the real-time temperature sequence; the first-order difference of the temperature at the current moment and the first-order difference of the temperature at the previous moment are calculated, and the difference between the two is used as the second-order difference of the temperature at the current moment, thereby characterizing the temperature change acceleration; the difference between the second-order difference of the temperature at the current moment and the second-order difference of the temperature at the previous sampling moment is calculated to obtain the third-order difference of the temperature at the current moment, thereby characterizing the temperature change jerk. This calculation method only requires temperature data from three consecutive sampling moments to obtain high-order dynamic characteristics in real time, with low computational load, and is suitable for the high-speed computing environment of embedded controllers.

[0008] In one embodiment of the present invention, when determining the adaptive adjustment amount of the PID coefficients, the following weighting coefficients are preset: proportional acceleration weighting coefficient, proportional jerk weighting coefficient, integral acceleration weighting coefficient, integral jerk weighting coefficient, derivative acceleration weighting coefficient, and derivative jerk weighting coefficient. The product of temperature change acceleration and the proportional acceleration weighting coefficient is used as the proportional acceleration adjustment component, and the product of temperature change jerk and the proportional jerk weighting coefficient is used as the proportional jerk adjustment component. The sum of these two constitutes the adaptive adjustment amount of the proportional coefficient. Similarly, the adaptive adjustment amounts of the integral coefficient and the derivative coefficient are obtained using the corresponding weighting coefficients. By reasonably configuring the six weighting coefficients, the influence of acceleration and jerk on each of the proportional, integral, and derivative channels can be adjusted independently, making parameter tuning more flexible and targeted.

[0009] Preferably, the initial values ​​of the weighting coefficients are set based on the thermal response time constant of the hot riveting head and the sampling period of the temperature sensor. Specifically, the thermal response time constant is obtained, the upper limit of the temperature change rate of the hot riveting head is determined, and the maximum temperature change within a single sampling period is calculated based on the sampling period. Using the maximum temperature change as a reference, the proportional acceleration weighting coefficient is set to the reciprocal of the reference value multiplied by a first preset coefficient, the proportional accelerometer weighting coefficient is set to the reciprocal of the square of the reference value multiplied by a second preset coefficient, and the integral acceleration weighting coefficient, integral accelerometer weighting coefficient, differential acceleration weighting coefficient, and differential accelerometer weighting coefficient are each set according to their respective preset coefficients. This setting method ensures that the weighting coefficients match the actual dynamic characteristics of the hot riveting head, and can obtain reasonable adaptive adjustment gain when using it for the first time or replacing it with a hot riveting head of different specifications, avoiding excessive or sluggish parameter adjustment.

[0010] In one embodiment of the present invention, the adaptive adjustment of the proportional coefficient is summed with the current value of the proportional coefficient to obtain the updated proportional coefficient; the adaptive adjustment of the integral coefficient is summed with the current value of the integral coefficient to obtain the updated integral coefficient; and the adaptive adjustment of the derivative coefficient is summed with the current value of the derivative coefficient to obtain the updated derivative coefficient. Furthermore, the updated proportional coefficient, integral coefficient, and derivative coefficient are each constrained within their respective preset parameter ranges to prevent abnormal parameter jumps caused by instantaneous disturbances and to ensure the stability of the control process.

[0011] In one embodiment of the present invention, when calculating the control quantity, the difference between the set target temperature value and the temperature value at the current sampling time is obtained as the temperature deviation value at the current time. Using the temperature deviation values ​​at the previous sampling time and the temperature deviation values ​​at two previous sampling times, the integral cumulative value and the first-order difference value of the temperature deviation at the current time are calculated. The updated proportional coefficient is multiplied by the temperature deviation value at the current time to obtain the proportional control component; the updated integral coefficient is multiplied by the integral cumulative value to obtain the integral control component; and the updated differential coefficient is multiplied by the first-order difference value to obtain the differential control component. These three are then superimposed to form the control quantity. This control quantity is limited before being output to the heating actuator. The upper and lower limits of the limit are set according to the rated power of the heating actuator to prevent the drive signal from exceeding the hardware capability range, thus protecting the actuator and the hot riveting head.

[0012] This invention extracts high-order variation features of temperature sequences online and introduces acceleration and jerk into the adaptive adjustment of PID coefficients. It can automatically adjust the control intensity at each stage of the hot riveting process, including heating, holding and cooling. While quickly following the target temperature, it reduces temperature overshoot and steady-state error, significantly improving the temperature control quality and process consistency of the hot riveting process. Moreover, it eliminates the need for repeated manual tuning of PID parameters, effectively improving equipment deployment efficiency and automation level.

[0013] The technical effects and advantages provided by the present invention in the above technical solution are as follows: The temperature change acceleration and temperature change jerk at the current sampling moment are calculated using real-time temperature sequences, and these two higher-order variables are used as the basis for adaptive adjustment of PID parameters. Temperature change acceleration reflects the changing trend of the temperature rise and fall rate, indicating whether the system is accelerating its heating or decelerating its cooling; temperature change jerk further reveals the variation in acceleration itself, corresponding to the inflection point of the rate of change caused by thermal load shocks or sudden changes in contact state. Introducing these two higher-order dynamic characteristics into the parameter adjustment process allows the controller to no longer rely solely on the amplitude and first-order rate of change of the temperature deviation, but to proactively adjust the proportional, integral, and derivative coefficients in the early stages, before the deviation has accumulated significantly but before the temperature curve shows an accelerating or shock trend. When the temperature change acceleration increases, the adjustment quantity changes accordingly, and the controller corrects the output in advance, suppressing overshoot or accelerating recovery; when the temperature change jerk shows a peak, the adjustment quantity quickly responds to the sudden change trend, avoiding control lag. This approach enables the PID controller to sense and respond to changes in dynamic temperature trends, effectively reducing temperature fluctuation amplitude and shortening the time to re-enter steady state when the hot riveting head contacts different workpieces, causing sudden changes in thermal load. The adaptive adjustment of each PID coefficient is combined with the corresponding current coefficient value through superposition, forming a continuous parameter update path. Unlike the method of directly switching to different fixed parameter groups based on preset conditions, superposition updates allow the parameter adjustment range of each sampling cycle to be dynamically determined by the high-order temperature change characteristics calculated in real time, resulting in a smoother coefficient change process. The proportional, integral, and derivative coefficients do not jump between a few preset levels, but evolve gradually along a trajectory guided by acceleration and jerk. When the temperature change tends to level off, the adjustment naturally decreases, and the parameters remain relatively stable; when the temperature change trend intensifies, the adjustment increases accordingly, and the parameters quickly adapt to the new thermal load state. This continuous parameter self-tuning mechanism avoids the control quantity abruptness that may be caused by parameter mutations, reduces the impact of actuator actions, and at the same time preserves the integrity and real-time performance of the original PID controller structure, achieving autonomous adaptation to complex thermal load disturbances with minimal computational overhead. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0015] Figure 1 This is a flowchart of the present invention; Figure 2 This is a flowchart of the PID temperature control acceleration and jerk weight coefficient setting for the hot riveting head; Figure 3 It is the adaptive adjustment curve of the proportional, integral, and derivative coefficients of the PID controller; Figure 4 It shows the real-time temperature change curve of the hot riveting head and a comparison chart of the temperature change rate and the upper limit of the rate. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] See Figure 1 This invention provides a temperature self-tuning method for a hot riveting process based on data acquisition and PID control. The method acquires real-time digital temperatures of the riveting head at multiple consecutive sampling moments, collected by a temperature sensor via an analog-to-digital converter, forming a real-time temperature sequence. Based on the real-time temperature sequence, the temperature change acceleration and temperature change jerk at the current sampling moment are calculated. Based on the temperature change acceleration and jerk, the adaptive adjustment values ​​of the proportional, integral, and derivative coefficients of the PID controller are determined. These adaptive adjustment values ​​are then superimposed with their corresponding current values ​​to obtain updated proportional, integral, and derivative coefficients. Using these updated coefficients, the PID controller calculates and outputs a control quantity to drive the heating actuator to adjust the temperature of the hot riveting head.

[0018] In practice, the real-time digital temperature of the hot riveting head is acquired by a temperature sensor via an analog-to-digital converter at multiple consecutive sampling moments. These real-time digital temperature values ​​constitute a real-time temperature sequence. The current sampling moment is denoted as sampling moment k. The real-time temperature sequence contains temperature values ​​from at least four consecutive sampling moments, denoted as the temperature value T(k) at the current sampling moment k, the temperature value T(k-1) at the previous sampling moment k-1, the temperature value T(k-2) at the second previous sampling moment k-2, and the temperature value T(k-3) at the third previous sampling moment k-3. All these temperature values ​​are obtained by directly reading the data from the real-time temperature sequence at the corresponding positions.

[0019] When calculating the acceleration of temperature change at the current sampling moment, the difference between the temperature value T(k) at the current sampling moment and the temperature value T(k-1) at the previous sampling moment is calculated to obtain the first-order difference of the temperature at the current moment. T(k), where T(k) = T(k) - T(k-1). Calculate the difference between the temperature value T(k-1) at the previous sampling time and the temperature value T(k-2) at the two previous sampling times to obtain the first-order temperature difference at the previous time. T(k-1), where, T(k-1) = T(k-1) - T(k-2). Calculate the first-order difference of the temperature at the current moment. The first-order difference between T(k) and the temperature at the previous moment The difference between T(k-1) yields the second-order temperature difference at the current moment. This second-order temperature difference is the acceleration of temperature change at the current sampling moment, denoted as a(k), and its expression is a(k) = T(k)- T(k-1)=T(k)-2T(k-1)+T(k-2).

[0020] Calculating the jerk of temperature change at the current sampling moment requires the second-order temperature difference from the previous sampling moment. The first-order temperature difference between the two previous sampling moments, T(k-2) and T(k-3), is obtained by calculating the difference between them. T(k-2), where, T(k-2) = T(k-2) - T(k-3). Calculate the first-order difference of temperature at the previous moment. T(k-1) and the first-order difference of the temperature at the previous two time points The difference between T(k-2) and T(k-2) yields the second-order temperature difference a(k-1) from the previous moment, expressed as a(k-1) = T(k-1)- T(k-2) = T(k-1) - 2T(k-2) + T(k-3). Then, calculate the difference between the second-order temperature difference a(k) at the current moment and the second-order temperature difference a(k-1) at the previous moment to obtain the third-order temperature difference at the current moment. The value of the third-order temperature difference at the current moment is the accelerometer of temperature change at the current sampling moment. The accelerometer of temperature change can be expressed by the following formula:

[0021] Where j(k) represents the accelerometer of temperature change at the current sampling time k, T(k) is the real-time digital temperature at the current sampling time k, T(k-1) is the real-time digital temperature at the previous sampling time k-1, T(k-2) is the real-time digital temperature at the second-to-last sampling time k-2, and T(k-3) is the real-time digital temperature at the third-to-last sampling time k-3. The real-time digital temperatures T(k), T(k-1), T(k-2), and T(k-3) are all obtained directly from the real-time temperature sequence. Each time a new sampling value is generated in the real-time temperature sequence, the existing temperature values ​​in the sequence shift one position to the past, and the newly acquired temperature value occupies the position of the current sampling time k.

[0022] It is understandable that the above calculation process is executed at each sampling time. In a digital signal processor or microcontroller, the temperature values ​​at four consecutive sampling times can be obtained from the real-time temperature sequence first, and then j(k) can be directly solved according to the above formula. The calculation result of the temperature change acceleration a(k) is retained and used as the second-order difference of the temperature at the next sampling time k+1 to participate in the solution of the temperature change acceleration. The specific values ​​of the temperature change acceleration and the temperature change jerk are dynamically updated as the real-time temperature sequence changes, reflecting the drastic degree of temperature change of the hot riveting head in real time.

[0023] In practice, the proportional acceleration weight coefficient, proportional jerk weight coefficient, integral acceleration weight coefficient, integral jerk weight coefficient, derivative acceleration weight coefficient, and derivative jerk weight coefficient are pre-generated and stored. These six weight coefficients are written into non-volatile memory during system initialization and are directly read by the processor each time adaptive adjustment calculations are performed. These weight coefficients are not modified online during the PID control program's execution.

[0024] After obtaining the temperature change acceleration and temperature change jerk at the current sampling moment, the following three sets of parallel or sequential multiply-accumulate operations are performed: When calculating the adaptive adjustment of the proportional coefficient, the temperature change acceleration is multiplied by the proportional acceleration weighting coefficient, and the product is used as the proportional acceleration adjustment component; the temperature change jerk is multiplied by the proportional jerk weighting coefficient, and the product is used as the proportional jerk adjustment component. The proportional acceleration adjustment components are then added together, and the sum is used as the adaptive adjustment of the proportional coefficient. When calculating the adaptive adjustment of the integral coefficient, the temperature change acceleration is multiplied by the integral acceleration weighting coefficient, and the product is used as the integral acceleration adjustment component; the temperature change jerk is multiplied by the integral jerk weighting coefficient, and the product is used as the integral jerk adjustment component. The integral acceleration adjustment components are then added together, and the sum is used as the adaptive adjustment of the integral coefficient. When calculating the adaptive adjustment amount of the differential coefficients, the temperature change acceleration is multiplied by the differential acceleration weighting coefficient, and the product is used as the differential acceleration adjustment component; the temperature change jerk is multiplied by the differential jerk weighting coefficient, and the product is used as the differential jerk adjustment component. The differential acceleration adjustment components are added together, and the sum is used as the adaptive adjustment amount of the differential coefficients.

[0025] In some implementations, the above three sets of operations can be described by a unified expression, where the acceleration of temperature change at the current sampling time is denoted as a, the acceleration of temperature change is denoted as j, and the adaptive adjustment amount of the proportional coefficient is... Adaptive adjustment amount of integral coefficient Adaptive adjustment amount of differential coefficients They are respectively:

[0026] in, This represents the proportional acceleration weighting coefficient. This represents the proportional accelerometer weighting coefficient. This represents the integral acceleration weighting coefficient. This represents the integral accelerometer weighting coefficient. This represents the differential acceleration weighting coefficient. This represents the differential acceleration weighting coefficient. 'a' is the acceleration due to temperature change calculated using the second-order difference of the current temperature, and 'j' is the acceleration due to temperature change calculated using the third-order difference of the current temperature. Proportional acceleration weighting coefficient. The numerical value of the acceleration due to temperature change 'a' is mapped to the acceleration-related adjustment contribution of the proportional gain, with the proportional accelerometer weighting coefficient as the key factor. The numerical value of the jerk j due to temperature change is mapped to the jerk-related adjustment contribution of the proportional coefficient, and the sum of the two constitutes the adaptive adjustment amount of the proportional coefficient. Integral acceleration weighting coefficient The numerical value of the acceleration 'a' due to temperature change is mapped to the acceleration-related adjustment contribution to the integral coefficient, and the integral acceleration weighting coefficient is used. The numerical value of the jerk j due to temperature change is mapped to the jerk-related adjustment contribution of the integral coefficient, and the sum of the two constitutes the adaptive adjustment amount of the integral coefficient. Differential acceleration weighting coefficient The numerical value of the acceleration 'a' due to temperature change is mapped to the acceleration-related adjustment contribution of the differential coefficient, and the differential acceleration weighting coefficient is used. The numerical value of the jerk j due to temperature change is mapped to the jerk-related adjustment contribution of the differential coefficient, and the sum of the two constitutes the adaptive adjustment amount of the differential coefficient. .

[0027] It can be understood that the signs and absolute values ​​of the six weighting coefficients directly determine the direction and intensity of the adjustment of the temperature change acceleration and temperature change jerk by the three PID coefficients. When a weighting coefficient is positive, an increase in the corresponding acceleration or jerk component will cause the corresponding PID coefficient to adjust in the increasing direction; when a weighting coefficient is negative, an increase in the corresponding acceleration or jerk component will cause the corresponding PID coefficient to adjust in the decreasing direction. Proportional acceleration weighting coefficients Integral acceleration weighting coefficient and differential acceleration weighting coefficient The larger the absolute value of the accelerometer, the greater the adjustment force of the temperature change acceleration on the corresponding PID coefficient. Proportional accelerometer weighting coefficient. Integral accelerometer weighting coefficient and differential acceleration weighting coefficient The larger the absolute value of the accelerometer, the greater the adjustment force of the corresponding PID coefficient on the temperature change accelerometer.

[0028] The specific values ​​of each weighting coefficient are determined based on the thermal response characteristics of the hot riveting head and the temperature sampling period. The optional determination method is executed during the system initialization phase, and the weighting coefficients remain constant after execution. Within each sampling period, the PID control reads the values ​​of temperature change acceleration and temperature change jerk from memory, performs six multiplications and three additions with the six weighting coefficients respectively, and obtains the adaptive adjustment values ​​of the proportional coefficient, integral coefficient, and derivative coefficient, which are used for subsequent PID coefficient updates.

[0029] See Figure 3 In the figure, the horizontal axis represents the sampling time in seconds, and the vertical axis represents the dimensionless coefficients of the proportional, integral, and derivative coefficients. The red solid line, green dashed line, and blue dotted line represent the changing trends of the proportional, integral, and derivative coefficients, respectively. The data in the figure corresponds to the PID coefficient adaptive adjustment process based on temperature change acceleration and temperature change jerk described in the above embodiment.

[0030] As shown in the graph, all three curves experienced an initial phase of rapid change followed by a relatively stable phase. Specifically, the proportional coefficient (solid red line) rose rapidly from 0 to approximately 3 seconds, reaching a peak of about 1000, then quickly fell back and gradually stabilized, fluctuating between approximately 500 and 700. The integral coefficient (dashed green line) also rose rapidly in the initial stage, reaching a peak of nearly 400, then slowly fell back and stabilized at around 250. The differential coefficient (dotted blue line) dropped sharply in the initial stage, reaching a low of approximately -750, then rebounded and stabilized, with a stable value in the range of approximately -350 to -450.

[0031] In practice, the PID controller maintains three variables in the program data area: the current value of the proportional coefficient, the current value of the integral coefficient, and the current value of the derivative coefficient. Each time the sampling cycle enters the coefficient update phase, the current values ​​of these three variables are read. The read operation is completed by accessing the corresponding memory address, and the value read is the coefficient value retained after the end of the previous control cycle.

[0032] The adaptive adjustment of the proportional coefficient is added to its current value using an addition operation, and the sum is used as the initial update value of the proportional coefficient. Similarly, the adaptive adjustment of the integral coefficient is added to its current value, and the sum is used as the initial update value of the integral coefficient. Finally, the adaptive adjustment of the derivative coefficient is added to its current value, and the sum is used as the initial update value of the derivative coefficient.

[0033] Retrieve the pre-stored preset parameter ranges for the proportional coefficient, integral coefficient, and derivative coefficient. The preset parameter ranges for the proportional coefficient, integral coefficient, and derivative coefficient include a lower limit and an upper limit. The initial updated value of the proportional coefficient is compared with the lower and upper limits. If the initial updated value is greater than the upper limit, the updated proportional coefficient is the upper limit; if it is less than the lower limit, the updated proportional coefficient is the lower limit; and if it lies between the lower and upper limits, the updated proportional coefficient is the initial updated value itself. The initial update value of the integral coefficient is compared with the lower and upper limits of the integral coefficient. If the initial update value is greater than the upper limit, the updated integral coefficient is the upper limit; if it is less than the lower limit, it is the lower limit; and if it lies between the lower and upper limits, it is the initial update value itself. Similarly, the initial update value of the differential coefficient is compared with the lower and upper limits. If the initial update value is greater than the upper limit, the updated differential coefficient is the upper limit; if it is less than the lower limit, it is the lower limit; and if it lies between the lower and upper limits, it is the initial update value itself.

[0034] In some implementations, the process of obtaining the initial update value of the scaling factor, comparing and constraining it can be represented by the scaling factor as an example:

[0035] in, This represents the updated scaling factor. This indicates the current value of the scaling factor. This represents the adaptive adjustment amount of the proportional coefficient. This indicates the lower limit of the proportionality coefficient. This indicates the upper limit of the proportionality coefficient. To perform calculations to obtain the larger value, The operation is performed to select the smaller value. First, take the smaller value between the sum of the current value of the scaling factor, the adaptive adjustment amount of the scaling factor, and the upper limit of the scaling factor. Then, use the outer max operation to combine this smaller value with the lower limit of the scaling factor. Compare and take the larger value, thus constraining the final result to [ , Within the range.

[0036] It is understandable that the integral coefficient and the derivative coefficient adopt the same constraint structure as the proportional coefficient. That is, the updated integral coefficient is obtained by using the sum of the current value of the integral coefficient and the adaptive adjustment amount of the integral coefficient, and then limiting it with the lower limit and the upper limit of the integral coefficient; the updated derivative coefficient is obtained by using the sum of the current value of the derivative coefficient and the adaptive adjustment amount of the derivative coefficient, and then limiting it with the lower limit and the upper limit of the derivative coefficient.

[0037] The lower and upper limits of the proportional gain coefficient are determined as follows: the lower limit is set based on the minimum proportional gain required by the hot riveting head heating system under undisturbed steady-state conditions; the upper limit is set based on the maximum proportional gain allowed by the hot riveting head heating system to avoid overshoot oscillations during the step response from the current temperature to the target temperature. The lower limit of the integral coefficient is set based on the minimum integral action intensity required to eliminate steady-state temperature error; the upper limit is set based on the maximum integral action intensity to prevent the integral controller output from saturating when the temperature deviates significantly from the target. The lower limit of the derivative coefficient is set to zero; the upper limit is set based on the maximum derivative action intensity allowed by the temperature measurement noise amplification not exceeding a set threshold. These six boundary values ​​are obtained through open-loop step response testing and noise measurement of the hot riveting head heating system. They are written into the controller's non-volatile memory during system initialization, and the PID controller directly reads and uses them in each sampling cycle without modification.

[0038] The updated proportional coefficient, integral coefficient, and derivative coefficient are written to the memory locations where the current values ​​of the proportional coefficient, integral coefficient, and derivative coefficient of the PID controller are located, respectively, overwriting the original current values, for use in subsequent PID control quantity calculations.

[0039] In practice, the difference between the set target temperature value and the real-time digital temperature at the current sampling moment is obtained as the temperature deviation value at the current moment. The set target temperature value is the temperature that the riveting head needs to reach and maintain during the hot riveting process. This value is written into the controller memory during the process parameter configuration stage, and the PID controller reads it from the memory in each sampling cycle. The real-time digital temperature at the current sampling moment is the temperature value corresponding to the current sampling moment K in the real-time temperature sequence, which is acquired by the temperature sensor through an analog-to-digital converter.

[0040] The temperature deviation value at the previous sampling time is obtained as the temperature deviation value at the previous time. This value is stored in a memory buffer for the previous cycle's temperature deviation value, which is updated with the current temperature deviation value at the end of each sampling cycle. The temperature deviation values ​​at the two sampling times before the previous time are obtained as the temperature deviation value at the two sampling times before the previous time. These values ​​are stored in a memory buffer for the temperature deviation values ​​at the two sampling times before the previous cycle, which is updated with the previous temperature deviation value at the end of each sampling cycle. Both buffers are initialized to zero during controller power-on initialization.

[0041] When calculating the cumulative integral value of the temperature deviation at the current moment, the cumulative integral value from the previous sampling moment is obtained as the cumulative integral value for that moment. This cumulative integral value is stored in a memory buffer, which contains the sum of all historical temperature deviation values ​​up to the previous sampling moment. The cumulative integral value from the previous moment is added to the current temperature deviation value, and the sum is taken as the cumulative integral value of the current temperature deviation. The update of the cumulative integral value in each sampling period is accomplished by adding the newly generated current temperature deviation value to the cumulative integral value from the previous period, and the result is written back to the cumulative integral value buffer, overwriting the original value.

[0042] When calculating the first-order difference of the temperature deviation at the current moment, the temperature deviation at the current moment is subtracted from the temperature deviation at the previous moment, and the difference is taken as the first-order difference of the temperature deviation at the current moment.

[0043] After obtaining the updated proportional coefficient, updated integral coefficient, and updated derivative coefficient, the PID control components are calculated. The product of the updated proportional coefficient and the current temperature deviation is calculated, and this product is used as the proportional control component. The product of the updated integral coefficient and the cumulative integral value is calculated, and this product is used as the integral control component. The product of the updated derivative coefficient and the first-order difference value is calculated, and this product is used as the derivative control component.

[0044] The proportional control component, integral control component, and derivative control component are added together, and the sum is used as the initial value of the control quantity. The above control quantity calculation process can be expressed by the following formula:

[0045] in, Indicates the initial value of the control variable. This represents the updated scaling factor. This represents the temperature deviation value at the current sampling time k. This represents the updated integral coefficient. This represents the cumulative integral value at the current sampling time K. Indicates the updated differential coefficients. This represents the first-order difference value at the current sampling time k. It is obtained by superimposing the current value of the proportional coefficient with the adaptive adjustment amount of the proportional coefficient and then applying a limiting constraint. It is obtained by superimposing the current value of the integral coefficient with the adaptive adjustment amount of the integral coefficient and then applying a limiting constraint. It is obtained by superimposing the current value of the differential coefficient with the adaptive adjustment amount of the differential coefficient and then subjecting it to amplitude limiting constraints.

[0046] In some implementations, the initial control value is limited before being output to the heating actuator. The rated power of the heating actuator is obtained, and an upper and lower limit value for the limiting process is set based on this power. The upper limit value corresponds to the maximum control value required when the heating actuator is operating at full power, and the lower limit value corresponds to the minimum control value required when the heating actuator stops heating. The initial control value is compared with the upper and lower limits. If the initial control value is greater than the upper limit, the final output control value is the upper limit; if the initial control value is less than the lower limit, the final output control value is the lower limit; and if the initial control value is between the upper and lower limits, the final output control value is the initial value itself. The final output control value is converted into an analog signal or duty cycle signal by a digital-to-analog converter or pulse width modulation module to drive the heating actuator to adjust the heating power of the hot riveting head.

[0047] It is understandable that, during the calculation of the integral cumulative value, optional anti-integral saturation processing can be introduced if the calculated integral cumulative value is too large or too small. The anti-integral saturation processing method is as follows: after adding the current temperature deviation value to the previous integral cumulative value to obtain the sum, it is determined whether the sum exceeds the preset upper limit or is lower than the preset lower limit of the integral cumulative value. If it exceeds, the sum is limited to the upper or lower limit of the integral cumulative value before being written back to the integral cumulative value buffer. The preset upper and lower limits of the integral cumulative value are calculated inversely based on the output limiting range of the heating actuator. That is, by dividing the upper and lower limits of the limiting processing by the updated integral coefficient, the corresponding boundary values ​​are obtained, ensuring that when the integral cumulative value is constrained within this boundary value range, the integral control component will not drive the control quantity beyond the output limiting value of the heating actuator.

[0048] In specific implementation, please refer to Figure 2The thermal response time constant of the hot riveting head and the sampling period of the temperature sensor are obtained. The thermal response time constant of the hot riveting head is obtained as follows: with the hot riveting head at room temperature and in an unloaded state without contacting the workpiece, a step power signal is applied to the heating actuator. Simultaneously, the real-time digital temperature of the hot riveting head is continuously acquired by the temperature sensor and analog-to-digital converter. The time taken for the real-time digital temperature of the hot riveting head to rise from the initial temperature value at the moment the step power signal is applied to 63.2% of the difference between the initial temperature value and the steady-state temperature value is recorded. This time length is the thermal response time constant. The thermal response time constant is measured during the equipment commissioning phase and written to the controller's non-volatile memory after measurement. The sampling period of the temperature sensor is the time interval between two consecutive temperature samples from the analog-to-digital converter. This sampling period is set during the analog-to-digital converter configuration phase. The analog-to-digital converter is triggered by a timer to operate at a fixed frequency. The value of the sampling period is determined by the timer counting frequency and the number of counting cycles, and remains unchanged after being set.

[0049] Based on the thermal response time constant, the upper limit of the temperature change rate of the hot riveting head is determined. The upper limit is determined as follows: the maximum rate of temperature change per unit time during the process of the hot riveting head rising from its initial temperature to its steady-state temperature under the action of a step power signal is taken as the upper limit. This upper limit is obtained by analyzing the step response curve; specifically, a first-order difference operation is performed on the step response curve, and the maximum absolute value of the difference result is taken as the upper limit.

[0050] Based on the sampling period of the temperature sensor and the upper limit of the temperature change rate, the maximum temperature change of the hot riveting head within a single sampling period is calculated. The calculation method involves multiplying the upper limit of the temperature change rate by the sampling period of the temperature sensor; the product is the maximum temperature change of the hot riveting head within a single sampling period. Let the sampling period of the temperature sensor be denoted as... The upper limit of the rate of temperature change is The maximum temperature change of the hot riveting head within a single sampling period. satisfy = × . This indicates the maximum possible change in the temperature of the hot riveting head within a single sampling period, expressed in the same temperature unit as the real-time digital temperature.

[0051] With the maximum temperature change As reference quantities, initial values ​​are set for the proportional acceleration weighting coefficient, proportional jerk weighting coefficient, integral acceleration weighting coefficient, integral jerk weighting coefficient, differential acceleration weighting coefficient, and differential jerk weighting coefficient, respectively. The setting method is as follows: The proportional acceleration weighting coefficient is set to the reciprocal of the baseline quantity multiplied by a first preset coefficient. The first preset coefficient is denoted as... Proportional acceleration weighting coefficient .in, This represents the proportional acceleration weighting coefficient. This indicates the maximum temperature change. This represents the first preset coefficient.

[0052] The proportional accelerometer weighting coefficient is set to the reciprocal of the square of the reference value multiplied by a second preset coefficient. The second preset coefficient is denoted as... proportional accelerometer weighting coefficient in, This represents the proportional accelerometer weighting coefficient. This indicates the maximum temperature change. This represents the second preset coefficient.

[0053] The integral acceleration weighting coefficient is set to the reciprocal of the baseline quantity multiplied by a third preset coefficient. The third preset coefficient is denoted as... Integral acceleration weighting coefficient in, This represents the integral acceleration weighting coefficient. This indicates the maximum temperature change. This represents the second preset coefficient.

[0054] The integral accelerometer weighting coefficient is set to the reciprocal of the square of the reference quantity multiplied by the fourth preset coefficient. The fourth preset coefficient is denoted as... Integral accelerometer weighting coefficient in, This represents the integral accelerometer weighting coefficient. This indicates the maximum temperature change. This represents the fourth preset coefficient.

[0055] The differential acceleration weighting coefficient is set to the reciprocal of the baseline quantity multiplied by the fifth preset coefficient. The fifth preset coefficient is denoted as... Differential acceleration weighting coefficient in, This represents the differential acceleration weighting coefficient. This indicates the maximum temperature change. This represents the fifth preset coefficient.

[0056] The differential acceleration weighting coefficient is set to the reciprocal of the square of the reference quantity multiplied by the sixth preset coefficient. The sixth preset coefficient is denoted as... Differential accelerometer weighting coefficient in, This represents the differential accelerometer weighting coefficient. This indicates the maximum temperature change. This represents the sixth preset coefficient.

[0057] First preset coefficient Second preset coefficient Third preset coefficient Fourth preset coefficient Fifth preset coefficient and the sixth preset coefficient All are dimensionless, purely numerical coefficients. First preset coefficients. The value range is a real number between -1.0 and -0.1, and the second preset coefficient is... The value range is a real number between -0.5 and -0.05, and the third preset coefficient is... The value range is a real number between -0.5 and -0.05, the fourth preset coefficient. The value range is a real number between -0.2 and -0.02, the fifth preset coefficient. The value range is a real number between 0.05 and 0.5, the sixth preset coefficient. The values ​​range from 0.02 to 0.2. The specific values ​​of the six preset coefficients are determined by adjusting the PID controller's temperature control performance during the hot riveting head step response test. The adjustment process is as follows: Initial values ​​are set for each of the six preset coefficients within their respective ranges. The PID temperature control program is run, and the overshoot, settling time, and steady-state temperature fluctuation amplitude during the rise of the hot riveting head temperature from the initial temperature value to the target temperature value are observed. The first preset coefficient is then fine-tuned sequentially. Second preset coefficient Reduce the overshoot and then fine-tune the third preset coefficient sequentially. and the fourth preset coefficient To reduce the amplitude of steady-state temperature fluctuations, the fifth preset coefficient is adjusted sequentially. and the sixth preset coefficient To shorten the adjustment time, repeat the above adjustments until the temperature control performance meets the process requirements. Record the values ​​of the six preset coefficients at this point and write them into the controller's non-volatile memory.

[0058] It is understandable that using the reciprocal of the maximum temperature change as the benchmark factor for the weighting coefficient involving temperature change acceleration makes the adjustment of the PID coefficient by the temperature change acceleration proportional to the reciprocal of the temperature change. This suppresses the adjustment amplitude when the temperature change is large and enhances the adjustment amplitude when the temperature change is small. Using the reciprocal of the square of the maximum temperature change as the benchmark factor for the weighting coefficient involving temperature change acceleration makes the adjustment of the PID coefficient by the temperature change acceleration proportional to the reciprocal of the square of the temperature change, resulting in a stronger nonlinear compensation characteristic in response to the drasticness of temperature changes.

[0059] See Figure 4 In the graph, the horizontal axis represents time in seconds; the left vertical axis represents temperature in degrees Celsius; and the right vertical axis represents the rate of temperature change in degrees Celsius per second. The solid curve in the graph represents the real-time temperature, the dashed curve represents the rate of temperature change, and the blue dashed line represents the upper limit of the rate of temperature change.

[0060] The real-time temperature digital curve starts from an initial temperature of approximately 25°C and shows a trend of rapid increase followed by stabilization over time. The temperature rises rapidly from 0 to approximately 15 seconds, then gradually approaches the target temperature of approximately 250°C, and stabilizes at around 250°C after 20 seconds, reflecting the process of the hot riveting head temperature rising from room temperature to the steady-state target temperature.

[0061] The temperature change rate curve exhibits a high fluctuation peak during the initial heating phase (0 to approximately 3 seconds), reaching a maximum of nearly 220℃ / s, followed by a rapid decline. After about 10 seconds, the temperature change rate decreases significantly, approaching zero with minimal fluctuation, indicating that the temperature change rate slows down considerably as the temperature approaches steady state. The temperature change rate remains consistently below the preset upper limit, with the blue dashed line value being approximately 55℃ / s, ensuring that the temperature change rate during the control process does not exceed the system's set safety range.

[0062] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A temperature self-tuning method for hot riveting process based on data acquisition and PID control, characterized in that, The method includes: The real-time digital temperature of the hot riveting head at multiple consecutive sampling moments is obtained from the temperature sensor via an analog-to-digital converter, forming a real-time temperature sequence; Based on the real-time temperature sequence, calculate the temperature change acceleration and temperature change jerk at the current sampling moment; Based on the temperature change acceleration and the temperature change jerk, determine the adaptive adjustment amount of the proportional coefficient, the adaptive adjustment amount of the integral coefficient, and the adaptive adjustment amount of the derivative coefficient of the PID controller. The adaptive adjustment amounts of the proportional coefficient, integral coefficient, and derivative coefficient are superimposed on their respective current coefficient values ​​to obtain the updated proportional coefficient, integral coefficient, and derivative coefficient. Using updated proportional, integral, and derivative coefficients, the PID controller calculates and outputs a control quantity to drive the heating actuator to adjust the temperature of the hot riveting head.

2. The temperature self-tuning method in the hot riveting process based on data acquisition and PID control according to claim 1, characterized in that, The step of calculating the temperature change acceleration and temperature change jerk at the current sampling moment based on the real-time temperature sequence includes: Obtain the temperature value at the current sampling time, the temperature value at the previous sampling time, and the temperature value at the second-to-last sampling time in the real-time temperature sequence; Calculate the difference between the temperature value at the current sampling time and the temperature value at the previous sampling time to obtain the first-order difference of the temperature at the current time; The difference between the temperature value at the previous sampling time and the temperature value at the two previous sampling times is calculated to obtain the first-order difference of the temperature at the previous time. The difference between the first-order temperature difference at the current moment and the first-order temperature difference at the previous moment is calculated to obtain the second-order temperature difference at the current moment, which is used as the acceleration of temperature change. The difference between the second-order temperature difference at the current moment and the second-order temperature difference at the previous sampling moment is calculated to obtain the third-order temperature difference at the current moment, which is used as the accelerometer for the temperature change.

3. The temperature self-tuning method in the hot riveting process based on data acquisition and PID control according to claim 1, characterized in that, The step of determining the adaptive adjustment amounts of the proportional coefficient, integral coefficient, and derivative coefficient of the PID controller based on the temperature change acceleration and the temperature change jerk includes: Obtain the preset proportional acceleration weight coefficient, proportional jerk weight coefficient, integral acceleration weight coefficient, integral jerk weight coefficient, differential acceleration weight coefficient, and differential jerk weight coefficient; The proportional acceleration adjustment component is obtained by multiplying the temperature change acceleration by the proportional acceleration weighting coefficient. The proportional jerk adjustment component is obtained by multiplying the temperature change jerk by the proportional jerk weighting coefficient. The sum of the proportional acceleration adjustment component and the proportional jerk adjustment component is used as the adaptive adjustment amount of the proportional coefficient. The integral acceleration adjustment component is obtained by multiplying the temperature change acceleration by the integral acceleration weighting coefficient. The integral jerk adjustment component is obtained by multiplying the temperature change jerk by the integral jerk weighting coefficient. The sum of the integral acceleration adjustment component and the integral jerk adjustment component is used as the adaptive adjustment amount of the integral coefficient. The differential acceleration adjustment component is obtained by multiplying the temperature change acceleration by the differential acceleration weighting coefficient. The differential accelerometer adjustment component is obtained by multiplying the temperature change accelerometer with the differential accelerometer weighting coefficient. The sum of the differential acceleration adjustment component and the differential jerk adjustment component is used as the adaptive adjustment amount of the differential coefficient.

4. The temperature self-tuning method in the hot riveting process based on data acquisition and PID control according to claim 1, characterized in that, The step of superimposing the adaptive adjustment amounts of the proportional coefficient, integral coefficient, and derivative coefficient with their respective current coefficient values ​​to obtain updated proportional coefficients, integral coefficients, and derivative coefficients includes: Obtain the current values ​​of the proportional coefficient, integral coefficient, and derivative coefficient of the PID controller at the current operating moment; The adaptive adjustment amount of the proportional coefficient is summed with the current value of the proportional coefficient to obtain the updated proportional coefficient; The updated integral coefficients are obtained by summing the adaptive adjustment amount of the integral coefficients with the current value of the integral coefficients. The adaptive adjustment amount of the differential coefficient is summed with the current value of the differential coefficient to obtain the updated differential coefficient.

5. The temperature self-tuning method in the hot riveting process based on data acquisition and PID control according to claim 4, characterized in that, The step of using updated proportional, integral, and derivative coefficients, and having the PID controller calculate and output a control quantity to drive the heating actuator to adjust the temperature of the hot riveting head, includes: The difference between the set target temperature value and the temperature value at the current sampling time is obtained as the temperature deviation value at the current time; The temperature deviation value at the previous sampling time is obtained as the temperature deviation value at the previous time, and the temperature deviation value at the two previous sampling times is obtained as the temperature deviation value at the two previous times. Based on the current temperature deviation value, the previous temperature deviation value, and the temperature deviation values ​​of the two previous times, calculate the integral cumulative value of the current temperature deviation and the first difference value of the current temperature deviation. The proportional control component is obtained by multiplying the updated proportional coefficient by the temperature deviation value at the current moment. The product of the updated integral coefficient and the cumulative integral value is calculated to obtain the integral control component; Calculate the product of the updated differential coefficients and the first-order difference value to obtain the differential control component; The sum of the proportional control component, the integral control component, and the derivative control component is taken as the control quantity.

6. The temperature self-tuning method in the hot riveting process based on data acquisition and PID control according to claim 5, characterized in that, The step of calculating the integral cumulative value of the temperature deviation at the current moment and the first-order difference value of the temperature deviation at the current moment based on the current temperature deviation value, the previous temperature deviation value, and the temperature deviation values ​​of the two previous moments includes: Obtain the cumulative integral value at the previous sampling time as the cumulative integral value at the previous time. The integral cumulative value of the previous moment is added to the temperature deviation value of the current moment to obtain the integral cumulative value of the temperature deviation at the current moment; The difference between the current temperature deviation value and the previous temperature deviation value is calculated to obtain the first-order difference value of the current temperature deviation.

7. The temperature self-tuning method in the hot riveting process based on data acquisition and PID control according to claim 3, characterized in that, Obtaining the proportional acceleration weight coefficient, the proportional jerk weight coefficient, the integral acceleration weight coefficient, the integral jerk weight coefficient, the differential acceleration weight coefficient, and the differential jerk weight coefficient includes: Obtain the thermal response time constant of the hot riveting head and the sampling period of the temperature sensor; Based on the thermal response time constant, determine the upper limit of the temperature change rate of the hot riveting head; Based on the sampling period and the upper limit of the temperature change rate, calculate the maximum temperature change of the hot riveting head within a single sampling period; Based on the maximum temperature change, the initial values ​​of the proportional acceleration weight coefficient, the proportional jerk weight coefficient, the integral acceleration weight coefficient, the integral jerk weight coefficient, the differential acceleration weight coefficient, and the differential jerk weight coefficient are respectively set.

8. The temperature self-tuning method in the hot riveting process based on data acquisition and PID control according to claim 7, characterized in that, The step of setting initial values ​​for the proportional acceleration weighting coefficient, the proportional accelerometer weighting coefficient, the integral acceleration weighting coefficient, the integral accelerometer weighting coefficient, the differential acceleration weighting coefficient, and the differential acceleration weighting coefficient based on the maximum temperature change includes: The maximum temperature change is used as the baseline. The proportional acceleration weighting coefficient is set to the reciprocal of the reference quantity multiplied by a first preset coefficient; The proportional acceleration weighting coefficient is set to the reciprocal of the square of the reference quantity multiplied by a second preset coefficient; The integral acceleration weighting coefficient is set to the reciprocal of the reference quantity multiplied by a third preset coefficient; The integral acceleration weighting coefficient is set to the reciprocal of the square of the reference quantity multiplied by a fourth preset coefficient; The differential acceleration weighting coefficient is set to the reciprocal of the reference quantity multiplied by a fifth preset coefficient; The differential acceleration weighting coefficient is set to the reciprocal of the square of the reference quantity multiplied by a sixth preset coefficient.

9. The temperature self-tuning method in the hot riveting process based on data acquisition and PID control according to claim 4, characterized in that, The updated proportional coefficient, integral coefficient, and derivative coefficient are each constrained within their respective preset parameter ranges.

10. The temperature self-tuning method in the hot riveting process based on data acquisition and PID control according to claim 5, characterized in that, The control quantity is limited before being output to the heating actuator, and the upper and lower limits of the limiting process are set according to the rated power of the heating actuator.

Images