A temperature change rate-based hot riveting horn synchronous temperature control method and system

Abstract

The present application relates to the technical field of welding, in particular to a hot riveting welding head synchronous temperature control method and system based on temperature change rate, comprising: triggering operation of full-power heating pulse on the welding head, calculating pulse duration according to start time stamp and end time stamp, calculating temperature rise, and removing spike noise through digital filtering algorithm; collecting environmental temperature and relative humidity in real time, calculating equivalent heat capacity; calculating initial temperature difference between any welding heads, constructing dynamic target temperature curve; calculating actual temperature change rate, synchronously extracting target change rate of dynamic target curve at corresponding moment, calculating change rate deviation of both, judging whether intervention adjustment is needed, if intervention adjustment is needed, calculating output signal by using incremental PID algorithm, and adjusting heating power according to the output signal. The present application effectively improves the quality stability of hot riveting welding by accurately quantifying the thermal characteristics of heterogeneous welding heads, dynamically planning individualized temperature rise trajectory, and high-frequency closed-loop regulation.

Description

Technical Field

[0001] This invention relates to the field of welding technology, and in particular to a method and system for synchronous temperature control of hot riveting welding heads based on the rate of temperature change. Background Technology

[0002] In hot riveting welding production scenarios, electric vehicle welding production lines often need to use hot riveting welding heads of different brands and models to meet diverse process requirements. However, these welding heads exhibit significant heterogeneity, mainly manifested in equivalent heat capacity differences exceeding 300%. Since heat capacity directly determines the heat storage capacity and heating characteristics of the welding head, heterogeneous welding heads show extremely large differences in heating rate and thermal response time under the same heating conditions, making it extremely difficult to achieve synchronized temperature reaching.

[0003] Traditional temperature control methods often use a uniform heating curve, failing to consider the individual characteristics of the welding head. This can easily lead to problems such as some welding heads heating up too quickly, causing overshoot, while others heat up too slowly and fail to reach the target temperature simultaneously. Furthermore, the complex working environment of the welding head, including fluctuations in ambient temperature and humidity, differences in heat dissipation conditions at different locations (such as the difference in heat dissipation efficiency between areas near fans and enclosed areas), electromagnetic interference, and equipment vibration, introduces temperature measurement noise and temperature rise calculation errors, further reducing temperature control accuracy.

[0004] In addition, the existing system lacks the ability to quantitatively model the thermal characteristics of the welding head in real time, and cannot dynamically adapt to the thermal capacity drift phenomenon after long-term use of the welding head. Moreover, the temperature control adjustment relies on fixed PID parameters, and the response to temperature difference deviation is lagging. It is difficult to achieve the high-precision synchronization requirement of ±2℃ temperature difference of multiple welding heads within a short time window of 0.5-3 seconds, which seriously affects the stability of product welding quality and production line efficiency. Summary of the Invention

[0005] This invention effectively improves the stability of hot riveting welding quality by precisely quantifying the thermal characteristics of heterogeneous welding heads, dynamically planning personalized heating trajectories, and using high-frequency closed-loop control.

[0006] The technical solution proposed in this invention is: a method for synchronous temperature control of a hot riveting head based on the rate of temperature change, the method comprising: Each welding head to be synchronized is independently triggered by a full-power heating pulse. The pulse start time and end time are recorded. The pulse duration is calculated based on the start and end times. The temperature rise is calculated based on the initial temperature at the start of the pulse and the real-time temperature at the end of the pulse. The actual temperature rise is corrected by combining the position coefficient and the position of each welding head. Peak noise is removed by digital filtering algorithm. The ambient temperature and relative humidity are collected in real time, and the equivalent heat capacity of each welding head is calculated by combining the actual temperature rise and pulse duration. The initial temperature difference between any welding heads is calculated using the initial temperature of the welding head. The welding heads are then classified according to the initial temperature difference. The welding head with the largest equivalent heat capacity is selected as the reference channel, and the global reference rate is calculated by combining the thermal coupling coefficient matrix. A thermodynamic model of the welding head is established using COMSOL, and a dynamic target temperature curve is constructed. The real-time temperature of each welding head is collected to calculate the actual temperature change rate. The target change rate of the dynamic target curve at the corresponding time is extracted simultaneously to calculate the deviation between the two change rates. Based on the change rate deviation, it is determined whether intervention and adjustment are required. If intervention and adjustment are required, the PID controller coefficient is dynamically adjusted according to the absolute value of the deviation. The output signal is calculated using an incremental PID algorithm, and the heating power is adjusted according to the output signal.

[0007] Preferably, the specific process of correcting the actual temperature rise is as follows: The average temperature within a sampling period before the pulse starts is used as the initial temperature, and the instantaneous temperature within a sampling period after the pulse ends is used as the final temperature. The difference between the two is calculated to obtain the original temperature rise. For marked suspected noise points, a digital filtering algorithm is used to filter them to obtain the filtered original temperature rise. The heat dissipation level is divided according to the location of the welding head, and a corresponding position coefficient is set for different heat dissipation level areas. The filtered original temperature rise is multiplied by the corresponding position coefficient to obtain the final actual temperature rise.

[0008] Preferably, the specific content of the digital filtering algorithm is as follows: Traverse the temperature sampling sequence and calculate the temperature difference between adjacent sampling points. When the temperature difference between adjacent sampling points exceeds the peak determination threshold, the sampling point is marked as a suspected peak noise point. For the marked noise point, extract several normal sampling points before and after it to form a dataset. After sorting the window data, take the median value to replace the noise point. Normal points not marked as noise are directly retained in their original values. For edge points at the beginning and end of the sequence, if they are determined to be noise, the mean of the effective data on one side is used for correction. After correction, calculate the temperature difference stability of the filtered sequence. If all adjacent temperature differences are ≤ the peak determination threshold, the filtering is effective. Otherwise, expand the window and repeat the correction until the condition is met.

[0009] Preferably, the specific calculation process for the equivalent heat capacity is as follows: The equivalent heat capacity of the welding head is calculated by using the heat conversion efficiency constant, temperature rise, power, and pulse duration. The heat conversion efficiency constant is corrected based on the ambient temperature and relative humidity, and the equivalent heat capacity of the welding head is corrected based on the corrected heat conversion efficiency constant. The system automatically performs a rapid calibration when it is first started each day, performs a deep calibration according to the preset welding number interval, and updates the equivalent heat capacity parameter library.

[0010] Preferably, the specific process for obtaining the dynamic target temperature curve is as follows: After heat capacity identification, several temperature data points are continuously collected at a fixed frequency, and the average value is taken as the initial temperature of each welding head. The initial temperature difference between any two welding heads is calculated. After determining the maximum initial temperature difference, the welding heads are divided into near-reference group and far-reference group. The welding head with the largest equivalent heat capacity is selected as the reference channel. A thermal coupling coefficient matrix is ​​introduced, and the global reference heating rate is calculated in combination with the full power parameters of the reference channel. The dynamic target temperature curve includes a main heating segment and a buffer segment. The main heating segment generates the target trajectory with the slope of the reference heating rate. The main heating time is extended for welding heads in the far-reference group with extreme temperature differences. When the welding head temperature approaches the target temperature, it automatically switches to the buffer segment. A thermodynamic model is established using COMSOL to predict the thermal coupling effect and optimize the parameters of the buffer segment.

[0011] Preferably, the intervention and adjustment judgment process is as follows: After the system enters the real-time control phase, it collects the real-time temperature of each welding head at a high-frequency sampling frequency. After smoothing the noise using a moving average filter, it calculates the actual temperature change rate based on the temperature difference between two consecutive sampling times to ensure that the accuracy of the actual temperature change rate calculation meets the requirements. Simultaneously, it extracts the target change rate of the dynamic target temperature curve at the corresponding time and obtains the target value at any time through an interpolation algorithm. It calculates the absolute value of the deviation between the actual change rate and the target change rate in real time. When the absolute value of the deviation is less than or equal to the first deviation threshold, it is determined to be in a normal state and no intervention is required. When the absolute value of the deviation is greater than the first deviation threshold, it is marked as an abnormal deviation, triggering the real-time tuning mechanism of the PID parameters and initiating intervention. When the absolute value of the deviation is greater than the second deviation threshold, the emergency adjustment mode is activated.

[0012] Preferably, the specific calculation process of the output signal is as follows: When the absolute value of the deviation is greater than the third deviation threshold, the proportional coefficient is increased and the integral coefficient is decreased to speed up the response. When the first deviation threshold is less than the absolute value of the deviation and less than or equal to the third deviation threshold, the initial parameters are maintained. When the absolute value of the deviation is less than or equal to the first deviation threshold, the proportional coefficient is decreased and the integral coefficient is increased to eliminate steady-state error. When the absolute value of the deviation is greater than the integral pause threshold, the integral term is paused, and only proportional and derivative adjustments are used to quickly suppress the deviation. The incremental power adjustment signal is calculated using an incremental PID algorithm and converted into a heating power command. The power supply duration of the welding head heating element is controlled by PWM pulse width modulation technology. The cumulative error is calculated by summing the instantaneous deviations over several consecutive cycles. When the absolute value of the cumulative error is greater than the third deviation threshold, the power compensation amount is calculated using a compensation coefficient and superimposed on the PID output signal. The single compensation amount is ≤2% of the rated power. The final output signal must ensure that the power output is less than or equal to the upper limit of the rated power. If the limit is exceeded, the upper limit is output and the deviation is recorded.

[0013] The present invention also provides a method for synchronous temperature control of a hot riveting head based on the rate of temperature change, wherein the system is used to execute the method for synchronous temperature control of a hot riveting head based on the rate of temperature change.

[0014] The present invention also provides a computer-readable storage medium storing a computer program that is executed by a processor to implement the aforementioned method for synchronous temperature control of a hot riveting head based on the rate of temperature change.

[0015] The beneficial effects of this invention are: 1. By independently executing standardized full-power heating pulses on each welding head to be synchronized, combined with digital filtering to eliminate peak noise, position coefficient correction for heat dissipation differences, and environmental parameter optimization calibration constants, accurate calculation of equivalent heat capacity is achieved. This process does not rely on preset power parameters for the welding heads and can complete the quantitative modeling of thermal characteristics for all heterogeneous welding heads (heat capacity difference > 300%) within 3 seconds. The automatic calibration mechanism after daily and every 1000 welding cycles further ensures parameter stability, providing reliable basic data support for subsequent synchronous temperature control and improving the accuracy of thermal characteristic modeling.

[0016] 2. Based on the initial temperature difference grading results and the equivalent heat capacity selection benchmark channel, a personalized target curve including a main heating segment and a buffer segment is generated. The main heating segment eliminates the initial temperature difference using a global benchmark rate, and extends the welding time for extreme temperature differences to ensure catch-up performance. The buffer segment effectively suppresses temperature overshoot through slope design of 0.3-0.4 times the benchmark rate and COMSOL simulation optimization. This segmented dynamic adaptation strategy solves the problem that fixed curves cannot cope with the discreteness of the welding head state, achieving the synchronous temperature target of ±2℃ temperature difference within a 0.5-3 second time window, balancing the synchronization rate and the risk of overshoot.

[0017] 3. Real-time monitoring of temperature change rate using 1kHz high-frequency sampling, dynamic tuning of PID parameters triggered by deviation thresholds, and precise adjustment of heating power combined with a cumulative error compensation algorithm ensure that the actual heating rate matches the target curve. Simultaneously, power limits (110% of rated value), temperature limits (target + 5℃), and an abnormal shutdown mechanism are set. Root cause analysis is performed on substandard welded heads, and maintenance suggestions are pushed. This closed-loop control mechanism not only reduces temperature deviation from ±1.5℃ to ±0.8℃, but also significantly reduces equipment failure rate through multiple safety constraints, increasing the product qualification rate of the production line to over 99%, ensuring temperature control accuracy, and enhancing system safety and reliability. Attached Figure Description

[0018] Figure 1 This is a flowchart of a synchronous temperature control method for a hot riveting head based on the rate of temperature change, according to the present invention. Figure 2 This is a flowchart illustrating the temperature control process of a synchronous temperature control method for hot riveting heads based on the rate of temperature change, according to the present invention. Detailed Implementation

[0019] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0020] It is understood that the term "a" should be understood as "at least one" or "one or more," that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0021] like Figure 1 and Figure 2 As shown, for the heterogeneity issue of different brands and unknown power hot riveting welding heads used in the production line (with heat capacity differences >300%), the system independently executes a full-power heating pulse trigger operation for each welding head to be synchronized. Heating pulse duration. Strictly control the value within the range of 150-250ms, preferably 200ms ± 10% (i.e., 180ms ≤ ≤220ms). This parameter setting is based on two considerations: first, to ensure that the pulse duration is sufficient to generate a measurable temperature change (avoiding the temperature rise signal being overwhelmed by noise due to excessively short pulses); and second, to avoid prolonged heating affecting the initial temperature stability of the welding head (preventing overheating from altering the subsequent temperature control reference state). Pulse execution is achieved through a relay module, outputting a full-power heating signal (voltage 220V±5%, current dynamically adjusted according to the welding head power), and simultaneously recording the pulse start timestamp. and end timestamp Ensured by using a high-precision clock chip (accuracy ≤1μs) The timing accuracy is ≤1ms. A temperature and humidity sensor (sampling frequency 1Hz, accuracy ±1℃ / ±5%RH) is deployed synchronously to collect ambient temperature data in real time. And relative humidity (RH). The system triggers heating pulses sequentially according to the welding head device ID, with an interval of ≥5 seconds between adjacent pulses to avoid heat accumulation affecting measurement accuracy. The pulse execution status is fed back in real time through indicator lights and software logs.

[0022] Throughout the entire duration of the heating pulse application, data acquisition is completed using the NTC temperature sensor built into the welding head (temperature range -50℃ to 300℃, accuracy ±0.5℃, sampling frequency ≥1kHz, data transmission via shielded cable to avoid electromagnetic interference), including the initial temperature stabilization of the welding head before the pulse begins. (Take the average temperature within 100ms before the pulse starts (preset initial temperature sampling time), calculated using the following formula:) ,in, For the pulse number The second sampling temperature (to ensure initial temperature stability) and the real-time temperature at the end of the pulse. (The instantaneous temperature value within 5ms after the pulse ends (preset end temperature sampling time) is captured, and the peak temperature is captured by hardware triggering to avoid the influence of thermal inertia.) Calculate the temperature rise. (Accurate to 0.1℃), spike noise is removed using a digital filtering algorithm.

[0023] During the application of the heating pulse, the data collected by the temperature sensor may be affected by electromagnetic interference, equipment vibration, etc., resulting in spike noise (manifested as instantaneous temperature jumps >5℃ and duration <10ms). This needs to be processed by a digital filtering algorithm to ensure the accuracy of the temperature rise calculation. An adaptive median filtering algorithm is adopted, which can both remove isolated spikes and retain the original temperature trend. The specific process is as follows: Traverse the temperature sampling sequence and calculate the temperature difference between adjacent sampling points: ; in: For the first One original temperature sample value (sampling frequency ≥ 1 kHz, i.e., sampling interval) ), For sampling point number ( This represents the total number of samples within the pulse period, with a 200ms pulse corresponding to... ).when ,mark The point shown is a suspected spike noise point; the rest are normal points. The threshold for determining the peak value. If the temperature difference between adjacent samples exceeds this value, it is judged as suspected spike noise.

[0024] For the marked suspected spike noise points, median filtering is used for correction. Data from one normal sampling point before and after the noise point are extracted to form a windowed dataset. Sort the window data to get Take the median. As the filtered value .

[0025] For normal points not marked as noise, retain the original values ​​directly. For the beginning and end edge points of the sequence ( or If it is determined to be noise, only one-sided valid data (e.g.) is used. Or take or (mean correction).

[0026] Calculate the temperature stability of the filtered sequence If all If the condition is met, the filtering is effective; otherwise, the median filtering correction is repeated (the window size can be expanded to 5 points) until the condition is met.

[0027] The filtered temperature sequence is directly used to calculate the initial temperature. ,make sure Unaffected by spike noise, accuracy is controlled within ±0.3℃. Real-time temperature at the end of the pulse. Temperature value taken from the end of the filtered pulse To ensure temperature rise Calculation error ±0.1℃. The algorithm execution time is ≤1ms, matching the sensor sampling frequency (1kHz), and does not affect the real-time performance of rapid heat capacity identification (modeling is completed within 3 seconds). This algorithm can eliminate more than 98% of spike noise, reducing the temperature data fluctuation range from over ±2℃ before filtering to within ±0.5℃, providing reliable raw data support for equivalent heat capacity calculation and subsequent dynamic target curve generation.

[0028] To address the differences in heat dissipation conditions caused by the dispersed positions of the welding heads, a position coefficient is added. Calibration, and pre-classification of heat dissipation levels for each welding head location on the production line (e.g., areas near fans). enclosed area Other areas The actual temperature rise calculation was corrected to .Will The corresponding welding head's device ID and pulse duration The data is bound to storage to form a related dataset including welding head, pulse, and temperature rise, providing raw input for subsequent calculations and supporting subsequent traceability and analysis.

[0029] Based on the collected pulse parameters and temperature rise data, the equivalent heat capacity of each welding head is calculated using formulas, thereby quantifying the characteristics of the welding head with unknown parameters and providing basic parameters for subsequent temperature control. According to thermal principles, the heat absorbed by the welding head during the heating process... With temperature rise Equivalent heat capacity satisfy Due to the application of a fixed power and pulse duration Calories ( (where is the heat conversion efficiency constant), the simplified formula for calculating the equivalent heat capacity is: ; in, Equivalent heat capacity (unit: J / ℃, reflecting the welding head's ability to store heat; the larger the value, the slower the welding head heats up). The calibration constant (unit: J) is obtained through factory calibration of the same model welding head and is pre-stored in the system database. Different brands of welding heads correspond to different constants. Value, for example, brand A's Brand B By introducing an environmental parameter correction term, the optimization is as follows: ; in, , The reference temperature is 25℃. The calibration constant at that time.

[0030] After the calculation is completed, the system generates a unique identifier for each welding head. The parameters are associated with the device ID and stored in the parameter library, completing the heat capacity modeling of all welding heads within 3 seconds. The system automatically performs a "quick calibration" upon its first startup each day, and a "deep calibration" after every 1000 welding operations, updating the parameters accordingly. Parameter library. When If an anomaly is detected, pulse heating and data acquisition are automatically re-executed to ensure modeling accuracy. This parameter will directly serve as the core input for subsequent dynamic target curve generation steps, used to match the heating rate requirements of different welding heads and solve the problem of synchronous temperature control caused by differences in heat capacity.

[0031] This process achieves three core values ​​through a closed-loop workflow of pulse application, temperature rise measurement, and heat capacity calculation: First, it eliminates the need to obtain welding head power parameters in advance, solving the problem of adapting to heterogeneous equipment; second, it completes modeling within 3 seconds, meeting the needs of rapid production line changeovers; and third... The parameters are directly linked to the generation of the subsequent target curve, providing a quantitative basis for synchronous temperature control and realizing the logical connection between characteristic perception and parameter transmission.

[0032] After rapid heat capacity identification, the system first collects the initial temperature of all hot riveting welding heads involved in synchronous temperature control to clarify the differences in the initial state of each welding head, providing a basis for subsequent curve adaptation. Using the temperature sensor built into the welding head (accuracy ±0.5℃), 10 temperature data points are continuously collected at a frequency of 100Hz during the stabilization phase 3 seconds after heat capacity identification (a preset waiting time to ensure temperature stability) before formal heating. The average value is taken as the initial temperature of each welding head. ( Representing the (One welding head), the data acquisition process avoids periods of drastic ambient temperature fluctuations. Calculate the initial temperature difference between any two welding heads. Statistically plot the temperature difference distribution histogram and determine the maximum initial temperature difference. This is used to determine the dispersion of the initial state of the welding head, providing a reference for subsequent differentiated temperature control. The typical value should be ≤5℃; exceeding this triggers preheating leveling. Welding heads are divided into two different temperature difference levels: "near-reference group" (difference from the lowest initial temperature ≤3℃) and "far-reference group" (difference >3℃), providing a classification basis for subsequent curve adaptation. The data includes the welding head ID, A structured data table of temperature difference levels is used to develop a health status dashboard for heat capacity parameters. Abnormal welding heads are identified by color coding (e.g., red when the k value deviates from the threshold by ±15%). The initial temperature distribution is displayed through a visual interface, allowing operators to quickly determine the equipment status.

[0033] To ensure that multiple welding heads can reach the target temperature simultaneously, the welding head with the weakest heating capacity should be selected as the reference channel, and its heating rate should be used as the reference benchmark to avoid some welding heads being unable to keep up due to the reference being too high.

[0034] The heating capacity of a welding head is directly related to its equivalent heat capacity (inversely proportional; the larger the equivalent heat capacity, the slower the heating). Therefore, based on the equivalent heat capacity... Screening for heat capacity The largest welding head is used as the reference channel (i.e., the welding head with the slowest heating, numbered). ),Right now The heating rate of the reference channel is used as the global reference rate. This rate is determined by the heat capacity characteristics of the reference channel, reflecting its natural temperature rise capability under full-power heating. The target curves for all subsequent welds are designed based on this rate. The calculation formula is as follows: For an array of n welding heads Introducing the thermal coupling coefficient matrix, it is optimized to ,in, The full-power heating power of the welding head (an inherent parameter of the equipment, obtained through factory calibration). For welding head Butt welding head The heat-affected ratio. As a global reference rate, it is stored in the control parameter library to provide core parameters for the subsequent generation of target curves, ensuring that all welding heads are designed with the reference rate as the basis for heating trajectory.

[0035] Based on the heating rate of the reference channel and the initial temperature difference of each welding head, a dynamic target temperature curve is generated for each welding head, including a "main heating section" and a "slope decreasing buffer section", to ensure synchronous temperature reaching and avoid overshoot.

[0036] This stage uses the heating rate of the reference channel. To ensure a uniform slope, all welding heads heat up at the same rate. For welding heads with lower initial temperatures, this stage reduces the temperature difference with other welding heads through continuous heating; for welding heads with higher initial temperatures, this stage provides adjustment space for subsequent buffer sections. Its core function is to eliminate synchronization obstacles caused by initial temperature differences by unifying the heating rate.

[0037] When the welding head temperature approaches the target temperature (typically within 10°C), it automatically switches to the buffer zone. The heating rate in the buffer zone is adjusted to the reference rate. The temperature is set at 0.3-0.4 times the target temperature to reduce the heating inertia as the temperature approaches the target temperature, preventing the temperature from exceeding the target value (i.e., overshoot) due to continuous high-speed heating, and ensuring that the temperature converges smoothly to the target value. A thermodynamic model of the welding head is established using COMSOL to predict the thermal coupling effect and optimize the buffer section parameters, thereby constructing a dynamic target temperature curve.

[0038] Based on the initial temperature differences of each welding head, the switching timing between the main heating section and the buffer section is dynamically adjusted. The welding head with a lower initial temperature extends the duration of the main heating section to catch up fully, while the welding head with a higher initial temperature enters the buffer section earlier to prevent overshoot. Ultimately, all welding heads reach the target temperature synchronously (temperature difference ±2℃) within a time window of 0.5-3 seconds.

[0039] Each dynamic target temperature curve consists of two segments, achieving precise temperature control through dynamic switching. In the main heating segment, all welding heads are driven to heat up synchronously at a reference rate, reducing the initial temperature difference and ensuring that those heating slower do not fall behind those heating faster. The formula for the main heating segment trajectory is as follows: ,in, Heating time (unit: seconds) For the first A welding head at The target temperature at any given time. The formula for the main heating phase duration of the "far reference group" welding head is optimized as follows: When the welding head temperature is real-time It automatically switches to the buffer segment, where, Preset target temperature (adjustable from 20℃ to 280℃). ℃ (the maximum value is taken for the high-temperature range, such as a target temperature of 280℃) ℃). The formula for the trajectory of the buffer segment is: ,in, Temperature at trigger time For trigger time, (Buffer slope). Reduce the heating inertia when approaching the target temperature to avoid overshoot and ensure stable temperature convergence. Dynamically adjust curve parameters based on the initial temperature difference to achieve personalized temperature control. Extend the main heating phase duration for welding heads in the "far reference group" (initial temperature difference > 3℃), calculated using the following formula: ( (Based on the main heating time of the benchmark group), ensuring sufficient catching up. For weld heads "near the benchmark group" (initial temperature difference ≤ 3℃), calculate the buffer trigger time in advance. To avoid premature deceleration that could disrupt the calling and comparison process, the system generates a visual target curve for each welding head, stored in a curve library. This curve includes parameters such as target temperature at various times, segmented slope, and switching timing, and supports real-time calling and comparison.

[0040] This stage achieves three major breakthroughs through a logical chain of initial state perception, benchmark establishment, and trajectory generation: First, it solves the problem that fixed curves cannot cope with discrete states by adapting to the differentiated initial temperature difference; second, it balances the synchronization rate and overshoot risk through segmented slope design; and third, the output dynamic target curve directly serves as the reference benchmark for subsequent real-time control, achieving seamless connection from planning to execution and providing accurate trajectory guidance for synchronous temperature control.

[0041] After the dynamic target curve is generated, the system enters the real-time control stage. The primary task is to continuously monitor the actual temperature rise of each welding head and compare it with the preset rate of change of the target curve to provide a basis for subsequent adjustments.

[0042] The real-time temperature of each welding head is obtained through high-frequency temperature sampling (sampling frequency ≥ 1kHz). Noise was smoothed using a moving average filter (window size 5). The actual temperature change rate was calculated based on the temperature difference between two consecutive sampling times. The formula is: ; in, The sampling interval is fixed at 1 ms to ensure the accuracy of the rate of change calculation is ±0.1℃ / s. Real-time calculation of temperature change acceleration. It is used for early warning of thermal runaway.

[0043] Synchronously extract the target change rate of the dynamic target curve at the corresponding time. (The main heating section is the reference rate) The buffer segment is (0.3 0.4 The target value at any given time is obtained through an interpolation algorithm. The deviation of the rate of change between the two values ​​is calculated in real time. ; Real-time assessment of whether intervention or adjustment is needed, when When the deviation reaches the first deviation threshold, it is considered to be in a normal state; if it exceeds the threshold, it is marked as an abnormal deviation and the adjustment mechanism is triggered.

[0044] When the deviation between the actual temperature change rate and the target value exceeds the threshold, the system automatically activates the parameter tuning mechanism to ensure temperature control accuracy.

[0045] when At that time, the PID parameters are triggered for real-time tuning, enabling precise adjustment of heating power, rapid correction of deviations, and prevention of synchronization failure due to accumulated deviations. For extreme deviation scenarios ( (Second deviation threshold)) Activate emergency adjustment mode, temporarily relax the PID parameter range and limit the single power adjustment amplitude to ≤20%.

[0046] An incremental PID algorithm is used to control the output: ; in, This is the deviation value. (Proportionality coefficient, initial value 3.0) (Integral coefficient, initial value 0.5) (Derivative coefficients, initial value 1.0) are PID parameters. An integral separation strategy is introduced: when... When the integral term is paused at the (integral pause threshold), the deviation is quickly suppressed using only proportional and derivative adjustments.

[0047] The system dynamically adjusts the proportional gain of the PID controller based on the absolute value of the deviation. Integral coefficient and differential coefficients ,when (Third deviation threshold, large deviation) When increasing (To 4.0), Reduce (Up to 2.0) to expedite response; when (At moderate deviation) maintain the initial parameters; when (Small deviation) decrease (Up to 2.0), Increase (Up to 0.8) to eliminate steady-state error. Parameter adjustment rules are pre-stored in the control algorithm library and can be dynamically modified via a host computer.

[0048] The heating power is adjusted by regulating the output signal after PID parameter tuning, so that the actual temperature change rate of each welding head matches the target curve, ultimately achieving synchronous temperature reach. The tuned PID output signal... This is converted into a heating power command (adjustable from 0-100% full power, controlling the power supply duration of the welding head heating element through PWM pulse width modulation technology, with a power resolution ≤1%, for example, outputting a 50% command corresponds to the heating element operating at half power). The power drive module controls the output power of the welding head heating element, correcting the actual heating rate in real time. Within a preset time window of 0.5-3 seconds, it ensures that the deviation between the actual temperature and the target temperature of all welding heads is ≤±2℃, the temperature difference between any two welding heads is ≤±2℃, the temperature difference overshoot is ≤3℃, and the fallback time is ≤0.5 seconds, meeting the temperature synchronization requirements of the hot riveting process. The current power output is: ; When temperature change accelerates Furthermore, the power is reduced by 20% in advance for three consecutive cycles. To avoid system oscillation caused by sudden power changes, an upper limit for the single power adjustment range (≤10% of rated power) and upper and lower limits for output power are set (lower limit 5% of rated power, upper limit 110% of rated power). The adjustment command is sent to the actuator via a high-speed communication bus (response time ≤5ms) to ensure timely control.

[0049] The system performs a monitoring, comparison, and adjustment cycle every 10ms, eliminating steady-state deviations through a cumulative error compensation algorithm. One cycle ( The sum of instantaneous deviations (i.e., 50ms): ; choose It can balance response speed and noise resistance, avoiding the impact of single-cycle data fluctuations on judgment. When When the average deviation over five consecutive cycles is greater than 0.3℃ / s, compensation adjustment is initiated to ensure that small deviations do not accumulate. Compensation coefficient. The power duty cycle needs to be adjusted additionally for every 1℃ / s of cumulative deviation, calibrated experimentally. This compensates for the power increment. positive Corresponding power increase, negative The corresponding power decreases. The compensation amount is then added to the PID control result to obtain the final power output. ,in Power output calculated by the PID algorithm for the current cycle (range 0-100% of rated power). Maximum compensation amount per cycle. Rated power to prevent temperature fluctuations caused by sudden power changes. Automatic reset when the welding head temperature reaches the target temperature within ±2℃. =0, stop compensation to prevent over-adjustment. Total power after compensation. The output power must not exceed 110% of the rated power. If it exceeds the limit, the output power will be set to the upper limit and the deviation will be recorded. Through cumulative error compensation, the temperature deviation at the end of the target time window will be reduced from ±1.5℃ to ±0.8℃, ensuring that the synchronization requirement of "the temperature deviation between each welding head and the target temperature ≤ ±2℃ and the temperature difference between welding heads ≤ ±2℃" is met.

[0050] For example, if a welding head has a positive deviation of 0.3℃ / s for five consecutive control cycles (the actual rate of change is lower than the target value), the system will add 0.5% power compensation on the basis of PID parameter tuning to gradually reduce the deviation to the threshold range.

[0051] To prevent equipment damage caused by abnormal heating power, multiple protection mechanisms are implemented. The heating power of a single welding head will not exceed 110% of its rated power (obtained from the equipment parameter library). When the welding head temperature exceeds the target temperature +5℃, the heating power will be forcibly cut off and an alarm will be triggered. If the maximum power output still cannot meet the target rate of change for 10 consecutive cycles (100ms), the equipment is deemed abnormal, the welding head's operation will be suspended, and maintenance will be requested. For welding heads that fail to meet standards, a root cause analysis dimension is added: distinguishing between fault types such as "environmental interference," "thermal capacity drift," and "thermal coupling," and providing targeted maintenance suggestions.

[0052] At the end of the preset time window (0.5-3 seconds), the system automatically verifies the temperature of each welding head. All welding head temperatures that meet the requirement of "deviation from target temperature ≤ ±2℃ and temperature difference between welding heads ≤ ±2℃" are marked as "synchronously met". If any welding head fails to meet the standard, the deviation value is recorded and the cause is analyzed (e.g., excessive initial temperature difference, heat capacity modeling error, etc.), providing data support for subsequent process optimization.

[0053] This process achieves three core values ​​through a closed-loop workflow of real-time monitoring, deviation adjustment, and power control: First, 1kHz high-frequency sampling and 10ms fast adjustment ensure real-time response to temperature changes, solving the problem of lag in traditional control; second, dynamic tuning of PID parameters based on rate of change deviation improves temperature control accuracy under complex operating conditions; and third, multiple safety constraints and synchronous verification mechanisms ensure equipment operation safety and product quality stability, ultimately achieving high-precision synchronous temperature control for multiple heterogeneous welding heads.

[0054] Example 1: A certain automotive parts hot riveting production line is equipped with four hot riveting welding heads of different brands (numbered H1-H4), which have problems such as large differences in heat capacity (measured difference reaches 320%) and significant initial temperature differences. This embodiment, based on the technical steps of the solution, verifies the effect of synchronous temperature control through specific data, and the target temperature. The temperature must be reached synchronously within 3 seconds (temperature difference ≤ ±2℃).

[0055] Select heating pulse duration (i.e., 0.2s), a 220V full-power signal is output through the relay module, and the pulse start timestamp is recorded synchronously. End timestamp The timing accuracy is 0.5ms. The temperature and humidity sensor collects ambient temperature data. The relative humidity (RH) is 60%. Trigger pulses are applied according to the sequence H1-H2-H3-H4, with a 5-second interval.

[0056] Taking H1 as an example: Initial temperature sampling 100ms before pulse - The average temperature was 30.2℃, and the temperature was 5ms after pulse interpretation. .

[0057] Calculate adjacent temperature differences ,Discover (A jump of 5.3℃) is identified as noise. Window data. After sorting After median replacement After filtering , , H1 is close to the fan, position coefficient , .

[0058] H1 represents brand A. Environmental correction k-value Substituting the equivalent heat capacity formula into... .

[0059] The welding head H2 is calculated by analogy. H2 represents other regions. , For brand B, , Welding head H3 H2 is near the fan , For brand A, , Welding head H4 H4 is a closed region. , For brand B, , .

[0060] Three seconds after heat capacity identification, the initial temperature of each welding head is collected. , , , Perform temperature difference calculations. Grouping was performed, with H4 designated as the "far baseline group" (based on the lowest temperature difference). The rest are "near the baseline group".

[0061] H4 Selected as the reference channel The simulation of the 8-head array yielded... , , H4 rated power ,but .

[0062] Main heating section formula H4 (far reference group) main heating time ( (Take 1). Trigger temperature difference ℃ (target 180℃), trigger temperature . Buffer segment trajectory .

[0063] H4 is in Real-time temperature , Real rate of change After moving average correction Main heating section Instantaneous deviation .

[0064] Cumulative error (M=5 cycles) No compensation is required.

[0065] H2 in Extreme deviations are handled in a timely manner. , , Trigger PID control. It was downgraded from 3.0 to 4.0. Reduced to 2.0. Power output correction. Power consumption decreased by 4.5%.

[0066] Temperatures were measured after 3 seconds: H1=179.2℃, H2=180.5℃, H1=178.9℃, H1=180.1℃. The maximum temperature difference from the target was 1.1℃, and the maximum temperature difference between the welding heads was 1.6℃, which meets the ±2℃ requirement.

[0067] Example 2: A new energy vehicle power battery pack production line needs to achieve IP67-level annular sealing welding between the top cover (glass fiber reinforced PP composite material, 3mm thick) and the casing (6061 aluminum alloy, 5mm thick). The weld circumference is 1200mm, and the tensile strength after welding is required to be ≥25MPa. Six hot riveting welding heads of different brands (H1-H6) are configured. H1, H3, and H5 are installed on the left side of the production line (near the cooling fan, wind speed 1.2m / s), and H2, H4, and H6 are installed on the right side (enclosed area, ambient temperature is 3-5℃ higher than the left side). All welding heads have a full power of 30W, and the actual measured difference in equivalent heat capacity is 315% (H4 maximum 7.0J / ℃, H3 minimum 2.2J / ℃). They need to reach 185℃ synchronously within 3 seconds, and the temperature difference between any two welding heads should be ≤±2℃.

[0068] During implementation, H1-H6 were first independently triggered with full-power heating pulses. According to the scheme requirements, the heating pulse duration was strictly controlled within the range of 150-250ms, preferably 200ms ± 10%, and actually set to 200ms. A 220V ± 5% voltage signal was output through the relay module (measured voltage of H1: 218V; measured voltage of H4: 222V). Synchronization was performed with an accuracy ≤ 1. The STM32H743 clock chip records the start and end timestamps of the pulses. The start timestamp of H1 is 0.000000s, and the end timestamp is 0.200001s. The pulse duration calculation accuracy is ≤1ms. To avoid heat accumulation, the pulse interval between adjacent welding heads is set to 5.2 seconds (in actual testing, the initial temperature of H2 increased by 0.8℃ due to the influence of H1 when the interval was 5 seconds; after adjusting to 5.2 seconds, there was no temperature interference). Simultaneously, an SHT30 temperature and humidity sensor (sampling frequency 1Hz, accuracy ±1℃ / ±5%RH) is deployed to collect real-time ambient temperature (25.8℃) and relative humidity (52%). The pulse execution status is fed back in real-time through a red indicator light (not executed), a green indicator light (executed), and the software log.

[0069] Throughout the entire heating pulse application period, temperature data was collected using the NTC temperature sensor built into the welding head (temperature range -50℃ to 300℃, accuracy ±0.5℃, sampling frequency 1kHz, data transmitted via shielded cable): For H1, the average of 100 samples within 100ms before the pulse started was taken as the initial temperature, calculated to be 30.2℃ (sampling range 29.9-30.5℃); the instantaneous temperature within 5ms after the pulse ended was taken as the final temperature, measured at 37.5℃ (sampling range 37.4-37.6℃), with an initial temperature rise of 7.3℃. For H4, the initial temperature was 34.5℃ (sampling range 34.3-34.7℃), the final temperature was 39.2℃ (sampling range 39.1-39.3℃), and the initial temperature rise was 4.7℃. During the data collection process, H2 experienced spike noise due to electromagnetic interference from the production line. The temperature at the 142nd sampling point jumped from 36.9℃ to 42.2℃, with an adjacent temperature difference of 5.3℃. The adaptive median filtering algorithm was used to process this noise point: one normal sampling point before and after the noise point (36.9℃, 37.1℃) was extracted to form a 3-point window dataset {36.9℃, 42.2℃, 37.1℃}. After sorting, the median of 37.0℃ was used to replace the noise point. After correction, the maximum adjacent temperature difference of H2 was 4.8℃ ≤ 5℃ (the spike detection threshold of the algorithm). ), to ensure effective filtering.

[0070] To address the differences in heat dissipation at different welding head locations, position coefficients are preset according to the plan: position coefficients for H1, H3, and H5 on the left side near the fan. =1.1, the position coefficients of H2, H4, and H6 in the closed region on the right. =0.9, multiply the filtered original temperature rise by the position coefficient to obtain the actual temperature rise: H1=7.3×1.1=8.03℃, H4=4.7×0.9=4.23℃. Then, based on the collected pulse parameters and temperature rise data, the equivalent heat capacity is calculated. According to the scheme formula, the heat conversion efficiency constant is first corrected. H1, H3, and H5 represent brand A, based on... =120J, substituting the environmental parameters, we get... =120×[1+0.002×(25.8-25)+0.001×(52-50)]=120.432J; H2, H4, and H6 are brand B, benchmark =150J, after correction =150×[1+0.002×0.8+0.001×2]=150.54J. Then, according to the equivalent heat capacity formula of the scheme... ( Calculations: H1 = 120.43 × 0.2 / 8.03 ≈ 3.00 J / ℃, H4 = 150.54 × 0.2 / 4.23 ≈ 7.11 J / ℃. Since this is the first startup of the day, the system automatically executes the "rapid calibration" scheme, repeating the pulse test on H1 three times. The average equivalent heat capacity is 3.02 J / ℃, with an error ≤ 1%. H4, having been welded 1023 times (over 1000 times) since the last calibration, undergoes "deep calibration." Comparing it with a new welding head of the same model (known heat capacity 7.0 J / ℃), the value is corrected to 7.08 J / ℃ and updated to the equivalent heat capacity parameter library.

[0071] After heat capacity identification, wait 3.1 seconds (in actual testing, the temperature of H4 still fluctuated by 0.3℃ after 3 seconds; extending the wait by 0.1 seconds ensured temperature stability). Continuously collect 10 temperature data points for each welding head at a frequency of 100Hz and take the average as the initial temperature: H1 = 30.5℃ (30.3-30.7℃), H4 = 34.8℃ (34.6-35.0℃). Calculate the initial temperature difference between any two welding heads. The maximum difference between H3 (29.9℃) and H4 (34.8℃) is 4.9℃. According to the scheme rules, the welding heads are divided into "near reference group" (H1, H3, H5, temperature difference from the lowest initial temperature ≤ 3℃) and "far reference group" (H2, H4, H6, temperature difference from the lowest initial temperature > 3℃). Select H4, with the largest equivalent heat capacity, as the reference channel. Establish a 6-welding-head thermodynamic model using COMSOL (mesh resolution 0.1mm) and obtain the thermal coupling coefficient matrix. =0.05、 =0.06、 =0.04、 =0.05、 =0.07, combined with H4's full power of 30W, the global baseline temperature rise rate is calculated according to the scheme formula: =30 / [7.08×(1+0.05+0.06+0.04+0.05+0.07)]=30 / (7.08×1.27)≈3.33℃ / s.

[0072] Subsequently, a dynamic target temperature curve was constructed, and the trajectory of the main heating segment was generated according to the scheme formula: H1 was T=30.5+3.33t, and H4 was T=34.8+3.33t; for H4 of the far reference group, the main heating time was extended according to the scheme, and the calculation formula was as follows: =1.2+4.9 / 3.33≈2.68 seconds; When the welding head temperature approaches the target temperature of 185℃, the buffer section trigger temperature difference is set to 10℃ according to the scheme (i.e., switching when the temperature reaches 175℃), and the buffer section rate is 0.35 times the base rate, i.e. 1.16℃ / s. Through COMSOL simulation prediction, the temperature at the end of the H4 buffer section is 184.9℃, the overshoot is ≤0.1℃, and the curve parameters are effective.

[0073] After the system enters the real-time control phase, it collects the real-time temperature of each welding head at a high-frequency sampling frequency of 1kHz. A moving average filter with a window size of 5 is used to smooth noise. The actual temperature change rate is calculated based on the temperature difference between two consecutive sampling times (1ms interval), with an accuracy of ±0.1℃ / s. H3 previously experienced an abnormal actual temperature change rate (300℃ / s) due to a loose sensor; after re-fixing, it recovered to 3.3℃ / s. Under normal operating conditions, H6, due to poor heat dissipation in the enclosed area, had an actual temperature change rate of 3.9℃ / s at t=2.2 seconds, while the target rate was 3.32℃ / s, resulting in a deviation of 0.58℃ / s > 0.5℃ / s (the first deviation threshold of the scheme). The PID parameters were then tuned in real-time according to the scheme: proportional coefficient... The integral coefficient has been adjusted from 3.0 to 3.8. Adjusted from 0.5 to 0.3, differential coefficient Keeping the value 1.0 constant, the incremental power regulation signal is calculated using an incremental PID algorithm: =3.8×(0.58-0.45)+0.3×0.58+1.0×(0.58-2×0.45+0.32)=0.668%, that is, the power adjustment increment is 0.668%. Through PWM pulse width modulation technology, the power of H6 is further reduced by 0.668% from the current output of 29.5W (which was already slightly lower than the full power of 30W due to the basic adjustment). The actual adjusted power = 29.5W-(29.5W×0.668%)≈29.30W. The single adjustment range is ≤10% of the rated power (30W×10%=3W), which meets the requirement of "upper limit of single power adjustment range" of the scheme.

[0074] Then, the cumulative instantaneous deviation for five consecutive cycles (50 ms) was calculated: the deviations for each cycle of H6 were 0.58℃ / s, 0.52℃ / s, 0.48℃ / s, 0.45℃ / s, and 0.42℃ / s, respectively, with a cumulative error of... =0.58+0.52+0.48+0.45+0.42=2.45℃ / s>1.5℃ / s (Scheme cumulative error threshold). Based on the scheme compensation coefficient. =0.005 Calculate the compensation amount: =0.005×2.45=0.01225 (i.e. 1.225%), after being superimposed on the PID adjustment result, the final power of H6 =29.30W+(29.30W×1.225%)≈29.66W, and the single compensation amount ≈0.36W≤2% of the rated power (30W×2%=0.6W), which meets the constraint of "upper limit of single compensation amount" of the scheme.

[0075] During production, H2 experienced a sudden malfunction of its cooling fan, causing the temperature to spike to 187℃ (over 185℃ + 5℃). The system's multiple protection mechanisms were immediately triggered: the heating power was cut off within 10ms, and an audible and visual alarm was activated. Production resumed after the fan was replaced. After the 3-second synchronization window, the measured temperatures of each welding head were: H1=184.8℃, H2=184.9℃, H3=184.5℃, H4=185.1℃, H5=184.7℃, and H6=184.9℃, with a maximum temperature difference of 0.6℃ ≤ ±2℃, indicating that synchronization was achieved. Post-weld sampling inspection: 10 products underwent IP67 testing (immersion in 1m water for 30 minutes), and all showed no leakage. The average tensile strength was 27.3MPa, meeting production requirements. All steps in the solution were essential for achieving the standard, with no redundant steps.

[0076] Example 3: A Tier 1 supplier's VCU housing production line needs to connect an ADC12 die-cast aluminum housing (150×100×50mm) to a PA66+GF30 end cap (2mm thick). The requirements are airtightness ≤0.5kPa / 5min, vibration test (10-2000Hz, 10g) with no loosening, and four welding heads (H7-H10) configured. H7 and H9 are close to the heat dissipation holes (22±1℃), and H8 and H10 are located inside the housing (25±1℃). The full power is 28W, the equivalent heat capacity difference is 290%, and the temperature needs to reach 170℃ synchronously within 2.5 seconds with a temperature difference ≤±2℃.

[0077] During implementation, a 200ms full-power pulse (H7 voltage 219V, H10 voltage 221V) was first applied to H7-H10, with an interval of 5.1 seconds. Temperature and humidity sensors recorded 23℃ and 58%RH. NTC sensor data showed: H7 initial temperature 29.3℃, ending temperature 36.1℃, initial temperature rise 6.8℃; H10 initial temperature 32.5℃, ending temperature 37.3℃, initial temperature rise 4.8℃. H8 exhibited 5.2℃ noise; after median filtering correction, the noise was calculated based on the position coefficients (H7 and H9 position coefficients). =1.1, H8, H10 position coefficients =0.9) Calculate the actual temperature rise: H7=7.48℃, H10=4.32℃.

[0078] Corrected heat conversion efficiency constants: H7, H9 (Brand D, =125J) k=125.5J; H8, H10 (Brand E, =145J)k=145.58J. Calculate the equivalent heat capacity according to the formula: H7=125.5×0.2 / 7.48≈3.36J / ℃, H10=145.58×0.2 / 4.32≈6.74J / ℃, H7 is 3.42J / ℃ due to depth calibration after more than 1000 welding cycles.

[0079] Three seconds after thermal capacity identification, the initial temperatures H7 = 29.6℃ and H10 = 32.8℃ were collected, with a temperature difference of 3.2℃. H10 was designated as the far reference group. H10 was selected as the reference channel, and the thermal coupling coefficient was obtained using COMSOL. =0.04、 =0.05、 =0.03, calculate =28 / [6.74×(1+0.04+0.05+0.03)]≈3.71℃ / s. Dynamic curve construction: H10 main heating time 2.6 seconds, 160℃ switching buffer segment (rate 1.29℃ / s).

[0080] During the real-time control phase, the rate of change is calculated after 1kHz sampling and filtering: H9 has a deviation e(t) of 0.6℃ / s at t=1.8 seconds, triggering PID tuning. =3.5, =0.4, =1.0), calculated using incremental PID ( =0.5℃ / s, =0.4℃ / s): =3.5×(0.6-0.5)+0.4×0.6+1.0×(0.6-2×0.5+0.4)=0.35+0.24+0.0=0.59%, reducing the power of H9 from 28W to 28-28×0.59%≈27.8348W. The cumulative error of H10 over 5 consecutive cycles is 1.6℃ / s, with a compensation of 0.005×1.6=0.8%. The combined power is approximately 28+28×0.8%≈28.224W. Temperatures after 2.5 seconds: H7=169.7℃, H10=170.1℃, temperature difference 0.4℃. The airtightness pass rate after welding is 99.8%, and the vibration test shows no faults.

[0081] The processes described above with reference to the flowcharts in the embodiments disclosed in this invention can be implemented as computer software programs. The embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wire segments, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical fibers, RF, etc., or any suitable combination thereof.

[0082] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0083] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been shown and explained in the embodiments. Without departing from the stated principles, the implementation of the present invention may have any variations or modifications.

Claims

1. A temperature change rate-based hot riveting horn synchronous temperature control method, characterized by, The method includes: Each welding head to be synchronized is independently triggered by a full-power heating pulse. The pulse start time and end time are recorded. The pulse duration is calculated based on the start and end times. The temperature rise is calculated based on the initial temperature at the start of the pulse and the real-time temperature at the end of the pulse. The actual temperature rise is corrected by combining the position coefficient and the position of each welding head. Peak noise is removed by digital filtering algorithm. The ambient temperature and relative humidity are collected in real time, and the equivalent heat capacity of each welding head is calculated by combining the actual temperature rise and pulse duration. The initial temperature difference between any welding heads is calculated using the initial temperature of the welding head. The welding heads are then classified according to the initial temperature difference. The welding head with the largest equivalent heat capacity is selected as the reference channel, and the global reference rate is calculated by combining the thermal coupling coefficient matrix. A thermodynamic model of the welding head is established using COMSOL, and a dynamic target temperature curve is constructed. The real-time temperature of each welding head is collected to calculate the actual temperature change rate. The target change rate of the dynamic target curve at the corresponding time is extracted simultaneously to calculate the deviation between the two change rates. Based on the change rate deviation, it is determined whether intervention and adjustment are required. If intervention and adjustment are required, the PID controller coefficient is dynamically adjusted according to the absolute value of the deviation. The output signal is calculated using an incremental PID algorithm, and the heating power is adjusted according to the output signal.

2. The temperature rate of change based hot rivet horn synchronized temperature control method of claim 1, wherein, The specific process for correcting the actual temperature rise is as follows: The average temperature within a sampling period before the pulse starts is taken as the initial temperature, and the instantaneous temperature within a sampling period after the pulse ends is taken as the final temperature. The difference between the two is calculated to obtain the original temperature rise. For the marked suspected noise points, a digital filtering algorithm is used to filter them to obtain the original temperature rise after filtering. The heat dissipation level is divided according to the location of the welding head, and a corresponding position coefficient is set for the different heat dissipation level areas. The original temperature rise after filtering is multiplied by the corresponding position coefficient to obtain the final actual temperature rise.

3. The temperature rate of change based hot rivet horn synchronized temperature control method of claim 2, wherein, The specific details of the digital filtering algorithm are as follows: Traverse the temperature sampling sequence, calculate the temperature difference between adjacent sampling points, and mark the sampling point as a suspected spike noise point when the temperature difference between adjacent sampling points exceeds the spike determination threshold. For each marked noise point, extract several normal sampling points before and after it to form a dataset. After sorting the window data, take the median value to replace the noise point. Normal points not marked as noise are retained in their original values; edge points at the beginning and end of the sequence are corrected for noise by using the mean of the valid data from one side. After correction, the temperature difference stability of the filtered sequence is calculated. If all adjacent temperature differences are less than or equal to the peak determination threshold, the filtering is effective. Otherwise, the expanded window is repeatedly corrected until the condition is met.

4. The method for synchronous temperature control of a hot riveting head based on the rate of temperature change according to claim 3, characterized in that, The specific calculation process for the equivalent heat capacity is as follows: The equivalent heat capacity of the welding head is calculated using the heat conversion efficiency constant, temperature rise, power, and pulse duration; the heat conversion efficiency constant is corrected based on the ambient temperature and relative humidity; and the equivalent heat capacity of the welding head is then corrected based on the corrected heat conversion efficiency constant. The system automatically performs a rapid calibration upon its first startup each day, performs a depth calibration according to the preset welding interval, and updates the equivalent heat capacity parameter library.

5. The method for synchronous temperature control of a hot riveting head based on the rate of temperature change according to claim 4, characterized in that, The specific process for obtaining the dynamic target temperature curve is as follows: After the heat capacity identification is completed, several temperature data are continuously collected at a fixed frequency and the average value is taken as the initial temperature of each welding head. The initial temperature difference between any two welding heads is calculated. After determining the maximum initial temperature difference, the welding heads are divided into near reference group and far reference group. The welding head with the largest equivalent heat capacity is selected as the reference channel. A thermal coupling coefficient matrix is ​​introduced, and the global reference heating rate is calculated in combination with the full power parameters of the reference channel. The dynamic target temperature curve includes a main heating segment and a buffer segment. The main heating segment generates the target trajectory with the reference heating rate as the slope. For welding heads with extreme temperature differences, the main heating time is extended. When the welding head temperature approaches the target temperature, it automatically switches to the buffer segment. A thermodynamic model is established using COMSOL to predict thermal coupling effects and optimize the parameters of the buffer segment.

6. The method for synchronous temperature control of a hot riveting head based on the rate of temperature change according to claim 5, characterized in that, The intervention and adjustment judgment process is as follows: After the system enters the real-time control stage, the real-time temperature of each welding head is collected at a high-frequency sampling frequency. After smoothing the noise by using a moving average filter, the actual temperature change rate is calculated based on the temperature difference between two consecutive sampling times to ensure that the accuracy of the actual temperature change rate calculation meets the requirements. Simultaneously extract the target change rate of the dynamic target temperature curve at the corresponding time, and obtain the target value at any time through an interpolation algorithm; The absolute value of the deviation between the actual rate of change and the target rate of change is calculated in real time; when the absolute value of the deviation is less than or equal to the first deviation threshold, it is determined to be a normal state and no intervention or adjustment is required. When the absolute value of the deviation is greater than the first deviation threshold, it is marked as an abnormal deviation, triggering the real-time PID parameter tuning mechanism and initiating intervention and adjustment; when the absolute value of the deviation is greater than the second deviation threshold, the emergency adjustment mode is activated.

7. The method for synchronous temperature control of a hot riveting head based on the rate of temperature change according to claim 6, characterized in that, The specific calculation process for the output signal is as follows: When the absolute value of the deviation is greater than the third deviation threshold, the proportional coefficient is increased and the integral coefficient is decreased to speed up the response. When the first deviation threshold is less than the absolute value of the deviation and less than the third deviation threshold, the initial parameters are maintained. When the absolute value of the deviation is less than the first deviation threshold, the proportional coefficient is decreased and the integral coefficient is increased to eliminate steady-state error. When the absolute value of the deviation is greater than the integral pause threshold, the integral term is paused, and the deviation is quickly suppressed by using only proportional and derivative control. The incremental power adjustment incremental signal is calculated using an incremental PID algorithm, which is then converted into a heating power command. The power supply duration of the welding head heating element is controlled by PWM pulse width modulation technology. The cumulative error is calculated by summing the instantaneous deviations over several consecutive cycles. When the absolute value of the cumulative error is greater than the third deviation threshold, the power compensation amount is calculated by the compensation coefficient and superimposed on the PID output signal. The single compensation amount is ≤2% of the rated power. The final output signal must ensure that the power output is less than or equal to the upper limit of the rated power. If it exceeds the limit, the output should be based on the upper limit and the deviation should be recorded.

8. A synchronous temperature control system for hot riveting heads based on temperature change rate, characterized in that, The system is used to execute the synchronous temperature control method for hot riveting heads based on the rate of temperature change as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that is executed by a processor to implement the synchronous temperature control method for hot riveting heads based on the rate of temperature change as described in any one of claims 1-7.

Images