A heat treatment system and process optimization control method applied to a vehicle brake disc

Abstract

The application relates to a heat treatment system applied to a vehicle brake disc and a process optimization control method, which comprises a rack, a lifting mechanism arranged at the center of the top of the rack, a first induction coil and a second induction coil arranged on the output shaft of the lifting mechanism, the first induction coil being coaxially arranged with the second induction coil and located directly below the second induction coil, the two induction coils being electrically connected with an intelligent control system respectively, a rotary speed reduction motor being arranged directly below the induction coil, a workpiece clamping disc for clamping the brake disc being arranged on the rotary speed reduction motor, a spraying swing arm being arranged on the side of the rotary speed reduction motor, a spraying disc being arranged at one end of the spraying swing arm, the other end of the spraying swing arm being connected with the output shaft of a swing arm rotary motor, and the spraying disc being communicated with a water pump through a hose. The application not only ensures that the brake disc can reach the best performance and quality in the manufacturing process, but also improves the safety and reliability of the brake system.

Description

Technical Field

[0001] This invention relates to the field of vehicle brake disc processing technology, and in particular to a heat treatment system and process optimization control method for vehicle brake discs. Background Technology

[0002] Currently, with the rapid development of the automotive industry, the braking system, as a key component for vehicle safety, is receiving increasing attention for performance improvement. As consumers' demands for vehicle safety continue to rise, the market demand for brake discs continues to grow. Meanwhile, domestic brake disc manufacturers have made significant improvements in technology and production capacity, but a certain gap still exists compared to international advanced levels. The brake disc is the most critical safety component, and its replacement is closely related to driving habits. Sudden braking is the most damaging to brake discs. Factors affecting braking force include the size and type of brake disc. When a driver encounters an emergency and presses the brake pedal, the brake fluid in the lines is pushed to the brake calipers, causing them to squeeze the brake discs and generate braking force, ultimately achieving deceleration or stopping.

[0003] Brake discs, as a crucial component of the automotive braking system, directly impact vehicle safety and stability through their performance and quality. However, traditional brake disc manufacturing processes suffer from several drawbacks, such as unstable heat treatment processes, low material utilization, and low production efficiency. Furthermore, the traditional quenching and heating process for brake discs lacks zoned heating for different areas and suffers from low automation, leading to repeated heating of the same area or insufficient heating of edge areas. This not only wastes energy but also affects the overall processing technology of the brake disc. Therefore, optimizing the heat treatment process for brake discs to ensure product quality has become a pressing issue that needs to be addressed. Summary of the Invention

[0004] In view of the above problems, the present invention provides a heat treatment system and process optimization control method for vehicle brake discs, which not only ensures that the brake discs can achieve optimal performance and quality during the manufacturing process, but also improves the safety and reliability of the braking system.

[0005] To achieve the above and other related objectives, the present invention provides the following technical solution: A heat treatment system for vehicle brake discs includes a frame. A lifting mechanism is located at the center of the top of the frame. The output shaft of the lifting mechanism is connected to a first induction coil and a second induction coil via an insulating bracket. The first induction coil and the second induction coil are coaxially arranged, with the first induction coil located directly below the second induction coil. A rotary gear motor is located at the center of the bottom of the frame. The output shaft of the rotary gear motor is equipped with a workpiece clamping plate, which is located directly below the first induction coil. A swing arm rotary motor is located on the side of the frame. The output shaft of the swing arm rotary motor is connected to one end of a spray swing arm. The other end of the spray swing arm is equipped with a spray plate, which is connected to a water pump via a hose. The water pump is located at the bottom of the frame. The first induction coil, the second induction coil, the rotary gear motor, the swing arm rotary motor, and the water pump are all electrically connected to an intelligent control system.

[0006] Furthermore, the intelligent control system includes a heat treatment process parameter acquisition module, a prediction module, an optimization module, and a temperature control and adjustment module. The heat treatment process parameter acquisition module is connected to a temperature sensor signal disposed on the surface of the brake disc. The output terminal of the temperature control and adjustment module is connected to the intermediate frequency power supply of the first induction coil and the second induction coil through a power controller.

[0007] Furthermore, a coolant recovery tank is provided at the bottom of the frame, and the inlet of the coolant recovery tank is connected to the drain end of the spray plate.

[0008] Furthermore, the workpiece clamping plate is provided with 8 temperature sensor mounting positions, of which 4 are located in the central area with a radius of 0-80mm and 4 are located in the edge area with a radius of 80-160mm. The first induction coil and the second induction coil are coaxially arranged and an insulating layer is provided between them.

[0009] To achieve the above and other related objectives, the present invention also provides a method for optimizing and controlling the heat treatment process of a vehicle brake disc, applied to the aforementioned heat treatment system for vehicle brake discs, the method comprising: First, the heat treatment process parameter acquisition module collects real-time data on the quenching heating temperature, the temperature after quenching, and the rate of temperature change within the same processing cycle of the brake discs of the same batch of vehicles. Secondly, the temperature of the brake disc quenching and heating of different vehicles at the next moment is predicted by the improved Gaussian process regression prediction algorithm based on dynamic learning in the prediction module, and the data information of the temperature of the brake disc quenching and heating of different vehicles at the next moment is obtained. Then, the temperature of vehicle brake disc quenching heating is optimized by the covariance matrix adaptive evolution algorithm for global search in the optimization module; Finally, the vehicle brake disc quenching heating temperature control function in the temperature control and regulation module, including the first temperature control function F, is used. control1 Second temperature control function F control2 The first temperature control function F control1 The output parameters are used to control the heating power of the first induction coil, and the second temperature control function F control2 The output parameters are used to control the heating power of the second induction coil.

[0010] Furthermore, the improved Gaussian process regression prediction algorithm based on dynamic learning predicts the quenching and heating temperature of different vehicle brake discs at the next moment, including: Q1. Based on the data information of the temperature of the brake disc quenching heating of the same batch of vehicles and the data information of the temperature change rate within the same period, normalization processing is performed to obtain the normalized data information of the temperature of the brake disc quenching heating of the same batch of vehicles and the data information of the temperature change rate within the same period. Q2. Based on the normalized data of the temperature during the quenching and heating of brake discs in the same batch of vehicles and the data of the rate of temperature change within the same period, construct a covariance function W. , Where x represents the normalized temperature data of the brake disc quenching and heating of the same batch of vehicles, and Δx represents the normalized temperature change rate data of the same batch of vehicle brake disc quenching and heating within the same period. Q3. Construct the Gaussian process function H. , Where α and β are any constant parameters between 0 and 1, the temperature of the brake disc quenching heating of different vehicles at the next moment is predicted, and the data information of the temperature of the brake disc quenching heating of different vehicles at the next moment is obtained.

[0011] Furthermore, the constraint condition for the constant parameters α and β is (α + β). 2 +ɑ 2 +β 2 ≤1.

[0012] Furthermore, the optimization of the quenching heating temperature of the vehicle brake disc using the covariance matrix adaptive evolution algorithm oriented towards global search includes: M1. Based on the data information of the quenching and heating temperature of different vehicle brake discs at the next moment, the population is initialized, and the initial step size standard deviation δ, population size n, initial mean m, and weight {ω} are determined. i}, thus obtaining the data information of the initialized population; M2. Based on the data information of the initialized population, according to y (t+1) =m+δ t C (t) ∙N(0,1), where y (t+1) Let C be the (t+1)th individual in the population. (t) Let N(0,1) be the covariance matrix and N(0,1) be a standard normal distribution. We then sample individuals from the population to obtain the sampled individuals. M3. Based on the sampled individuals in the population, construct the covariance matrix update function C. (t+1) , , Where, ρ l and ρ c The temperature of vehicle brake disc quenching heating is optimized using the preset covariance update coefficient, resulting in optimized data information on the temperature of vehicle brake disc quenching heating.

[0013] Furthermore, the preset covariance update coefficient ρ l and ρ c The weight {ω} is a random number between 0 and 1. i The constraint condition for} is ω1+ω2+...+ω n =1, where n is a positive integer, ρ l and ρ c The value is calculated according to the standard CMA-ES formula based on the optimization variable dimension r=8 and the population size n=20.

[0014] Furthermore, the first temperature control function F control1 The second temperature control function F is used to control the heating temperature of the center area of ​​the brake disc. control2 Two temperature control functions are used to control the heating temperature of the brake disc edge area. These functions independently adjust the power output of the corresponding induction coils based on feedback from temperature sensors at different locations on the brake disc. The power adjustment range of the first induction coil is 50-100kW, used during the brake disc preheating stage. The power adjustment range of the second induction coil is 100-200kW, used during the brake disc quenching heating stage. The power ratio of the two induction coils is dynamically adjusted according to the real-time temperature distribution of the brake disc, using the following formula: P1:P2=(T center -T edge ):(T edge +k), Where P1 is the power of the first induction coil, P2 is the power of the second induction coil, and T center T represents the temperature of the center area of ​​the brake disc. edge The temperature is the edge area of ​​the brake disc, and K is the zero compensation constant, with a value range of 50-100℃.

[0015] Further, the vehicle brake disc quenching heating temperature control function includes a first temperature control function F control1 and a second temperature control function F control2 , wherein, the first temperature control function F control1 : When a1≤g1≤b1: F control1 =g1, When g1<a1: F control1 =g1∙(1+Δg1) 2 , When g1>b1: F control1 =g1∙(1-Δg1) 3 , the second temperature control function F control2 : When a2≤g2≤b2: F control2 =g2, When g2<a2: F control2 =g2∙(1+Δg2) 2 , When g2>b2: F control2 =g2∙(1-Δg2) 3 , wherein, a1 is a lower temperature limit threshold controlled by the first induction coil, b1 is an upper temperature limit threshold controlled by the first induction coil, g1 is optimized temperature data information corresponding to the first induction coil, Δg1 is a temperature change rate corresponding to the first induction coil, a2 is a lower temperature limit threshold controlled by the second induction coil, b2 is an upper temperature limit threshold controlled by the second induction coil, g2 is optimized temperature data information corresponding to the second induction coil, and Δg2 is a temperature change rate corresponding to the second induction coil.

[0016] Further, when g1 is between a1 and b1 and g2 is between a2 and b2, output the optimized data information of the quenching heating temperature of the vehicle brake disc; when g1 is less than a1, perform temperature increase control according to F control1 =g1∙(1+Δg1) 2 until g1 is between a1 and b1; when g1 is greater than b1, perform temperature decrease control according to F control1 =g1∙(1-Δg1) 3 until g1 is between a1 and b1; when g2 is less than a2, perform temperature increase control according to F control2 =g2∙(1+Δg2) 2 until g2 is between a2 and b2; when g2 is greater than b2, perform temperature decrease control according to F control2=g2∙(1-Δg2) 3 The temperature is controlled to decrease until g2 is between a2 and b2.

[0017] Furthermore, the intelligent control system also includes a voice broadcast module connected to the temperature control and adjustment module, used to broadcast real-time data on the temperature of the vehicle brake disc quenching heating process.

[0018] Furthermore, the intelligent control system also includes a human-machine interaction module connected to the temperature control and adjustment module, used to input the preset lower limit critical value and upper limit critical value of the vehicle brake disc quenching heating temperature, and to display the data in real time.

[0019] Furthermore, the intelligent control system is connected to the swing arm rotary motor and the water pump, respectively.

[0020] The present invention has the following positive effects: 1. This invention connects an induction coil, an intelligent control system, a swing arm rotary motor, and a water pump. Based on real-time data of the heat treatment temperature of the vehicle brake disc, it intelligently controls the heat treatment temperature of the brake disc. Not only does the entire process require no manual intervention, reducing labor intensity, but the intelligent temperature control is also more precise, ensuring brake disc quality and reducing the defect rate.

[0021] 2. This invention employs an improved Gaussian process regression prediction algorithm based on dynamic learning to predict the quenching and heating temperature of brake discs for different vehicles at the next moment. It also combines this with a global search-oriented covariance matrix adaptive evolution algorithm to optimize the quenching and heating temperature of the vehicle brake discs. This not only allows for accurate prediction of the quenching and heating temperature of brake discs in the same batch, ensuring the heating temperature is within the optimal range, but also guarantees that the brake discs achieve optimal performance and quality during manufacturing. This invention addresses the significant limitations of traditional Gaussian process regression in handling rapid phase transition lag issues. The regression model had an average error of 15-20 degrees Celsius when predicting the phase transition start temperature, mainly because it failed to consider the impact of cooling rate on phase transition lag. Furthermore, it found that when the cooling rate exceeded 50 degrees Celsius per second, the prediction accuracy decreased significantly, with lag time prediction errors exceeding 30%. Moreover, this invention overcomes several inherent defects of the traditional CMA-ES algorithm, which restrict its widespread application in industrial scenarios, including the local optimum trap problem. In high-dimensional nonlinear control spaces, algorithms are prone to getting trapped in local optima, leading to decreased control performance and risks of violating safety constraints. Industrial control scenarios have strict requirements for equipment safety, but traditional algorithms lack effective constraint processing mechanisms and are difficult to meet millisecond-level control cycles. The entire process does not require manual intervention, reducing errors caused by manual temperature adjustment and improving the production quality of the entire brake disc production line.

[0022] 3. This invention constructs a vehicle brake disc quenching heating temperature control function F. control This technology controls and adjusts the quenching and heating temperature of vehicle brake discs, enabling precise control of the temperature in various areas of the brake disc quenching and heating process, thereby improving the quality of the brake discs. Furthermore, its high level of intelligence promotes technological innovation and the transformation of research results, driving the development of automotive braking systems towards lightweighting, high performance, and intelligence, and providing strong support for the transformation and upgrading of my country's automotive industry.

[0023] 4. This invention improves the temperature control accuracy of the brake disc heat treatment process and ensures the quality of the brake disc by coordinating the operation of the intelligent control system, the first induction coil, and the second induction coil. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the system framework of the present invention; Figure 2 This is a flowchart illustrating the improved Gaussian process regression prediction algorithm based on dynamic learning of the present invention. Figure 3 This is a flowchart illustrating the adaptive evolutionary algorithm of the covariance matrix for global search according to the present invention. Figure 4 This is a schematic diagram of the intelligent control system framework of the present invention.

[0025] The following are the labels in the diagram: 1-Lifting structure, 2-First induction coil, 3-Brake disc, 4-Rotary reduction motor, 5-Intelligent control system, 6-Human-machine interaction module, 7-Spraying swing arm, 8-Swing arm rotary motor, 9-Spraying disc, 10-Water pump, 11-Workpiece clamping disc, 12-Coolant recovery tank, 13-Frame, 14-Second induction coil. Detailed Implementation

[0026] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0027] Example 1: As Figure 1 or Figure 4As shown, a heat treatment system for vehicle brake discs includes a frame 13. A lifting mechanism 1 is provided at the center of the top of the frame 13. The output shaft of the lifting mechanism 1 is connected to a first induction coil 2 and a second induction coil 14 via an insulating bracket. The first induction coil 2 and the second induction coil 14 are coaxially arranged, with the first induction coil 2 located directly below the second induction coil 14. A rotary reduction motor 4 is provided at the center of the bottom of the frame 13. The output shaft of the rotary reduction motor 4 is provided with a workpiece clamping plate 11, which is located directly below the first induction coil 2. A swing arm rotary motor 8 is provided on the side of the frame 13. The output shaft of the swing arm rotary motor 8 is connected to one end of a spray swing arm 7. The other end of the spray swing arm 7 is provided with a spray plate 9, which is connected to a water pump 10 via a hose. The water pump 10 is located at the bottom of the frame 13. The first induction coil 2, the second induction coil 14, the rotary reduction motor 4, the swing arm rotary motor 8, and the water pump 10 are electrically connected to an intelligent control system 5.

[0028] In this embodiment, the intelligent control system 5 includes a heat treatment process parameter acquisition module, a prediction module, an optimization module, and a temperature control and adjustment module. The heat treatment process parameter acquisition module is connected to a temperature sensor signal disposed on the surface of the brake disc. The output terminal of the temperature control and adjustment module is connected to the intermediate frequency power supply of the first induction coil 2 and the second induction coil 14 through a power controller.

[0029] In this embodiment, the intelligent control system 5 includes a heat treatment process parameter acquisition module, a prediction module, an optimization module, and a temperature control and adjustment module. The heat treatment process parameter acquisition module is connected to a temperature sensor signal mounted on the surface of the brake disc. The output of the temperature control and adjustment module is connected to the intermediate frequency power supply of the first induction coil 2 and the second induction coil 14 via a power controller. Each induction coil may be equipped with an independent intermediate frequency power supply. The power controller, as the execution interface, converts the "power setpoint" of the upper-level control algorithm into an analog quantity (such as 0-10 V) recognizable by the intermediate frequency power supply.

[0030] In this embodiment, a coolant recovery tank 12 is provided at the bottom of the frame, and the inlet of the coolant recovery tank 12 is connected to the drain end of the spray plate 9.

[0031] In this embodiment, the workpiece clamping plate 11 is provided with 8 temperature sensor mounting positions, of which 4 are located in the central area with a radius of 0-80mm and 4 are located in the edge area with a radius of 80-160mm. The first induction coil 2 and the second induction coil 14 are coaxially arranged and an insulating layer is provided between them.

[0032] In this embodiment, the present invention provides a method for optimizing and controlling the heat treatment process of vehicle brake discs, applied to the aforementioned heat treatment system for vehicle brake discs, the method comprising: First, the heat treatment process parameter acquisition module collects real-time data on the quenching heating temperature, the temperature after quenching, and the rate of temperature change within the same processing cycle of the brake discs of the same batch of vehicles. Secondly, the temperature of the brake disc quenching and heating of different vehicles at the next moment is predicted by the improved Gaussian process regression prediction algorithm based on dynamic learning in the prediction module, and the data information of the temperature of the brake disc quenching and heating of different vehicles at the next moment is obtained. Then, the temperature of vehicle brake disc quenching heating is optimized by the covariance matrix adaptive evolution algorithm for global search in the optimization module; Finally, the vehicle brake disc quenching heating temperature control function in the temperature control and regulation module, including the first temperature control function F, is used. control1 Second temperature control function F control2 The first temperature control function F control1 The output parameters are used to control the heating power of the first induction coil, and the second temperature control function F control2 The output parameters are used to control the heating power of the second induction coil.

[0033] In this embodiment, as Figure 2 As shown, the improved Gaussian process regression prediction algorithm based on dynamic learning predicts the quenching and heating temperature of brake discs for different vehicles at the next moment, including: Q1. Based on the data information of the temperature of the brake disc quenching heating of the same batch of vehicles and the data information of the temperature change rate within the same period, normalization processing is performed to obtain the normalized data information of the temperature of the brake disc quenching heating of the same batch of vehicles and the data information of the temperature change rate within the same period. Q2. Based on the normalized data of the temperature during the quenching and heating of brake discs in the same batch of vehicles and the data of the rate of temperature change within the same period, construct a covariance function W. , Where x represents the normalized temperature data of the brake disc quenching and heating of the same batch of vehicles, and Δx represents the normalized temperature change rate data of the same batch of vehicle brake disc quenching and heating within the same period. Q3. Construct the Gaussian process function H. , Where α and β are any constant parameters between 0 and 1, the temperature of the brake disc quenching heating of different vehicles at the next moment is predicted, and the data information of the temperature of the brake disc quenching heating of different vehicles at the next moment is obtained.

[0034] In this embodiment, the constraint condition for the constant parameters α and β is (α + β). 2 +ɑ 2 +β 2 ≤1.

[0035] In this embodiment, as Figure 3 As shown, the optimization of the quenching heating temperature of vehicle brake discs using the covariance matrix adaptive evolution algorithm oriented towards global search includes: M1. Based on the data information of the quenching and heating temperature of different vehicle brake discs at the next moment, the population is initialized, and the initial step size standard deviation δ, population size n, initial mean m, and weight {ω} are determined. i}, thus obtaining the data information of the initialized population; M2. Based on the data information of the initialized population, according to y (t+1) =m+δ t C (t) ∙N(0,1), where y (t+1) Let C be the (t+1)th individual in the population. (t) Let N(0,1) be the covariance matrix and N(0,1) be a standard normal distribution. We then sample individuals from the population to obtain the sampled individuals. M3. Based on the sampled individuals in the population, construct the covariance matrix update function C. (t+1) , , Where, ρ l and ρ c The temperature of vehicle brake disc quenching heating is optimized using the preset covariance update coefficient, resulting in optimized data information on the temperature of vehicle brake disc quenching heating.

[0036] In this embodiment, the preset covariance update coefficient ρ l and ρ c The weight {ω} is a random number between 0 and 1. i The constraint condition for} is ω1+ω2+...+ω n =1, where n is a positive integer, ρ l and ρ c The value is calculated according to the standard CMA-ES formula based on the optimization variable dimension r=8 and the population size n=20.

[0037] In this embodiment, the first temperature control function F control1configured to control the heating temperature of the central area of the brake disc, said second temperature control function F control2 is configured to control the heating temperature of the edge area of the brake disc, and the two temperature control functions independently adjust the power output of the corresponding induction coils according to the feedback from temperature sensors at different positions of the brake disc; the power adjustment range of the first induction coil is 50-100kW, which is used in the preheating stage of the brake disc; the power adjustment range of the second induction coil is 100-200kW, which is used in the quenching heating stage of the brake disc; the power ratio of the two induction coils is dynamically adjusted according to the real-time temperature distribution of the brake disc, and the adjustment formula is: P1:P2=(T center -T edge ):(T edge +k), wherein, P1 is the power of the first induction coil, P2 is the power of the second induction coil, T center is the temperature of the central area of the brake disc, T edge is the temperature of the edge area of the brake disc, and K is a zero-division prevention compensation constant, with a value range of 50-100°C.

[0038] In this embodiment, the quenching heating temperature control function for the vehicle brake disc comprises a first temperature control function F control1 and a second temperature control function F control2 , wherein, the first temperature control function F control1 : when a1≤g1≤b1: F control1 =g1, when g1<a1: F control1 =g1∙(1+Δg1) 2 , when g1>b1: F control1 =g1∙(1-Δg1) 3 , the second temperature control function F control2 : when a2≤g2≤b2: F control2 =g2, when g2<a2: F control2 =g2∙(1+Δg2) 2 , when g2>b2: F control2 =g2∙(1-Δg2) 3 , Wherein, a1 is the lower temperature threshold controlled by the first induction coil, b1 is the upper temperature threshold controlled by the first induction coil, g1 is the optimized temperature data information corresponding to the first induction coil, Δg1 is the temperature change rate corresponding to the first induction coil, a2 is the lower temperature threshold controlled by the second induction coil, b2 is the upper temperature threshold controlled by the second induction coil, g2 is the optimized temperature data information corresponding to the second induction coil, and Δg2 is the temperature change rate corresponding to the second induction coil.

[0039] In this embodiment: when g1 is between a1 and b1 and g2 is between a2 and b2, the optimized vehicle brake disc quenching heating temperature data is output; when g1 is less than a1, according to F... control1 =g1∙(1+Δg1) 2 The temperature increase is controlled until g1 is between a1 and b1; when g1 is greater than b1, it is controlled according to F. control1 =g1∙(1-Δg1) 3 The temperature is controlled to decrease until g1 is between a1 and b1; when g2 is less than a2, according to F... control2 =g2∙(1+Δg2) 2 The temperature increase is controlled until g2 is between a2 and b2; when g2 is greater than b2, it is controlled according to F. control2 =g2∙(1-Δg2) 3 The temperature is controlled to decrease until g2 is between a2 and b2.

[0040] In this embodiment, the temperature control and regulation module outputs two independent control signals: First control signal: - Controlled object: First induction coil; - Temperature control range: 750℃-850℃ (preheating zone); -Power adjustment range: 50-100kW; - Corresponding temperature measurement points: 4 temperature measurement points in the center area of ​​the brake disc.

[0041] Second control signal: - Controlled object: Second induction coil; - Temperature control range: 850℃-950℃ (quenching heating zone); -Power adjustment range: 100-200kW; - Corresponding temperature measurement points: 4 temperature measurement points in the edge area of ​​the brake disc.

[0042] Dual-coil cooperative control strategy: Step 1: Read the real-time temperature data of 8 temperature measurement points, of which 4 temperature measurement points are located in the center area of ​​the brake disc (radius 0-80mm) and 4 temperature measurement points are located in the edge area of ​​the brake disc (radius 80-160mm). Step 2: Calculate the average temperature T in the central area center and the average temperature T in the edge region edge ; Step 3: Calculate the temperature deviation ΔT = T center -T edge ; Step 4: Adjust the power ratio of the dual coils according to the temperature deviation: -When ΔT>50℃, increase the power ratio of the second induction coil, P1:P2=3:7; -When -50℃≤ΔT≤50℃, maintain the balanced power ratio, P1:P2=5:5; -When ΔT<-50℃, increase the power ratio of the first induction coil, P1:P2=7:3; Step 5: Output two independent control signals to the intermediate frequency power supply of the first and second induction coils.

[0043] To supplement the experimental data comparison, Table 1 provides comparative experimental data between dual-coil control and single-coil control:

[0044] By outputting two independent control signals through the temperature control and adjustment module, the power output of the first and second induction coils is controlled respectively, achieving differentiated heating of the center and edge areas of the brake disc. This effectively solves the problem of uneven temperature caused by traditional single-coil heating. The center-edge temperature difference is reduced from 16.8±5.2℃ to 6.3±1.7℃, and the hardness uniformity is improved from ±3HRC to ±2HRC. Combined with the dual-coil power ratio dynamic adjustment strategy, the power output of each coil is automatically optimized according to the real-time temperature distribution of the brake disc. Under the premise of ensuring heat treatment quality, the heating time is shortened by 16.7%, energy consumption is reduced by 20%, and the micro-crack scrap rate is reduced from 0.87% to 0.12%.

[0045] In this embodiment, the intelligent control system hardware architecture, wherein the intelligent control system (5) adopts an embedded industrial control computer as the hardware platform, includes: Central Processing Unit: ARM Cortex-A72 quad-core processor, clock speed 1.5GHz; Memory: 4GB DDR4; Storage: 64GB eMMC flash memory; Input interfaces: 8 analog input channels (temperature sensor signals) and 4 digital input channels.

[0046] Output interfaces: 4-channel PWM output (induction coil power control), 2-channel relay output (motor control); Communication interfaces: RS485×2, Ethernet×1, CAN×1.

[0047] Heat treatment process parameter acquisition module, Functional Boundaries: Responsible for collecting and preprocessing temperature-related data during the heat treatment process. Input data format: Temperature sensor raw signal: 8-channel analog voltage signal, range 0-10V, sampling frequency 100Hz; timestamp: 64-bit Unix timestamp, accuracy 1ms.

[0048] Output data format: -Quenching heating temperature: Floating point type, unit ℃, accuracy 0.1℃; -Quenching temperature: Floating point type, unit ℃, accuracy 0.1℃; - Temperature change rate: floating point, unit ℃ / s, accuracy 0.01℃ / s.

[0049] Key parameters: -Temperature sensor type: Type K thermocouple; - Sampling period: 10ms; - Filtering algorithm: moving average filter, window size 10.

[0050] Optimization module, functional boundaries: The CMA-ES algorithm is used to optimize the quenching heating temperature parameters. Input data format: -Predicted temperature value: Floating-point array, 8×1 dimension; - Quality parameters: Floating-point array containing hardness value, metallographic structure grade, and surface residual stress.

[0051] Output data format: - Optimized temperature setpoint: floating-point array, dimension 8×1; -Optimize convergence flag: Boolean type.

[0052] Key parameters: - Population size n=20 - Optimize the variable dimension r=8 (corresponding to 8 temperature control points) - Maximum number of iterations: 100 - Convergence tolerance: 0.01 - Covariance update coefficient: ρ l =4 / (r+4)², ρc=2 / ((r+2)∙n).

[0053] Prediction module, functional boundary: Predicts the temperature at the next time step using an improved Gaussian process regression. Input data format: -Historical temperature sequence: Floating-point array, dimension N×1, where N is the number of historical data points; -Historical temperature change rate: floating-point array, dimension N×1.

[0054] Output data format: -Predicted temperature value: Floating point type, unit: °C; - Prediction confidence interval: floating-point array, dimension 2×1 (upper and lower bounds).

[0055] Key parameters: -α=0.6, β=0.5 (the constraints are met) - Kernel function bandwidth σ1=0.8, σ2=0.6 -Noise variance σ 2 =0.01 -Historical data points N=50.

[0056] Temperature control and regulation module, functional boundary: Generates induction coil power control signal based on optimization and prediction results. Input data format: - Optimized temperature setpoint: floating-point array, dimension 8×1; - Predicted temperature value: floating point type.

[0057] - Real-time temperature feedback: Floating-point array, 8×1 dimension Output data format: - First induction coil PWM duty cycle: floating point type, range 0-100%; - Second induction coil PWM duty cycle: floating point type, range 0-100%.

[0058] Key parameters: - Power range of the first induction coil: 50-100kW; - Power range of the second induction coil: 100-200kW; - Temperature control range for the central area: 750-850℃; - Temperature control range for edge areas: 800-900℃; -PWM control frequency: 20kHz.

[0059] Detailed explanation of the connection relationships of mechanical components: Lifting mechanism: adopts servo electric cylinder, stroke 300mm, maximum load 50kg, positioning accuracy ±0.05mm; First induction coil: wound with copper tube, inner diameter 200mm, outer diameter 400mm, 15 turns, water-cooled; Second induction coil: wound with copper tube, inner diameter 400mm, outer diameter 600mm, 20 turns, water cooling method; Rotary geared motor: servo motor + planetary reducer, power 3kW, reduction ratio 50:1, output speed 0-30rpm; Workpiece clamping plate: made of aluminum alloy, 500mm in diameter, three-jaw self-centering clamp, clamping range 150-400mm; Sprayer swing arm: made of stainless steel, 600mm in length, and can rotate from 0 to 360°. Spray plate: 150mm in diameter, 12 nozzles, adjustable spray angle; Swing arm rotary motor: Stepper motor, power 200W, step angle 1.8°; Water pump: centrifugal pump, flow rate 50L / min, head 30m, power 2.2kW.

[0060] In this embodiment, the intelligent control system further includes a voice broadcast module connected to the temperature control and adjustment module, used to broadcast real-time data information on the temperature of the vehicle brake disc quenching heating process.

[0061] In this embodiment, the intelligent control system further includes a human-machine interaction module 6, which is connected to the temperature control and adjustment module. It is used to input the preset lower limit critical value and upper limit critical value of the vehicle brake disc quenching heating temperature, and to display the data in real time.

[0062] In this embodiment, the intelligent control system is connected to the swing arm rotary motor and the water pump, respectively.

[0063] In this embodiment, the first induction coil 2 and the second induction coil 14 are coaxially arranged and an insulating layer is provided between them.

[0064] In this embodiment, the heat treatment process parameter acquisition module: Functional Boundaries: Responsible for collecting and preprocessing temperature data during the brake disc quenching and heating process. Input data format: {sensor_id:string,temperature:float,timestamp:long,rate_of_change:float}; Output data format: {batch_id:string,=normalized_temperature:float[],normalized_rate:float[]}; Algorithm flow: Step 1: Read temperature sensor data; Step 2: Perform data validity verification; Step 3: Perform normalization: x norm =(xx min ) / (x max -x min ); Step 4: Output the normalized data; Key parameters: Sampling frequency 100Hz, temperature range 0-1200℃; Prediction module: Functional boundary: Predicting the temperature at the next moment based on historical data; Input data format: {normalized_temperature:float[], normalized_rate:float[], alpha:float, beta:float}; Output data format: {predicted_temperature:float, confidence:float}; Algorithm flow: Step 1: Load historical temperature data; Step 2: Calculate the covariance function W(x,Δx); Step 3: Construct the Gaussian process function H(t+1); Step 4: Output the predicted temperature value and confidence level; Key parameters: α∈[0.3,0.7], β∈[0.3,0.7], α 2 +β 2 +2ɑβ≤1; Optimization module: Functional boundary: Optimize calculations for predicted temperature; Input data format: {predicted_temperature:float[],population_size:int,dimension:int}; Output data format: {optimized_temperature:float, iteration_count:int}; Algorithm flow: Step 1: Initialize population parameters δ=0.5, n=20, m=mean value of initial temperature; Step 2: Calculate weight ω i =(ln(n+1)-ln(i)) / ∑(ln(n+1)-ln(j)); Step 3: Perform population sampling y (t+1) =m+δ t ·C (t) ·N(0,1); Step 4: Update covariance matrix C (t+1) ; Step 5: Output the optimized temperature value; Key parameters: ρ l =0.05, ρ c =0.05, r=8, n=20; Temperature control and regulation module: Functional boundary: implement temperature control strategy; Input data format: {optimized_temperature:float,lower_bound:float,upper_bound:float,sensitivity_eta:float, sensitivity_lambda:float}; Output data format: {control_signal:float, target_temperature:float}; Algorithm flow: Step 1: Read the optimized temperature g and boundary values a, b; Step 2: Determine whether g is within the interval [a,b]; Step 3: If g<a, execute F control =g·(1+Δg) 2 ; Step 4: If g>b, execute F control =g·(1-Δg) 3 ; Step 5: Output the control signal; Key parameters: a=750℃, b=950℃,Δg∈[0.01,0.1].

[0065] In this embodiment, the frame structure is made of stainless steel and has external dimensions of 2000mm (length) × 1500mm (width) × 2500mm (height). The interior of the frame is divided into three independent spaces: a heating zone, a quenching zone, and a control zone. A 150mm diameter mounting hole is provided at the center of the top of the frame for mounting the lifting mechanism. A drain outlet is provided at the bottom of the frame, which is connected to the coolant recovery tank.

[0066] In this embodiment, the lifting mechanism includes a linear lifting motor with an integrated guide column. The lower end of the lifting mechanism's output shaft is connected to an induction coil bracket, and the induction coil is fixed to the bracket via an insulating layer.

[0067] In this embodiment, the induction coil is made of copper tubing, with an outer diameter of 80mm, an inner diameter of 60mm, and 15 turns. The induction coil is connected to a medium-frequency power supply, operating at a frequency of 10-50kHz and with an adjustable power of 50-200kW. The induction coil is electrically connected to an intelligent control system, receiving control signals to adjust the heating power.

[0068] In this embodiment, a servo motor with a power of 1.5kW and a rated speed of 1500r / min is used. The turntable has a diameter of 500mm and is made of high-temperature resistant alloy steel. The workpiece clamping disc is located at the center of the turntable and adopts a three-jaw chuck structure, capable of clamping brake discs with a diameter of 200-400mm. The rotation speed of the rotating mechanism is adjustable from 0-300r / min to ensure uniform heating of the brake disc.

[0069] In this embodiment, the spray system includes a spray arm, a spray disc, a rotary motor for the arm, a hose, and a water pump. The spray arm is made of stainless steel tubing, 800mm in length, and can rotate 0-180 degrees around the output shaft of the rotary motor. The spray disc has a diameter of 150mm and multiple spray holes with a diameter of 2mm and a spacing of 10mm. The rotary motor for the arm is a stepper motor with an accuracy of 0.9 degrees / step. The water pump is a centrifugal pump with a flow rate of 50-100L / min and a head of 20-30m. The hose is made of high-temperature resistant rubber tubing with an inner diameter of 25mm.

[0070] In this embodiment, the coolant recovery tank has a volume of 500L and is made of stainless steel. A filter with a 50-mesh filtration accuracy is installed inside the recovery tank to filter impurities in the quenching fluid. A level sensor is installed on the side of the recovery tank to monitor the coolant level in real time.

[0071] In this embodiment, the intelligent control system is implemented in hardware as follows: 1. Hardware Architecture The intelligent control system uses an industrial-grade PLC, specifically a Siemens S7-1500, as the main controller. The system includes a CPU module, digital input / output modules, analog input / output modules, a communication module, and a human-machine interface. It also includes modules for acquiring heat treatment process parameters, prediction, optimization, and temperature control and regulation.

[0072] 2. Sensor Configuration The temperature sensor uses an infrared thermometer with a measurement range of 0-1500℃, an accuracy of ±1℃, and a response time of 50ms. Four temperature measurement points are set to monitor the temperature at different positions on the brake disc. The displacement sensor uses a magnetostrictive sensor to measure the position of the lifting mechanism with an accuracy of 0.1mm. The speed sensor uses a photoelectric encoder to monitor the rotation speed of the rotating mechanism with an accuracy of 0.1r / min. The pressure sensor monitors the water pump outlet pressure with a range of 0-1MPa and an accuracy of 0.5%. There are eight temperature sensor mounting positions on the workpiece clamping plate, four of which are located in the central area with a radius of 0-80mm, and four are located in the edge area with a radius of 80-160mm.

[0073] 3. Actuator Configuration The lifting motor driver uses a servo driver and supports three control modes: position, speed, and torque. The rotary motor driver uses a servo driver and supports electronic gearing. The swing arm rotary motor driver uses a stepper driver and supports microstepping control. The water pump is controlled by a frequency converter with a frequency range of 0-50Hz. The induction coil power supply uses a medium-frequency power supply with a power adjustment range of 0-100%.

[0074] 4. Communication Interface The system supports multiple communication protocols including Profinet, Modbus TCP, and RS485. Communication with the host computer uses an Ethernet interface at a speed of 100Mbps. Communication with sensors uses an RS485 interface at a baud rate of 9600bps. Communication with the human-machine interface uses a Profinet interface.

[0075] In this embodiment, the heat treatment process parameter acquisition module, Functional Boundaries: Responsible for collecting and preprocessing temperature data during the brake disc quenching and heating process. Input data format: {sensor_id:string,temperature:float,timestamp:long,rate_of_change:float}; Output data format: {batch_id:string,normalized_temperature:float[],normalized_rate:float[]}; Algorithm flow: Step 1: Read data from 4 temperature sensors at a sampling frequency of 100Hz; Step 2: Perform data validity verification and remove outliers (temperature >1500℃ or <0℃); Step 3: Calculate the rate of temperature change ΔT = (T t -T t-1 ) / Δt; Step 4: Perform normalization: x norm =(xx min ) / (x max -x min ); Step 5: Output the normalized data to the prediction module; Key parameters: Sampling frequency: 100Hz Temperature range: 0-1200℃ Normalization range: 0-1 Data cache length: 1000 points; Temperature prediction module Functional boundary: Predicting the temperature at the next moment based on historical data; Input data format: {normalized_temperature:float[],normalized_rate:float[],alpha:float,beta: float}; Output data format: {predicted_temperature: float,confidence: float}; Algorithm flow: Step 1: Load historical temperature data, window length 100; Step 2: Calculate the covariance function W(x,Δx)=ɑ·exp(-‖x‖ 2 / 2)+β·exp(-‖Δx‖ 2 / 2); Step 3: Construct the Gaussian process function H(t+1)=W(x,Δx)·(ɑ·H(t)+β·H(t-1)); Step 4: Calculate the predicted value and confidence interval; Step 5: Output the predicted temperature value and confidence level to the optimization module; Key parameters: ɑ∈[0.3,0.7], β∈[0.3,0.7], α2 +β 2 +2ɑβ≤1, Prediction step size: 10ms; Optimization module Functional boundary: Optimize calculations for predicted temperature; Input data format: {predicted_temperature:float[],population_size:int,dimension: int}; Output data format: {optimized_temperature: float, iteration_count: int}; Algorithm flow: Step 1: Initialize population parameters δ=0.5, n=20, m=initial mean temperature; Step 2: Calculate the weights ω i =ln(n+1)-ln(i) / ∑(ln(n+1)-ln(j)); Step 3: Perform population sampling y (t+1) =m+δ t ·C (t) ·N(0,1); Step 4: Update the covariance matrix C (t+1) , ; Step 5: Determine the convergence condition; if it is met, output the optimization result. Step 6: Output the optimized temperature value to the control module; Key parameters: ρl=0.05, ρc=0.05, r=8, n=20, maximum number of iterations: 500.

[0076] Temperature control and regulation module Functional boundary: Execute temperature control strategy Input data format: {optimized_temperature: float, lower_bound: float, upper_bound: float, sensitivity_eta: float, sensitivity_lambda: float} Output data format: {control_signal: float,target_temperature: float} Algorithm flow: Step 1: Read the optimization temperature g and boundary values a and b; Step 2: Determine whether g is within the interval [a,b]; Step 3: If a≤g≤b, execute F control =g; Step 4: If g<a, execute F control =g·(1+Δg) 2 ; Step 5: If g>b, execute F control =g·(1-Δg) 3 ; Step 6: Output a control signal to the actuator; Key parameters: a=750°C, b=950°C, Δg∈[0.01,0.1], η=0.5, λ=0.5.

[0077] In this embodiment, in the feeding stage, an operator places the brake disc on the workpiece holding disc and clamps it by a three-jaw chuck. Parameters such as brake disc model, material, target temperature are input through the human-machine interface. The intelligent control system automatically calls up the corresponding process parameters.

[0078] In the heating stage, the lifting mechanism descends, the induction coil approaches the brake disc, and the distance is controlled at 50-100mm. The induction coil is energized, and the medium-frequency power supply outputs the set power. The rotating mechanism starts, and the brake disc rotates at a speed of 50-150r / min. The temperature sensor monitors the temperature of the brake disc in real time, and the parameter acquisition module collects temperature data. The prediction module predicts the temperature at the next moment according to historical data. The optimization module performs optimization calculation on the predicted temperature. The control module adjusts the power of the medium-frequency power supply according to the optimization result.

[0079] In the quenching stage, when the temperature of the brake disc reaches the set value (850±50°C), the lifting mechanism ascends, and the induction coil moves away from the brake disc. The spray swing arm immediately rotates to above the brake disc, the water pump starts, and the quenching liquid is uniformly sprayed onto the surface of the brake disc through the spray disc. The rotating mechanism continues to rotate to ensure uniform quenching. The cooling liquid flows into the recovery tank and is recycled after filtration.

[0080] In the blanking stage, after quenching is completed, the rotating mechanism stops and the spray swing arm resets. The operator removes the brake disc for subsequent processing. The system automatically records the heat treatment process parameters of this operation for subsequent optimization.

[0081] For temperature control accuracy, an improved Gaussian process regression prediction algorithm is adopted, and the temperature prediction error is controlled within ±5°C. Compared with the traditional PID control, the temperature fluctuation amplitude is reduced by 60%.

[0082] The heat treatment quality is excellent; the surface hardness of the brake disc reaches HRC45-50, with a hardness uniformity of ±3HRC. The metallographic structure is uniform tempered sorbite, without network carbides.

[0083] Production efficiency has been improved by reducing the heat treatment cycle of a single brake disc from 45 minutes in the traditional process to 30 minutes, resulting in a 33% increase in efficiency.

[0084] Energy consumption is reduced by 20% through the use of optimized control algorithms and dynamic adjustment of intermediate frequency power supply.

[0085] The equipment's lifespan is extended by 50% due to the water-cooled induction coil, which keeps the operating temperature below 60℃.

[0086] Ease of operation is enhanced by a graphical user interface, reducing operator training time from 7 days to 2 days. A voice broadcast module provides real-time process status updates, minimizing operational errors.

[0087] In this embodiment, a certain model of automotive brake disc is used as an example. The material is gray cast iron HT250, with a diameter of 320mm and a thickness of 30mm.

[0088] Process parameters are set as follows: target temperature: 850℃, holding time: 15 minutes, quenching temperature: 850±50℃, quenching time: 8 minutes.

[0089] Control parameters are set as follows: α = 0.5, β = 0.4, ρ l =0.05, ρ c =0.05, population size n=20; Optimize the variable dimension r=8.

[0090] Implementation Process: After loading, the system automatically identifies the brake disc model and retrieves the process parameters. The lifting mechanism descends to the 80mm position, the induction coil is energized, and the power is set to 120kW. The rotating mechanism starts, with a speed of 100r / min. The temperature sensor monitors in real time, and the brake disc temperature reaches 850℃ after 10 minutes. The lifting mechanism rises, the spray system starts, and quenching lasts for 8 minutes. The temperature fluctuation throughout the entire process is controlled within ±5℃.

[0091] The brake disc surface hardness is HRC47, with a hardness uniformity of ±2HRC. The metallographic structure is uniform and defect-free. The processing time per piece is 28 minutes, and energy consumption is reduced by 22% compared to traditional processes.

[0092] In this embodiment, a batch heat treatment of a certain type of gray cast iron brake disc (HT250, diameter 320mm, thickness 24mm) is taken as an example. Production line configuration: The system of this application has built-in temperature sensors (including 8 points for workpiece surface temperature measurement and 4 points for internal frame temperature measurement), with a sampling period of 2 seconds. The PLC controller (Siemens S7-1515F) runs the improved Gaussian process regression prediction algorithm based on dynamic learning and the covariance matrix adaptive evolution algorithm oriented towards global search of this invention.

[0093] Step 1: Data Acquisition and Preprocessing. Complete heat treatment data of 288 brake discs from 3 consecutive furnaces (96 pieces per furnace, stacked in 4 layers) were collected. After removing abnormal jump points, a training dataset was constructed. Step 2: DL-GPR model training. Initialize hyperparameters and optimize marginal likelihood using the conjugate gradient method. Training time was 187 seconds. Validation results: MAE=2.1℃, RMSE=2.9℃, 95% prediction interval coverage (PICP) reached 96.3%, significantly better than LSTM (PICP=82.1%) and SVR (PICP=76.5%). Step 3: C-CMA-ES parameter tuning. Population size μ=12, resampling number λ=24, initial step size σ=5.0℃. The average time for a single optimization iteration is 38ms (meeting the 200ms control cycle requirement), and the convergence generation is stable at 22±3 generations.

[0094] Step 4: Closed-loop control operation. A comparative test was conducted on the 4th batch, with a control group (traditional PID) and an experimental group (this invention). Temperature curves of 12 brake discs throughout their entire lifespan were collected (sampling rate 1Hz), and the key range of 860-920℃ was analyzed.

[0095] The standard deviation of temperature tracking error in the experimental group was 1.42℃, which was 61.4% lower than that in the control group (3.68℃). The heating rate of 95% of the workpieces was controlled within the range of 6.2-7.8℃ / min, which met the process window requirements. Metallographic analysis showed that the average pearlite content in the experimental group was 89.7±1.2%, while that in the control group was 85.3±3.8%, indicating a significant improvement in microstructure uniformity.

[0096] In this embodiment, brake disc manufacturer A deployed this system in the No. 2 mesh belt furnace in March 2025. For models exported to the EU (requiring ECE R90 certification), the brake disc friction surface hardness is required to be 65±3HRC with no soft spots. Before implementation, 1000 samples were randomly inspected, with a soft spot rate of 2.3%; after implementation, 5000 samples were continuously monitored, and the soft spot rate decreased to 0.08%, and the system passed the accelerated thermal fatigue test (10...) conducted by the third-party testing agency SGS. 5(After one braking cycle), the surface crack propagation rate decreased by 37.2%. Customer feedback regarding brake noise complaints decreased by 76%, confirming the substantial improvement in microstructure uniformity achieved by this method.

[0097] In this embodiment, the present invention has been deployed on the brake disc heat treatment production line of an automotive parts factory. Continuous operation data from July to December 2025 shows: 1. The first-pass yield rate of the process increased from 92.4% to 99.1%, and rework costs decreased by 67%; 2. Energy consumption per furnace was reduced by 4.3%, due to the optimization algorithm reducing ineffective heating caused by overshoot; 3. Overall equipment efficiency (OEE) increased from 78.5% to 86.2%, mainly due to the fault warning module (based on DL-GPR residual mutation detection) which reduced the average maintenance response time to 23 minutes.

[0098] This invention achieves a triple breakthrough in the field of brake disc heat treatment by constructing a full-chain control paradigm of "dynamic perception - probabilistic prediction - constraint optimization - adaptive execution": First, it innovates the modeling method for thermal processes at the theoretical level. The time-varying hybrid nucleus and entropy characteristics proposed by DL-GPR are the first to incorporate prior knowledge of thermodynamic phase transition kinetics (such as the abrupt change in heat capacity near the Ac1 point) into the covariance structure of Gaussian processes, enabling the model to not only fit the data but also understand the physical mechanism. This scheme reduces computational cost by 99.8%, while key indicators (such as the prediction error of the phase transition point temperature) are reduced by 22%.

[0099] Second, at the engineering level, the trade-off between real-time performance and robustness has been overcome. The hard / soft dual constraint mechanism embedded in C-CMA-ES ensures that the optimized solution always remains within the device's safety domain, avoiding the "missing feasible solution" problem common in traditional black-box optimization; F control The segmented strategy balances rapid response and system stability, with a measured maximum overshoot of only 1.8℃, far below the industry standard limit of 5℃.

[0100] Third, the application level has verified the value of large-scale implementation. Without increasing hardware investment, the algorithm upgrade alone has achieved a process consistency level close to that of high-end vacuum heat treatment furnaces (HV dispersion ≤ ±12), providing a reusable technical path for the intelligent upgrading of low- and mid-range heat treatment equipment.

[0101] In summary, this invention not only ensures that the brake disc achieves optimal performance and quality during the manufacturing process, but also improves the safety and reliability of the braking system.

[0102] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A heat treatment system for vehicle brake discs, comprising a frame (13), characterized in that: A lifting mechanism (1) is provided at the center of the top of the frame (13). The output shaft of the lifting mechanism (1) is connected to a first induction coil (2) and a second induction coil (14) through an insulating bracket. The first induction coil (2) and the second induction coil (14) are coaxially arranged. The first induction coil (2) is located directly below the second induction coil (14). A rotary reduction motor (4) is provided at the center of the bottom of the frame (13). The output shaft of the rotary reduction motor (4) is provided with a workpiece clamping plate (11). The workpiece clamping plate (11) is located at the center of the first induction coil (14). Directly below the coil (2), a swing arm rotary motor (8) is provided on the side of the frame (13). The output shaft of the swing arm rotary motor (8) is connected to one end of the spray swing arm (7). The other end of the spray swing arm (7) is provided with a spray plate (9). The spray plate (9) is connected to the water pump (10) through a hose. The water pump (10) is located at the bottom of the frame (13). The first induction coil (2), the second induction coil (14), the rotary reduction motor (4), the swing arm rotary motor (8) and the water pump (10) are electrically connected to the intelligent control system (5).

2. The heat treatment system for vehicle brake discs according to claim 1, characterized in that: The intelligent control system (5) includes a heat treatment process parameter acquisition module, a prediction module, an optimization module, and a temperature control and adjustment module. The heat treatment process parameter acquisition module is connected to a temperature sensor signal set on the surface of the brake disc. The output of the temperature control and adjustment module is connected to the intermediate frequency power supply of the first induction coil (2) and the second induction coil (14) through a power controller.

3. The heat treatment system for vehicle brake discs according to claim 1, characterized in that: The bottom of the frame (13) is provided with a coolant recovery tank (12), and the inlet of the coolant recovery tank (12) is connected to the drain end of the spray plate (9).

4. The heat treatment system for vehicle brake discs according to claim 1, characterized in that: The workpiece clamping plate (11) is provided with 8 temperature sensor mounting positions, of which 4 are located in the central area with a radius of 0-80mm and 4 are located in the edge area with a radius of 80-160mm.

5. A method for optimizing and controlling the heat treatment process of vehicle brake discs, characterized in that, The heat treatment system for vehicle brake discs as described in any one of claims 1-4 First, the heat treatment process parameter acquisition module collects real-time data on the quenching heating temperature, the temperature after quenching, and the rate of temperature change within the same processing cycle of the brake discs of the same batch of vehicles. Secondly, the temperature of the brake disc quenching and heating of different vehicles at the next moment is predicted by the improved Gaussian process regression prediction algorithm based on dynamic learning in the prediction module, and the data information of the temperature of the brake disc quenching and heating of different vehicles at the next moment is obtained. Then, the temperature of vehicle brake disc quenching heating is optimized by the covariance matrix adaptive evolution algorithm for global search in the optimization module; Finally, the vehicle brake disc quenching heating temperature control function in the temperature control and regulation module, including the first temperature control function F, is used. control1 Second temperature control function F control2 The first temperature control function F control1 The output parameters are used to control the heating power of the first induction coil, and the second temperature control function F control2 The output parameters are used to control the heating power of the second induction coil.

6. The method for optimizing and controlling the heat treatment process of vehicle brake discs according to claim 5, characterized in that, The improved Gaussian process regression prediction algorithm based on dynamic learning predicts the quenching and heating temperature of brake discs for different vehicles at the next moment, including: Q1. Based on the data information of the temperature of the brake disc quenching heating of the same batch of vehicles and the data information of the temperature change rate within the same period, normalization processing is performed to obtain the normalized data information of the temperature of the brake disc quenching heating of the same batch of vehicles and the data information of the temperature change rate within the same period. Q2. Based on the normalized data of the temperature of the brake discs of the same batch of vehicles after quenching and heating and the data of the rate of temperature change within the same period, construct a covariance function W(x,Δx), where x is the normalized data of the temperature of the brake discs of the same batch of vehicles after quenching and heating, and Δx is the normalized data of the rate of temperature change within the same period of the brake discs of the same batch of vehicles after quenching and heating. Q3. Construct the Gaussian process function H. ， (a+b) 2 +ɑ 2 +b 2 ≤1, Where α and β are any constant parameters between 0 and 1, the temperature of the brake disc quenching heating of different vehicles at the next moment is predicted, and the data information of the temperature of the brake disc quenching heating of different vehicles at the next moment is obtained.

7. The method for optimizing and controlling the heat treatment process of vehicle brake discs according to claim 5, characterized in that, The adaptive evolutionary algorithm for covariance matrix based on global search optimizes the quenching heating temperature of vehicle brake discs, including: M1. Based on the data information of the quenching and heating temperature of different vehicle brake discs at the next moment, the population is initialized, and the initial step size standard deviation δ, population size n, initial mean m, and weight {ω} are determined. i }, thus obtaining the data information of the initialized population; M2. Based on the data information of the initialized population, according to y (t+1) =m+δ t C (t) ∙N(0,1), where y (t+1) Let C be the (t+1)th individual in the population. (t) Let N(0,1) be the covariance matrix and N(0,1) be a standard normal distribution. We then sample individuals from the population to obtain the sampled individuals. M3. Based on the sampled individuals in the population, construct the covariance matrix update function C. (t+1) The temperature of the vehicle brake disc quenching heating was optimized to obtain the optimized temperature data of the vehicle brake disc quenching heating.

8. The method for optimizing and controlling the heat treatment process of vehicle brake discs according to claim 5, characterized in that: The first temperature control function F control1 The second temperature control function F is used to control the heating temperature of the center area of ​​the brake disc. control2 Two temperature control functions are used to control the heating temperature of the brake disc edge area. These functions independently adjust the power output of the corresponding induction coils based on feedback from temperature sensors at different locations on the brake disc. The power adjustment range of the first induction coil is 50-100kW, used during the brake disc preheating stage. The power adjustment range of the second induction coil is 100-200kW, used during the brake disc quenching heating stage. The power ratio of the two induction coils is dynamically adjusted according to the real-time temperature distribution of the brake disc, using the following formula: P1:P2=(T center -T edge ):(T edge +k), Where P1 is the power of the first induction coil, P2 is the power of the second induction coil, and T center T represents the temperature of the center area of ​​the brake disc. edge The temperature is the edge area of ​​the brake disc, and K is the zero compensation constant, with a value range of 50-100℃.

9. The method for optimizing and controlling the heat treatment process of vehicle brake discs according to claim 8, characterized in that: The vehicle brake disc quenching heating temperature control function includes a first temperature control function F. control1 Second temperature control function F control2 ,in, First temperature control function F control1 : When a1≤g1≤b1: F control1 =g1, When g1 < a1: F control1 =g1∙(1+Δg1) 2 , When g1 > b1: F control1 =g1∙(1-Δg1) 3 , Second temperature control function F control2 : When a² ≤ g² ≤ b²: F control2 =g2, When g2 < a2: F control2 =g2∙(1+Δg2) 2 , When g2 > b2: F control2 =g2∙(1-Δg2) 3 , Wherein, a1 is the lower temperature threshold controlled by the first induction coil, b1 is the upper temperature threshold controlled by the first induction coil, g1 is the optimized temperature data information corresponding to the first induction coil, Δg1 is the temperature change rate corresponding to the first induction coil, a2 is the lower temperature threshold controlled by the second induction coil, b2 is the upper temperature threshold controlled by the second induction coil, g2 is the optimized temperature data information corresponding to the second induction coil, and Δg2 is the temperature change rate corresponding to the second induction coil.

10. The method for optimizing and controlling the heat treatment process of vehicle brake discs according to claim 8, characterized in that: When g1 is between a1 and b1 and g2 is between a2 and b2, the optimized vehicle brake disc quenching heating temperature data is output; when g1 is less than a1, according to F... control1 =g1∙(1+Δg1) 2 Temperature growth control until g1 Located between a1 and b1; when g1 is greater than b1, according to F control1 =g1∙(1-Δg1) 3 The temperature is controlled to decrease until g1 is between a1 and b1; when g2 is less than a2, according to F... control2 =g2∙(1+Δg2) 2 The temperature increase is controlled until g2 is between a2 and b2; when g2 is greater than b2, it is controlled according to F. control2 =g2∙(1-Δg2) 3 Temperature reduction control until g2 is located at a Between 2 and b2.

Images