Electronic cigarette self-adaptive precise temperature control system based on fuzzy PID (Proportion Integration Differentiation) control

Abstract

The invention relates to the technical field of electronic cigarette control, and particularly discloses an electronic cigarette self-adaptive precise temperature control system based on fuzzy PID control, which comprises a temperature acquisition module, a disturbance quantization module, a fuzzy reasoning module, a parameter self-adaptive PID control module and a power driving module. By quantizing the disturbance intensity and adjusting PID parameters in real time based on a fuzzy rule, the system can adaptively enhance the anti-interference capability and the steady-state precision. The system has quick response to sudden disturbance, control oscillation possibly caused by frequent parameter adjustment is avoided, optimal balance between response speed and control stability is achieved, and the system is particularly suitable for portable equipment such as electronic cigarettes with complex and changeable working conditions. According to the scheme, rapid, stable and overshoot-free accurate control over the atomization temperature is achieved, and the use experience and consistency of the electronic cigarette are improved.

Description

Technical Field

[0001] This invention belongs to the field of electronic cigarette control technology, specifically relating to an adaptive and precise temperature control system for electronic cigarettes based on fuzzy PID control. Background Technology

[0002] In the field of electronic atomization devices, precise temperature control is a core technology for ensuring atomization effects, improving user experience, and guaranteeing safety. The performance of the temperature control system directly determines the vaporization efficiency of the atomized liquid, the amount of smoke, and the level of harmful substance formation, and therefore has always been a key area of ​​technological research and development in this field.

[0003] As a typical portable electronic atomization device, electronic cigarettes require a temperature control system that can rapidly respond to and stably maintain the heating temperature of the atomization chamber within limited space and power consumption constraints. This system is typically based on a closed-loop control principle, using sensors to collect temperature feedback signals in real time. The controller then calculates and outputs corresponding power drive signals to counteract external disturbances and bring the actual temperature close to the preset target value.

[0004] Traditional proportional-integral-derivative (PI-DE) controllers are used as the core control algorithm. However, in actual e-cigarette use, user inhalation behavior is highly random and uncertain. The intensity, duration, and mode of inhalation constantly change, causing complex and frequent airflow and temperature disturbances within the atomization chamber. Traditional controllers, with their fixed parameters, exhibit significant control lag when faced with such time-varying, nonlinear, and strong disturbances, making it difficult to achieve rapid and overshoot-free dynamic adjustments. This directly leads to large fluctuations in atomization temperature during inhalation, affecting not only the stability and consistency of the vapor's flavor but also potentially generating undesirable substances due to sudden temperature spikes. Summary of the Invention

[0005] The purpose of this invention is to provide an adaptive and precise temperature control system for electronic cigarettes based on fuzzy PID control, so as to solve the technical problems of control lag, large temperature fluctuation and difficulty in achieving rapid and overshoot-free dynamic adjustment when the controller with fixed parameters in the prior art faces time-varying and nonlinear strong disturbances caused by the user's inhalation behavior.

[0006] This invention provides an adaptive precision temperature control system for electronic cigarettes based on fuzzy PID control. The system includes a temperature acquisition module, a disturbance quantization module, a fuzzy inference module, a parameter adaptive PID control module, and a power drive module. The temperature acquisition module acquires the actual temperature value of the atomizing chamber in real time. The disturbance quantization module calculates and outputs a quantized index characterizing the intensity of the current external disturbance based on the acquired actual temperature value and its variation characteristics. The fuzzy inference module receives the quantized index output by the disturbance quantization module and, according to a preset fuzzy rule base, calculates and outputs the proportional coefficient adjustment, integral coefficient adjustment, and derivative coefficient adjustment in real time.

[0007] The parameter-adaptive PID control module receives the actual temperature value output from the temperature acquisition module and the preset target temperature value, calculates the temperature deviation and its rate of change. Simultaneously, this module receives three coefficient adjustment values ​​output from the fuzzy inference module, dynamically adjusting its internal proportional, integral, and derivative coefficients. Based on the dynamically adjusted parameters, this module performs proportional-integral-derivative (PID) calculations to generate a real-time control output signal. The power drive module receives the control output signal generated by the parameter-adaptive PID control module and converts it into a corresponding pulse width modulation (PWM) signal to drive the heating element, thereby achieving closed-loop control of the atomizing chamber temperature.

[0008] Furthermore, the quantization process of the disturbance quantization module is as follows: the module internally sets a length of... A sliding time window for each sampling period. The value is a positive integer greater than or equal to 5. At each sampling moment, the disturbance quantization module obtains the latest actual temperature value from the temperature acquisition module and calculates the first-order difference between the current temperature value and the temperature value at the previous sampling moment, i.e., the instantaneous temperature change rate. Simultaneously, this module calculates the standard deviation of all temperature samples within the current sliding time window, serving as a statistical measure of temperature fluctuation within that window. The disturbance quantization module weights and fuses the absolute value of the calculated instantaneous temperature change rate with the temperature standard deviation to generate a comprehensive disturbance intensity quantification index. The value range of this quantification index is normalized and mapped to a continuous interval from 0 to 1, where 0 represents no disturbance or extremely weak disturbance, and 1 represents the state corresponding to the system's preset maximum disturbance intensity threshold.

[0009] Furthermore, the fuzzy rule base construction process of the fuzzy inference module is as follows: the perturbation quantification index is used as the only input variable, and the proportional coefficient adjustment, integral coefficient adjustment and differential coefficient adjustment are used as three independent output variables.

[0010] The universe of discourse for both input and output variables is defined as a continuous interval from 0 to 1. Three fuzzy subsets are defined for the intensity of the input variable perturbation, named weak, medium, and strong, respectively.

[0011] Three fuzzy subsets are defined for the adjustment amount of the output variable proportional coefficient, named micro-increase, medium-increase, and large-increase respectively.

[0012] Three fuzzy subsets are defined for the adjustment of the integral coefficients of the output variables, named micro-reduction, medium-reduction, and large-reduction respectively.

[0013] Three fuzzy subsets are defined for the adjustment of the differential coefficients of the output variables, named micro-increase, medium-increase, and large-increase respectively.

[0014] The fuzzy rule base contains three core rules, which are described as follows: If the disturbance intensity is weak, the proportional coefficient adjustment is slightly increased, the integral coefficient adjustment is slightly decreased, and the differential coefficient adjustment is slightly increased. If the disturbance intensity is medium, then the proportional coefficient adjustment is medium increase, the integral coefficient adjustment is medium decrease, and the differential coefficient adjustment is medium increase. Article 3 states that if the disturbance intensity is strong, the proportional coefficient adjustment will increase significantly, the integral coefficient adjustment will decrease significantly, and the differential coefficient adjustment will increase significantly. The fuzzy inference module uses the centroid method for defuzzification calculation, converting the fuzzy inference results into precise coefficient adjustment values ​​for output.

[0015] Furthermore, the parameter dynamic adjustment process of the parameter adaptive PID control module is as follows: The parameter adaptive PID control module has a preset proportional coefficient reference value. Integral coefficient benchmark value and differential coefficient benchmark value ; In each control cycle, the parameter adaptive PID control module receives three coefficient adjustments from the fuzzy inference module, denoted as follows: , and ; The parameter adaptive PID control module calculates the PID parameters used in the current control cycle in real time according to the following formula: Current scaling factor ; Current integral coefficient ; Current differential coefficients ; in, , , The preset gain coefficient is used to adjust the magnitude of the adaptive adjustment; after calculating the dynamic parameters, the parameter adaptive PID control module adjusts the current temperature deviation. The integral and derivative of the temperature deviation are used to calculate the control output according to the standard PID control algorithm. The calculation formula is: .

[0016] Furthermore, the power drive module includes a pulse width modulation (PWM) signal generator and a metal-oxide-semiconductor (MOSFET) drive circuit. The PWM signal generator receives the control quantity u output by the parameter adaptive PID control module. This control quantity, after amplitude limiting, is mapped to a duty cycle command, with the duty cycle command ranging from 0% to 100%. The PWM signal generator generates a PWM square wave signal with a corresponding duty cycle according to the duty cycle command. The MOSFET drive circuit receives this PWM square wave signal, amplifies its current, and then drives the heating resistance wire connected to the atomizing cavity. The on / off time of the heating resistance wire is directly controlled by the duty cycle of the PWM signal, thereby achieving linear adjustment of the average heating power.

[0017] Furthermore, the system also includes a preheating control submodule, which is integrated within the parameter adaptive PID control module. This submodule is activated when the system powers on or wakes from sleep mode. It first sets the target temperature to a preheating temperature value lower than the steady-state target value, typically set to 70% to 80% of the steady-state target value. Simultaneously, this submodule temporarily overrides the output of the fuzzy inference module, forcing the proportional, integral, and derivative coefficient adjustments to zero, even if the parameter adaptive PID control module uses its preset reference parameters. , and Control is then implemented. Once the actual temperature value reaches and stabilizes near the preheating temperature value for a preset period of 2 to 3 seconds, the preheating control submodule automatically exits, the system resumes switching the target temperature value to the steady-state target value, and resumes receiving the output of the fuzzy inference module, entering a complete adaptive precision temperature control mode.

[0018] Furthermore, the system operates within a nested dual-loop collaborative control architecture. The inner loop is a millisecond-level fast power regulation loop, consisting of a parameter-adaptive PID control module and a power drive module, responsible for rapidly suppressing high-frequency temperature fluctuations. The outer loop is a second-level parameter-adaptive loop, consisting of a disturbance quantization module and a fuzzy inference module, with an operating cycle 10 to 50 times that of the inner loop control cycle. The outer loop monitors disturbance pattern changes over longer time scales and periodically optimizes the parameters of the inner loop controller accordingly. The inner and outer loops collaborate through shared temperature acquisition data and parameter transmission interfaces to ensure the system has a rapid dynamic response capability when dealing with sudden strong disturbances and extremely high steady-state control accuracy during stable phases.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention constructs a real-time adaptive controller parameter mechanism by introducing a disturbance quantization module and a fuzzy inference module. The system can dynamically sense the change in disturbance intensity caused by the user's suction behavior and intelligently adjust the three control parameters (proportional, integral, and derivative) according to preset fuzzy rules. When facing strong disturbances, the system automatically increases the proportional and derivative actions to improve the response speed, while reducing the integral action to prevent integral saturation and overshoot. When the disturbance weakens, the system adjusts the parameters in the opposite direction to optimize steady-state accuracy. This online adaptive capability fundamentally overcomes the lag and mismatch problems of traditional fixed-parameter controllers when facing time-varying nonlinear disturbances, and realizes rapid, stable, and overshoot-free dynamic tracking of the target temperature.

[0020] 2. The dual-loop nested collaborative control architecture proposed in this invention features a finely divided control timing sequence. The inner loop focuses on millisecond-level rapid power adjustment, ensuring immediate suppression of high-frequency temperature disturbances; the outer loop operates at a slower cycle, focusing on analyzing disturbance characteristics on a second-level time scale and optimizing the inner loop controller parameters accordingly. This architecture design ensures both rapid system response to sudden disturbances and avoids control oscillations that may result from frequent parameter adjustments, achieving an optimal balance between response speed and control stability. It is particularly suitable for portable devices like electronic cigarettes, which operate under complex and variable conditions.

[0021] 3. This invention effectively solves the temperature shock problem during system cold start or hibernation wake-up by designing an independent preheating control submodule. This submodule uses a lower preheating temperature and fixed reference control parameters to enable the system to heat up to near the operating point at a gradual rate, avoiding temperature overshoot that may be caused by full-power heating and the resulting deterioration in taste or thermal stress in materials. After preheating, the system seamlessly switches to adaptive temperature control mode, ensuring a smooth temperature curve and consistent user experience throughout the entire usage process. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the overall technical solution architecture of the present invention; Figure 2 This is a schematic diagram of the core principle framework of the parameter adaptive mechanism based on perturbation intensity quantization and fuzzy inference in this invention; Figure 3 This is a schematic diagram of the multi-level interaction relationship and data flow of the dual-ring nested collaborative control architecture in this invention; Figure 4 This is a logical flow diagram of the perturbation quantization module in this invention; Figure 5 This is a flowchart illustrating the logic flow of the preheating and steady-state adaptive mode switching in this invention. Detailed Implementation

[0023] Example 1: The adaptive precision temperature control system for electronic cigarettes based on fuzzy PID control proposed in this invention is shown in the appendix. Figures 1 to 5 As shown, the system comprises a temperature acquisition module, a disturbance quantization module, a fuzzy inference module, a parameter adaptive PID control module, a power drive module, and a start-up preheating control submodule. These modules communicate with each other via an internal data bus and control signal interface, achieving high-precision, low-latency information exchange and command transmission, forming a complete closed-loop temperature control system. The system operates within a nested dual-loop collaborative control architecture. The inner loop is a millisecond-level fast power adjustment loop, while the outer loop is a second-level parameter adaptive loop. These two loops cooperate in terms of time scale and functional division, ensuring that the system maintains high-precision, overshoot-free, and fast-response dynamic control of the atomizing chamber temperature even when faced with strong time-varying nonlinear disturbances caused by user suction behavior.

[0024] Firstly, the temperature acquisition module, as the sensing front end of the entire temperature control system, has the core task of acquiring the actual temperature value of the atomizing cavity in real time and continuously. This module is typically composed of a high-precision negative temperature coefficient thermistor or platinum resistance temperature sensor, with a temperature measurement range covering room temperature to 300 degrees Celsius and a sampling frequency of no less than 200 Hz to ensure the capture of instantaneous temperature fluctuations caused by the user's suction action. The temperature acquisition module converts the raw analog voltage signal into a digital quantity via an analog-to-digital converter and outputs it to the data buffer of the system's main control unit, while simultaneously providing it to the disturbance quantization module and the parameter adaptive PID control module. To ensure the reliability of temperature measurement, this module incorporates a temperature drift compensation algorithm, which can dynamically correct the sensor output based on changes in ambient temperature, thereby eliminating measurement deviations caused by long-term operation.

[0025] The disturbance quantization module is used to evaluate and quantify the intensity of current external disturbances in real time. Its logical flow framework is shown in the attached figure. Figure 4 As shown, this module reads the latest actual temperature value from the temperature acquisition module at each sampling time. And calculate the first-order difference between the current time and the previous sampling time. This difference value represents the instantaneous temperature change rate. Meanwhile, the maintenance length of the disturbance quantization module is... A sliding time window for each sampling period. It is a positive integer greater than or equal to 5, typically 10. In each control cycle, this module calculates all temperature samples within the sliding window. Standard deviation This serves as a statistical measure of temperature fluctuation within that time window. Standard deviation The calculation formula is as follows: in, This is the arithmetic mean of the temperature samples within the sliding window. Subsequently, the perturbation quantization module calculates the absolute value of the instantaneous rate of temperature change. with standard deviation A weighted fusion process is performed to generate a comprehensive disturbance intensity index D. This fusion process employs the following linear weighting method: ,in and For the preset weighting coefficients, satisfy + =1, with typical values ​​of 0.6 and 0.4; and These are the system's preset maximum allowable instantaneous temperature change rate and maximum allowable temperature standard deviation, used to normalize the two components. Finally, the disturbance quantization module will... The value range is mapped to a continuous interval from 0 to 1, and the normalized perturbation strength quantification index is output. .when When the value is close to 0, it indicates that the system is in a stable state with no significant external disturbances; when... When the value is close to 1, it indicates that the system is experiencing a strong disturbance, such as when the user is performing deep suction.

[0026] The fuzzy inference module receives the output from the perturbation quantization module. As the sole input variable, and based on a pre-defined fuzzy rule base, the adjustment amount of the proportional coefficient is calculated in real time. Integral coefficient adjustment and the adjustment amount of the differential coefficient The fuzzy rule base construction process of this module strictly follows the description in the invention. Its input variable perturbation strength is defined with three fuzzy subsets: weak, medium, and strong, corresponding to... For membership functions within the intervals [0, 0.3], [0.2, 0.7], and [0.6, 1.0], triangular or trapezoidal membership functions are used to achieve a smooth transition. Each of the three output variables also has three fuzzy subsets defined. for Slight increase, medium increase, and large increase correspond to [0, 0.2], [0.1, 0.5], and [0.4, 1.0], respectively; for Slight reduction, medium reduction, and large reduction correspond to [0, 0.2], [0.1, 0.5], and [0.4, 1.0] respectively (Note that here, reduction indicates that the adjustment direction is negative, but the output value is still positive and will be processed by subsequent calculation logic). for The domains of incremental, medium, and large increases and same.

[0027] The fuzzy rule base contains three core rules, each employing a Mamdani-type structure of "if...then...". The fuzzy inference process uses the max-min synthesis method, while defuzzification uses the centroid method to convert the fuzzy output set into precise real values. The module's operating cycle is 10 to 50 times the inner loop control cycle, with a typical value of 50 milliseconds, to avoid frequent parameter fluctuations caused by high-frequency noise interference.

[0028] The parameter adaptive PID control module is the core execution unit of the entire temperature control system, and it has a preset proportional coefficient reference value. Integral coefficient benchmark value and differential coefficient benchmark value These reference values ​​are obtained through offline experimental tuning and are suitable for steady-state operation of the system under no or weak disturbance conditions. In each inner-loop control cycle (typically 1 millisecond), the module first obtains the current actual temperature value from the temperature acquisition module. and the preset target temperature value Compare and calculate the current temperature deviation. Simultaneously, this module calculates the rate of change of the deviation. And accumulate the integral term of the deviation. ,in The sampling period is specified. Subsequently, the parameter adaptive PID control module receives three coefficient adjustments from the fuzzy inference module. , , The PID parameters used in the current control cycle are dynamically updated according to the following formula: Current scaling factor ; Current integral coefficient ; Current differential coefficients .

[0029] in, , , The preset gain coefficient is used to adjust the sensitivity and amplitude of the adaptive adjustment. Its value ranges from 0.5 to 2.0, with typical values ​​of 1.2, 1.0, and 1.5. Through this adjustment mechanism, when the disturbance increases, and The size was increased to improve the system's response speed and interference immunity. The parameters are reduced to suppress integral saturation and overshoot; as the disturbance weakens, the parameters are adjusted in the opposite direction to optimize steady-state accuracy. After the parameter update is complete, the module calculates the control output u(k) according to the standard discrete PID control algorithm: This control quantity It is then sent to the power drive module for execution.

[0030] The power drive module consists of two parts: a pulse width modulation (PWM) signal generator and a metal-oxide-semiconductor (MOSFET) drive circuit. The PWM signal generator receives the control input from the parameter-adaptive PID control module. First, limit its amplitude to ensure ,in Corresponding to 0% duty cycle, This corresponds to a 100% duty cycle. Subsequently, the generator will limit the amplitude... Linear mapping to duty cycle instructions The calculation formula is: Based on this duty cycle command, the pulse width modulation signal generator generates a square wave signal with a fixed frequency (typically 20 kHz) and an adjustable duty cycle. This square wave signal is input to a metal-oxide-semiconductor field-effect transistor (MOSFET) driver circuit, which amplifies and levels the signal to drive a heating resistance wire connected to the atomizing cavity. The average heating power of the heating resistance wire... It is directly proportional to the duty cycle, that is ,in This is the rated power when heating at full power. By adjusting the duty cycle, the system achieves continuous and linear control of the heating power, thereby precisely regulating the temperature of the atomizing chamber.

[0031] The system also includes a preheating control submodule, the logic flow of which is shown in the attached figure. Figure 5 As shown. This submodule is integrated within the parameter adaptive PID control module and is automatically activated during system power-on initialization or wake-up from a low-power sleep state. After activation, this submodule first temporarily sets the target temperature value to the preheating temperature value. This value is typically the steady-state target temperature. 70% to 80%, for example when At 220 degrees Celsius, The temperature is set to 160 degrees Celsius. Simultaneously, this submodule forcibly disables the output of the fuzzy inference module. , , Setting all parameters to zero allows the adaptive PID control module to use only preset reference parameters. , , To take control.

[0032] In this mode, the system heats up slowly with lower power to avoid initial current surges and temperature overshoots caused by the low thermal resistance of the heating element when cold. When the actual temperature value... Stable for 2 to 3 seconds When the temperature is within ±5 degrees Celsius, the preheating control submodule is activated to determine that preheating is complete, and the preheating mode is automatically exited: the target temperature value is switched back. The output of the fuzzy inference module is reconnected to the parameter adaptive PID control module, and the system officially enters the complete adaptive precision temperature control mode. This switching process is smooth and seamless, without control jumps or temperature oscillations.

[0033] The dual-loop nested collaborative control architecture of the entire system is shown in the attached figure. Figure 3 As shown, the inner loop consists of a parameter-adaptive PID control module and a power drive module, with an operating cycle on the order of 1 millisecond. It is responsible for real-time suppression of high-frequency temperature fluctuations, ensuring the speed and stability of control. The outer loop consists of a disturbance quantization module and a fuzzy inference module, with an operating cycle on the order of 50 milliseconds. It is responsible for analyzing disturbance characteristics on a longer time scale and periodically optimizing the parameters of the inner loop controller. The inner and outer loops exchange information through a shared temperature data buffer: the inner loop continuously writes the latest temperature sample value to the buffer, and the outer loop periodically reads data from it for disturbance quantization. At the same time, the coefficient adjustment calculated by the outer loop is transmitted to the parameter-adaptive PID control module of the inner loop through a dedicated register interface. This architecture effectively separates the two major functions of fast execution and intelligent decision-making, avoiding control noise introduced by frequent parameter adjustments and ensuring the system's rapid adaptability to sudden strong disturbances.

[0034] In actual operation, when the user is not pumping, the system is in a low-disturbance condition, and the disturbance quantization module outputs... The output of the fuzzy inference module is close to 0. , , The parameters are relatively small, and the adaptive PID control module mainly relies on reference parameters for fine adjustment, keeping temperature fluctuations within ±1 degree Celsius. When the user begins to inhale, the airflow carries away a large amount of heat, causing a sudden drop in the temperature of the atomizing chamber. At this point, the temperature acquisition module detects a significant negative temperature fluctuation. With increase Perturbation quantization module output It quickly rose to above 0.8.

[0035] The fuzzy inference module triggers a strong perturbation rule based on this, resulting in a larger output. and At the same time, it outputs a large amount of data. (used to reduce) The parameter adaptive PID control module immediately increases... and Improve heating response speed and reduce To prevent the accumulation of excessive integral terms due to persistent deviations, the power drive module increases its duty cycle to quickly compensate for heat loss. After the suction process ends, the disturbance intensity decreases, and the system automatically adjusts its parameters to restore high-precision steady-state control. Throughout the process, the temperature curve shows no significant overshoot, and the recovery time is less than 1 second, significantly outperforming traditional fixed-parameter PID controllers.

[0036] Furthermore, the system possesses a comprehensive fault diagnosis and safety protection mechanism. The temperature acquisition module continuously monitors the validity of the sensor signal. If an open circuit, short circuit, or abnormal value exceeding the measurement range is detected, a safety interrupt is immediately triggered, shutting down the power drive module and entering a fault lockout state. The power drive module has a built-in overcurrent protection circuit; when the heating resistance wire current exceeds the safety threshold, the drive signal is automatically cut off. The parameter adaptive PID control module controls the output... Implement dual limiting to prevent the duty cycle from exceeding 100% and to avoid outputting negative power commands at low temperatures. All key parameters (such as...) All data (such as electronic cigarette devices, etc.) are stored in non-volatile memory and can be adjusted online via firmware upgrades to adapt to different models of electronic cigarette devices or user preference settings.

[0037] In summary, this embodiment constructs a complete closed-loop system consisting of temperature acquisition, disturbance quantization, fuzzy inference, parameter adaptive PID control, power drive, and start-up preheating submodules. Employing a dual-loop nested collaborative control architecture, it achieves adaptive, precise, rapid, and overshoot-free dynamic control of the e-cigarette atomization chamber temperature. This system intelligently senses changes in disturbance intensity caused by the user's inhalation behavior and optimizes control parameters in real time accordingly. It ensures rapid response under strong disturbance conditions and maintains high steady-state accuracy under stable conditions. Furthermore, the preheating mechanism avoids cold-start shocks, comprehensively improving the safety, flavor consistency, and user experience of e-cigarettes.

[0038] Example 2: Based on Example 1, this example further optimizes the quantization logic of the disturbance quantization module to improve the accuracy of disturbance identification under complex and variable suction modes. Specifically, the disturbance quantization module not only considers the first-order difference and standard deviation of temperature, but also introduces second-order difference features as auxiliary criteria to distinguish between persistent disturbances and instantaneous impact disturbances. At each sampling moment, the disturbance quantization module calculates the first-order difference... In addition, the second-order difference is calculated. The second-order difference reflects the acceleration of the rate of temperature change and can be used to determine whether a disturbance has a sustained trend. For example, continuous shallow suction by a user may produce a small but continuous temperature drop. In this case, the absolute value of the first-order difference is not large, but the second-order difference is close to 0. However, a single deep suction will produce a large negative first-order difference, and the second-order difference will also be negative, indicating that the cooling trend is accelerating.

[0039] The perturbation quantization module converts the absolute value of the second-order difference into... Incorporate into the weighted fusion model. New perturbation intensity index. The calculation formula is: ,in + + =1, typical weight allocation is =0.5, =0.3, =0.2; This is the preset maximum allowable absolute value of the second-order difference. By introducing the second-order difference, the system can more accurately distinguish different types of disturbance patterns, thus providing a more discriminative input for the fuzzy inference module. For example, when a high... But low When the system judges it as a transient disturbance, only a moderate increase in proportional effect is needed; while when and When the average value is high, it is determined to be a strong and persistent perturbation, requiring a significant enhancement of the proportional and derivative actions and a substantial reduction in the integral action. This optimization makes the parameter adjustment strategy more targeted, further reducing the temperature fluctuation amplitude in complex suction scenarios.

[0040] Furthermore, this embodiment expands the rule base of the fuzzy inference module. In addition to the original three rules, a new rule is added for extremely weak perturbation states: if the perturbation strength is extremely weak ( If the value is less than 0.1, the proportional gain adjustment is slightly reduced, the integral gain adjustment is slightly increased, and the derivative gain adjustment is slightly reduced. This rule aims to slightly reduce the proportional gain to decrease sensitivity to high-frequency noise when the system is in a super-steady state, while enhancing the integral action to eliminate residual static error, and moderately weakening the derivative action to avoid over-response to measurement noise. The fuzzy subsets of the input variable disturbance intensity are expanded from three to four: extremely weak, weak, medium, and strong, with corresponding universes of discourse of [0, 0.15], [0.1, 0.4], [0.3, 0.7], and [0.6, 1.0], respectively. Sufficient overlap is maintained between subsets to ensure inference continuity. The number of fuzzy subsets of the output variable is increased accordingly, but the adjustment direction logic remains consistent. This expansion enables the system to have higher control accuracy under extremely steady conditions, and the steady-state error can be further compressed to within ±0.5 degrees Celsius.

[0041] At the hardware implementation level, this embodiment uses a higher-performance 32-bit microcontroller as the main control chip. Its built-in hardware floating-point unit can accelerate the calculation of standard deviation and fuzzy inference, ensuring that even after the introduction of second-order difference and extended rule base, the outer loop control cycle can still be stably maintained at 50 milliseconds. The temperature acquisition module is upgraded to a dual-sensor redundant configuration, with two independent thermistors installed at different positions in the atomization cavity. The system fuses the two signals through a weighted average or optimal selection strategy, effectively overcoming the local thermal hysteresis problem that may exist in single-point temperature measurement. The pulse width modulation frequency of the power drive module is increased to 50 kHz to further reduce the current ripple of the heating resistance wire, reduce electromagnetic interference, and improve the linearity of power regulation.

[0042] Through the above improvements, this embodiment, while maintaining all the advantages of embodiment 1, significantly enhances the system's ability to identify the type and intensity of disturbances, making the adaptive parameter adjustment strategy more refined and intelligent, and is especially suitable for the demand for extreme temperature control performance in high-end electronic cigarette products.

Claims

1. An adaptive and precise temperature control system for electronic cigarettes based on fuzzy PID control, characterized in that, include: The temperature acquisition module is used to acquire the actual temperature value of the atomizing chamber in real time. The disturbance quantization module is used to calculate and output a quantitative index characterizing the intensity of the current external disturbance in real time based on the actual temperature value and its change characteristics collected by the temperature acquisition module. The fuzzy inference module receives the quantization index output by the perturbation quantization module and calculates and outputs the proportional coefficient adjustment, integral coefficient adjustment, and differential coefficient adjustment in real time according to the preset fuzzy rule library. The parameter adaptive PID control module receives the actual temperature value output by the temperature acquisition module and the preset target temperature value, and calculates the temperature deviation and its rate of change. At the same time, the module receives the proportional coefficient adjustment, integral coefficient adjustment, and derivative coefficient adjustment output by the fuzzy inference module, and dynamically adjusts its internal proportional coefficient, integral coefficient, and derivative coefficient. Based on the dynamically adjusted parameters, the module performs proportional-integral-derivative operations to generate a real-time control output signal. The power drive module receives the control output signal generated by the parameter adaptive PID control module and converts it into a corresponding pulse width modulation signal to drive the heating element to work, thereby realizing closed-loop control of the temperature of the atomizing chamber.

2. The adaptive precision temperature control system for electronic cigarettes based on fuzzy PID control according to claim 1, characterized in that, The quantization process of the disturbance quantization module is as follows: The module's internal length is set to... A sliding time window for each sampling period; At each sampling moment, the perturbation quantization module obtains the latest actual temperature value from the temperature acquisition module and calculates the first-order difference between the current temperature value and the temperature value at the previous sampling moment, i.e., the instantaneous temperature change rate. At the same time, this module calculates the standard deviation of all temperature samples within the current sliding time window, which serves as a statistical measure of temperature fluctuation within that window; The disturbance quantification module weights and fuses the absolute value of the instantaneous temperature change rate with the temperature standard deviation to generate a comprehensive disturbance intensity quantification index.

3. The adaptive precision temperature control system for electronic cigarettes based on fuzzy PID control according to claim 1, characterized in that, The fuzzy rule base construction process of the fuzzy inference module is as follows: The perturbation quantification index is used as the only input variable, and the proportional coefficient adjustment, integral coefficient adjustment, and derivative coefficient adjustment are used as three independent output variables. The universe of discourse for both input and output variables is defined as a continuous interval from 0 to 1; Three fuzzy subsets are defined for the intensity of the input variable disturbance, named weak, medium, and strong respectively; Three fuzzy subsets are defined for the adjustment amount of the output variable proportional coefficient, named micro-increase, medium-increase, and large-increase respectively; Three fuzzy subsets are defined for the adjustment of the integral coefficients of the output variables, named as micro reduction, medium reduction, and large reduction respectively; Three fuzzy subsets are defined for the adjustment of the differential coefficients of the output variables, named micro-increase, medium-increase, and large-increase respectively; The fuzzy rule base contains 3 rules, described as follows: Article 1: If the disturbance intensity is weak, the adjustment amount of the proportional coefficient is slightly increased, the adjustment amount of the integral coefficient is slightly decreased, and the adjustment amount of the differential coefficient is slightly increased. Article 2. If the disturbance intensity is medium, the adjustment amount of the proportional coefficient is medium increase, the adjustment amount of the integral coefficient is medium decrease, and the adjustment amount of the differential coefficient is medium increase. Article 3: If the disturbance intensity is strong, the adjustment amount of the proportional coefficient is greatly increased, the adjustment amount of the integral coefficient is greatly decreased, and the adjustment amount of the differential coefficient is greatly increased. The fuzzy inference module uses the centroid method for defuzzification calculation, converting the fuzzy inference results into precise coefficient adjustment values ​​for output.

4. The adaptive precision temperature control system for electronic cigarettes based on fuzzy PID control according to claim 1, characterized in that, The parameter dynamic adjustment process of the parameter adaptive PID control module is as follows: The parameter adaptive PID control module has a preset proportional coefficient reference value. Integral coefficient benchmark value and differential coefficient benchmark value ; In each control cycle, the parameter adaptive PID control module receives three coefficient adjustments from the fuzzy inference module, denoted as follows: , and ; The parameter adaptive PID control module calculates the PID parameters used in the current control cycle in real time according to the following formula: Current scaling factor ; Current integral coefficient ; Current differential coefficients ; in, , , The preset gain coefficient is used to adjust the magnitude of the adaptive adjustment; after calculating the dynamic parameters, the parameter adaptive PID control module adjusts the current temperature deviation. The integral and derivative of the temperature deviation are used to calculate the control output according to the standard PID control algorithm. .

5. The adaptive precision temperature control system for electronic cigarettes based on fuzzy PID control according to claim 1, characterized in that, The power drive module includes a pulse width modulation signal generator and a metal-oxide-semiconductor field-effect transistor drive circuit. The pulse width modulation signal generator receives the control quantity u output by the parameter adaptive PID control module. After the control quantity is limited, it is mapped to the duty cycle command. The pulse width modulation signal generator generates a pulse width modulation square wave signal with a corresponding duty cycle according to the duty cycle command; The metal-oxide-semiconductor field-effect transistor driving circuit receives the pulse width modulation square wave signal, amplifies it, and then drives the heating resistance wire connected to the atomizing cavity.

6. The adaptive precision temperature control system for electronic cigarettes based on fuzzy PID control according to claim 1, characterized in that, The system also includes a startup preheating control submodule, which is integrated into the parameter adaptive PID control module; the startup preheating control submodule is activated when the system is powered on or wakes up from hibernation.

7. The adaptive precision temperature control system for electronic cigarettes based on fuzzy PID control according to claim 6, characterized in that, The preheating control submodule first sets the target temperature value to a preheating temperature value lower than the steady-state target value. At the same time, the preheating control submodule temporarily overrides the output of the fuzzy inference module, forcibly setting the proportional coefficient adjustment, integral coefficient adjustment, and derivative coefficient adjustment to 0, so that the adaptive PID control module uses its preset reference parameters Kp0, Ki0, and Kd0 for control. When the actual temperature value reaches and stabilizes near the preheating temperature value for a preset period of 2 to 3 seconds, the preheating control submodule automatically exits, the system resumes switching the target temperature value to the steady-state target value, and resumes receiving the output of the fuzzy inference module.

8. The adaptive precision temperature control system for electronic cigarettes based on fuzzy PID control according to claim 1, characterized in that, The system operates in a dual-loop nested collaborative control architecture. The inner loop is a millisecond-level fast power regulation loop, consisting of a parameter adaptive PID control module and a power drive module, responsible for rapidly suppressing high-frequency temperature fluctuations. The outer loop is a second-level parameter adaptive loop, consisting of a disturbance quantization module and a fuzzy inference module, with an operating cycle 10 to 50 times that of the inner loop control cycle. The outer loop is responsible for monitoring disturbance pattern changes over a longer time scale and periodically optimizing and adjusting the parameters of the inner loop controller accordingly.

9. The adaptive precision temperature control system for electronic cigarettes based on fuzzy PID control according to claim 2, characterized in that, When the disturbance quantization module weights and fuses the absolute value of the instantaneous temperature change rate with the temperature standard deviation, it adopts a linear weighting form, wherein the absolute value component of the instantaneous temperature change rate and the temperature standard deviation component are normalized by the preset maximum allowable instantaneous temperature change rate and the maximum allowable temperature standard deviation, respectively.

10. The adaptive precision temperature control system for electronic cigarettes based on fuzzy PID control according to claim 3, characterized in that, The fuzzy subsets of the perturbation intensity of the input variable in the fuzzy inference module, namely weak, medium, and strong, correspond to the membership functions of the quantization index in the intervals of 0 to 0.3, 0.2 to 0.7, and 0.6 to 1.0, respectively. The fuzzy subsets of the output variable proportional coefficient adjustment amount, namely micro-increase, medium-increase, and large-increase, correspond to the universe of discourse of 0 to 0.2, 0.1 to 0.5, and 0.4 to 1.0, respectively.

Images