Fuzzy PID (Proportion Integration Differentiation) stable control method and system for neutron tube acceleration voltage

Abstract

The invention relates to the technical field of high-voltage power supply control of a petroleum logging instrument, in particular to a fuzzy PID (Proportion Integration Differentiation) stable control method and a fuzzy PID stable control system for neutron tube acceleration voltage, and aims to solve the technical problems of voltage drop and difficulty in quick recovery caused by internal resistance soft characteristics and high-temperature parameter time-varying characteristics of an existing voltage doubling circuit. The method comprises the following steps: acquiring a current real-time running state of a system; calculating a feed-forward compensation duty ratio required for counteracting the natural voltage drop under the current working condition; the dynamic proportional gain is calculated in real time, and feedback control output is obtained by combining the integral term; and generating a PWM duty ratio which is finally output to the inverter full bridge so as to control the acceleration voltage of the neutron tube. Through a feed-forward compensation mechanism, the output characteristic of the high-voltage power supply is improved physically and essentially, so that the output characteristic is close to the characteristic of an ideal voltage source.

Description

Technical Field

[0001] This invention relates to the field of high-voltage power supply control technology for oil well logging instruments, and more specifically, to a fuzzy PID stable control method and system for neutron tube accelerating voltage. Background Technology

[0002] In neutron tube operation, the stability of the accelerating voltage directly determines the consistency of neutron production, thus affecting the accuracy of logging data. The accelerating voltage is typically generated by a voltage multiplier circuit. In actual logging, especially under extreme conditions during drilling, the control system faces severe engineering challenges.

[0003] Voltage multiplier circuits are not ideal voltage sources, exhibiting significant soft-characteristic internal resistance and extremely high output impedance, typically reaching the megaohm level. When the ion source is turned on, the target current load abruptly jumps from 0 to a high level (e.g., 100μA). According to the voltage multiplier principle, this sudden load change inevitably leads to a physical voltage drop, and the magnitude of this drop exhibits a complex nonlinear relationship with the load current, drive frequency, and capacitor losses. Furthermore, the downhole high-temperature environment can reach 175℃, causing an increase in the dielectric loss tangent of the high-voltage capacitor, resulting in a decrease in the equivalent capacitance. This leads to slow voltage recovery in the high-temperature environment downhole due to insufficient driving force, even for PID parameters well-tuned on the surface, exhibiting time-varying parameter issues.

[0004] Existing control technologies typically employ conventional PID algorithms, treating the voltage multiplier circuit as a black box and adjusting it only based on voltage deviation. This approach cannot fundamentally overcome the inherent internal resistance voltage drop of the circuit, and the simple linear gain cannot simultaneously achieve rapid rise during startup and maintain low ripple during steady state, making it difficult to meet the requirements of high-precision logging. Summary of the Invention

[0005] This invention provides a fuzzy PID stable control method and system for neutron tube accelerating voltage, in order to solve the technical problem that the voltage drop caused by the soft internal resistance and time-varying high temperature parameters of existing voltage multiplier circuits is difficult to recover quickly.

[0006] In a first aspect, the present invention provides a fuzzy PID stable control method for neutron tube accelerating voltage, comprising: A multi-dimensional state-aware feedback link is constructed, and a high-frequency synchronous acquisition mechanism is established through the main control chip to obtain the current real-time operating status of the system. The real-time operating status includes high voltage feedback quantity, load interference quantity, and thermal state quantity. Dynamic impedance feedforward compensation is performed based on the physical inverse model of the voltage multiplier circuit. According to the load interference, thermal state and preset circuit parameters, the feedforward compensation duty cycle required to offset the inherent voltage drop under the current operating condition is calculated. Based on the exponential reaching law, a variable structure PID feedback regulation is performed. According to the voltage tracking error between the set target voltage and the high voltage feedback quantity, the dynamic proportional gain is calculated in real time, and the feedback control output is obtained by combining the integral term. A dual-channel vector synthesis output is performed, in which the feedforward compensation duty cycle and the feedback control output are vector-superimposed, and after amplitude limiting protection processing, the final output PWM duty cycle is generated to the inverter full bridge, and a PWM drive signal is generated accordingly to control the acceleration voltage of the neutron tube.

[0007] By establishing a high-frequency synchronous acquisition mechanism, the system's voltage output, load changes, and ambient temperature changes can be obtained in real time, providing a data foundation for subsequent precise control and avoiding the information loss problem of single-variable control.

[0008] Furthermore, a multi-dimensional state-aware feedback loop is constructed, including: using the parallel ADC channel of a DSP or FPGA controller to synchronously acquire data at a preset sampling rate.

[0009] Furthermore, constructing a multi-dimensional state-aware feedback loop also includes: The high voltage feedback quantity is obtained by acquiring real-time voltage through a high voltage divider arm and performing sliding window filtering. The real-time current flowing through the target circuit is collected as the load interference quantity; The ambient temperature around the power device and the voltage multiplier cylinder is collected as the thermal state quantity.

[0010] Furthermore, the real-time current flowing through the target circuit is acquired by connecting a sampling resistor in series with the target circuit or by using a current sensor.

[0011] Furthermore, dynamic impedance feedforward compensation is performed based on the physical inverse model of the voltage multiplier circuit, including: The temperature-corrected equivalent load factor is calculated based on the following formula: ; In the formula, The equivalent loading factor, The load interference amount, The inverter drive frequency, This is the nominal value of the voltage multiplier capacitor. The thermal decay coefficient of the capacitor. The thermal state quantity is denoted as .

[0012] By constructing an inverse model based on physical principles and introducing a temperature correction term, the inherent voltage drop caused by load mutations and temperature rises can be mathematically calculated and offset precisely, enabling the power supply to exhibit zero internal resistance hard characteristics in engineering performance and improving the system's anti-disturbance capability.

[0013] Furthermore, dynamic impedance feedforward compensation based on the physical inverse model of the voltage multiplier circuit also includes: The feedforward compensation duty cycle is calculated according to the following formula: ; In the formula, To compensate for the duty cycle, The geometric constant of the voltage multiplier series is . This is the normalized gain coefficient.

[0014] Furthermore, the dynamic proportional gain satisfies the following relationship: ; In the formula, For dynamic proportional gain, To maintain the gain in steady state, For transient burst gain, This is the variable structure sensitivity factor. The voltage tracking error is given, and , For the set target voltage, This refers to the high-voltage feedback quantity.

[0015] Furthermore, the feedback control output satisfies the following relationship: ; In the formula, For feedback control output, This is the integral gain coefficient.

[0016] By using a variable structure gain design based on the exponential reaching law, the controller can adaptively adjust the dynamic proportional gain according to the magnitude of the voltage tracking error: providing high gain to accelerate the dynamic response when the error is large, and reducing the gain to suppress the output ripple when the error is small. This ensures rapid recovery under load changes while suppressing steady-state overshoot and oscillation, thus reconciling the contradiction between the system's dynamic response speed and steady-state accuracy.

[0017] Furthermore, the output of the dual-channel vector synthesis satisfies the following relationship: ; In the formula, To determine the PWM duty cycle that will ultimately be output to the inverter full-bridge, For the amplitude limiting function, The feedforward compensation duty cycle is... This is the feedback control output.

[0018] By combining the coarse-tuned feedforward quantity with the fine-tuned feedback quantity through dual-channel vector synthesis and applying software limiting protection, the response speed and accuracy of the control signal are guaranteed, while preventing excessive drive signals from damaging the hardware circuits, thus ensuring the safe and reliable operation of the system.

[0019] Secondly, the present invention provides a fuzzy PID stable control system for neutron tube accelerating voltage, comprising: a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned fuzzy PID stable control method for neutron tube accelerating voltage is implemented.

[0020] The beneficial effects are as follows: This invention provides a composite control architecture based on inverse model dynamic impedance feedforward and variable structure nonlinear feedback. It does not treat voltage sags as simple errors, but rather calculates the feedforward compensation duty cycle required to offset voltage sags based on the physical equations of the voltage multiplier circuit. Simultaneously, a variable structure control law is used in the feedback loop to construct a nonlinear controller. Through the feedforward compensation mechanism, this invention fundamentally improves the output characteristics of the high-voltage power supply, making it closer to the characteristics of an ideal voltage source. Furthermore, the control method of this invention has strong adaptability, operating stably under various conditions such as no-load to full-load and room temperature to high temperature without parameter adjustment. Moreover, the algorithm structure is simple, computationally inexpensive, and executes quickly, making it suitable for implementing high-frequency closed-loop control on controllers such as FPGAs or DSPs. Attached Figure Description

[0021] Figure 1 A fuzzy PID stable control method for neutron tube accelerating voltage; Figure 2 The dynamic mechanism diagram of variable gain control; Figure 3 This is a comparison chart of the power supply output impedance characteristic curves. Detailed Implementation

[0022] An embodiment of the fuzzy PID stable control method for neutron tube accelerating voltage provided by this invention: like Figure 1 As shown, a fuzzy PID stable control method for neutron tube accelerating voltage includes the following steps: S101. Construct a multi-dimensional state-aware feedback link. Establish a high-frequency synchronous acquisition mechanism through the main control chip to obtain the current real-time operating status of the system. The real-time operating status includes high-voltage feedback quantity, load interference quantity, and thermal state quantity.

[0023] Specifically, in order to achieve precise control of the neutron tube accelerating voltage, a data acquisition mechanism capable of real-time sensing of the system's internal state and external environment is established. In this embodiment, the parallel ADC channel of the DSP or FPGA controller is used to synchronously acquire key system state quantities at a sampling rate of 10kHz: high voltage feedback quantity, load interference quantity, and thermal state quantity.

[0024] The specific data collection process is as follows: 1) Obtaining high voltage feedback: The real-time voltage signal output by the voltage multiplier circuit is acquired through a precision high voltage divider arm circuit. In order to filter out high frequency switching noise, the acquired raw data will be processed by digital sliding window filtering to obtain a smooth high voltage feedback value.

[0025] 2) Obtain the load interference quantity: By connecting a sampling resistor in series in the target circuit or using a current sensor, the real-time current flowing through the target circuit is collected. The real-time current directly reflects the load intensity of the ion beam after the ion source is turned on, and is the main source of disturbance that causes voltage drop; the real-time current is used as the load interference quantity.

[0026] 3) Acquiring thermal state parameters: Real-time ambient temperature is collected around the power switching devices and voltage multiplier cylinder using temperature sensors, and this ambient temperature is used as a thermal state parameter. For example, the ambient temperature around the MOSFET or IGBT can be collected in real time using a PT100 or thermistor. In the high-temperature environment downhole, temperature parameters are crucial for correcting capacitor parameters.

[0027] By constructing a multi-dimensional state-aware feedback loop, voltage, current, and temperature data can be refreshed at microsecond speeds, providing real-time and accurate physical basis for subsequent feedforward compensation and feedback regulation, ensuring that the control algorithm can perceive subtle changes in the system.

[0028] S102. Perform dynamic impedance feedforward compensation based on the physical inverse model of the voltage multiplier circuit. Calculate the feedforward compensation duty cycle required to offset the inherent voltage drop under the current operating conditions based on the load interference, thermal state, and preset circuit parameters.

[0029] Specifically, in this embodiment, reverse modeling is performed based on the voltage drop physical principle of the voltage multiplier circuit. The load interference and thermal state quantities collected in real time, as well as the inverter drive frequency, nominal value of the voltage multiplier capacitor and its thermal attenuation coefficient preset by the system, are substituted into the physical model to calculate the inherent voltage drop value under the current operating condition in real time, and normalize it into the feedforward compensation duty cycle, which is then injected into the system in advance.

[0030] The specific calculation process consists of two steps: The first step is to calculate the temperature-corrected equivalent load factor. The equivalent load factor represents the theoretical voltage drop per unit voltage multiplication stage under the current load disturbance and current thermal conditions. The relationship is as follows: ; in: This is the equivalent loading factor; This is the load interference, and its unit is A; This is the inverter drive frequency, and its unit is Hz; This is the nominal value of the voltage multiplier capacitor, for example, the nominal value of the voltage multiplier capacitor at 25°C, and its range is F; This is the capacitor thermal decay coefficient, and its unit is . The thermal decay coefficient of the capacitor was obtained by calibration through a high-temperature aging test. 25 represents the thermal state quantity, with units of °C; 25 represents the reference temperature, with units of °C.

[0031] In one example, at a certain moment, the system's operating parameters are as follows: ; ; ; ; .

[0032] Substitute into the formula to calculate the equivalent loading factor: .

[0033] The results indicate that under the current high temperature and high load conditions, the base voltage of each voltage multiplier circuit drops by 1.136V.

[0034] The second step is to calculate the feedforward compensation duty cycle, the relationship of which is shown below: ; in: This is for feedforward compensation of duty cycle; The geometric constant of the voltage multiplier series is related to the voltage multiplier series. Related, approximate Order of magnitude; This is the normalized gain coefficient, and its unit is... Its value is equal to ,in This is the maximum input voltage of the inverter.

[0035] In one example, continuing from the previous example, let's assume the voltage multiplier level... The geometric constant of the voltage multiplier stage is determined based on the specific circuit structure. ; ,but .

[0036] Substitute the formula to calculate the feedforward compensation duty cycle: ; The feedforward channel will directly output a feedforward compensation duty cycle of approximately 34.08% to preemptively offset impending voltage dips.

[0037] Through dynamic calculations based on a physical inverse model, the control system no longer passively waits for voltage drops before adjusting, but can proactively predict and inject the required energy based on the current load and temperature conditions. This feedforward mechanism smooths out the power supply output impedance at the algorithm level, eliminating the internal resistance effect of the voltage multiplier circuit.

[0038] S103. Based on the exponential approach law, a variable structure PID feedback regulation is performed. According to the voltage tracking error between the set target voltage and the high voltage feedback quantity, the dynamic proportional gain is calculated in real time, and the feedback control output is obtained by combining the integral term.

[0039] Specifically, in this embodiment, the control law is designed as a variable structure. The controller automatically calculates the dynamic proportional gain according to the magnitude of the voltage tracking error, thereby exhibiting different control characteristics within different error ranges.

[0040] The specific calculation process is as follows: The first step is to calculate the dynamic proportional gain, and the relationship is shown below: ; In the formula, For dynamic proportional gain; To maintain steady-state gain, it is used to provide moderate rigidity when the voltage is stable, preventing oscillations; Transient burst gain is used to provide maximum drive force when a sudden load change causes a voltage drop; it is typically set to... ; The variable structure sensitivity factor determines the steepness of the gain change; Let be the absolute value of the voltage tracking error, and , For the set target voltage, This refers to the high-voltage feedback quantity.

[0041] In one example, suppose the following parameters are set: , Sensitivity factor .

[0042] In steady-state scenarios: the voltage tracking error is very small, assuming... , .

[0043] At this point, the dynamic proportional gain approaches the steady-state gain, and the system maintains a low gain to avoid noise amplification.

[0044] In a sudden change scenario: a sudden load change causes a voltage drop, assuming... , .

[0045] At this point, the dynamic proportional gain is close to the transient burst gain, and the system outputs a very strong regulating force, quickly pulling the voltage back.

[0046] The second step is to calculate the feedback control output, the relationship of which is shown below: ; in, For feedback control output, This is the integral gain coefficient. This is the cumulative sum of voltage tracking errors. Step S103 combines dynamic proportional action and integral action, utilizing high gain to quickly eliminate large errors and using the integral term to eliminate steady-state errors.

[0047] By introducing a variable structure gain design with an exponential reaching law, the controller can intelligently identify the state of the system: in steady state, a low gain strategy is adopted to suppress output ripple, and in transient state, a high gain mode is switched to improve the ability to resist disturbances, thus balancing the steady-state accuracy and dynamic response speed of the system.

[0048] S104. Perform dual-channel vector synthesis output, vector superimpose the feedforward compensation duty cycle and the feedback control output, and generate the final output PWM duty cycle to the inverter full bridge after amplitude limiting protection processing, and generate PWM drive signal accordingly to control the acceleration voltage of the neutron tube.

[0049] Specifically, the feedforward compensation duty cycle based on the physical model is vector-superimposed with the feedback control output based on nonlinear error to obtain the composite duty cycle. Then, its effective range is constrained by the amplitude limiting function, and finally, the PWM drive signal used to drive the front-stage inverter full bridge of the voltage multiplier circuit is generated.

[0050] Its output logic satisfies the following relationship: ; in: This is the PWM duty cycle that is ultimately output to the inverter full bridge; To compensate for the duty cycle, it undertakes more than 90% of the load driving tasks. Its response speed depends on the ADC sampling rate, and it has no additional phase lag due to the use of synchronous direct drive architecture. The output is a feedback control, used to correct residual steady-state errors and suppress environmental disturbances; This is a limiting function that restricts the PWM duty cycle to a certain value. Within a range, for example =95%, to prevent excessive PWM duty cycle from causing shoot-through damage to the upper and lower transistors of the inverter full bridge.

[0051] The controller configures the internal PWM peripherals in real time according to the PWM duty cycle, generates the corresponding PWM drive signal and outputs it to the inverter full-bridge power stage.

[0052] In one example, continuing from the previous example, suppose... The combined PWM duty cycle is then: .

[0053] In other examples, if the calculated PWM duty cycle reaches 120% due to some anomaly, the limiting function keeps it at 95% to protect the circuit.

[0054] By employing dual-channel vector synthesis and limiting protection, not only is precise compensation and rapid adjustment of voltage dips achieved, but the safety of power devices is also ensured through a software limiting mechanism. The feedforward channel enhances the system's dynamic response speed, while the feedback channel guarantees steady-state control accuracy. Together, they achieve highly stable control of the neutron tube accelerating voltage.

[0055] The control method of the present invention has shown significant advantages in practical simulation and application. In an optional embodiment, such as... Figure 2 and Figure 3 As shown: Figure 2 The dynamic mechanism of variable gain control is demonstrated. The upper subplot shows the step change in load current at t=10ms, simulating a full-load impact. The middle subplot compares the voltage response, showing that traditional PID controllers experience a significant voltage drop at the abrupt change and recover slowly, while the present invention exhibits only a very slight fluctuation at the moment of the abrupt change, followed by a rapid recovery with slight oscillations, locking into steady state within milliseconds. The lower subplot shows the change in controller gain. At the moment of slight voltage fluctuation, the controller gain automatically and significantly increases, and then gradually returns to the reference value with slight fluctuations during the subsequent recovery process, demonstrating the variable structure controller's keen ability to capture system states.

[0056] Figure 3 The load regulation characteristics of the power supply are shown, with the horizontal axis representing the load current (ranging from 0-120μA) and the vertical axis representing the voltage sag. Traditional solutions exhibit a pronounced upward slope, indicating that the heavier the load, the greater the voltage sag; this phenomenon is known as the high-resistance soft characteristic. In contrast, the solution of this invention presents a nearly horizontal straight line, demonstrating that regardless of load changes, the voltage sag is always controlled within a very small range; this characteristic is known as the zero-resistance hard characteristic, intuitively proving the effectiveness of the feedforward compensation algorithm.

[0057] An embodiment of the fuzzy PID stabilization control system for neutron tube accelerating voltage provided by this invention: The fuzzy PID stable control system for neutron tube accelerating voltage includes a processor and a memory. The memory stores computer program instructions, which are executed by the processor to implement the aforementioned fuzzy PID stable control method for neutron tube accelerating voltage.

[0058] The fuzzy PID stabilization control system for neutron tube accelerating voltage also includes other components well known to those skilled in the art, such as communication interfaces. Their settings and functions are known in the art and will not be described in detail here.

[0059] In addition, in the description of this specification, "multiple" means at least two, such as two, three or more, etc., unless otherwise expressly and specifically defined.

Claims

1. A fuzzy PID stable control method for neutron tube accelerating voltage, characterized in that, include: A multi-dimensional state-aware feedback link is constructed, and a high-frequency synchronous acquisition mechanism is established through the main control chip to obtain the current real-time operating status of the system. The real-time operating status includes high voltage feedback quantity, load interference quantity, and thermal state quantity. Dynamic impedance feedforward compensation is performed based on the physical inverse model of the voltage multiplier circuit. According to the load interference, thermal state and preset circuit parameters, the feedforward compensation duty cycle required to offset the inherent voltage drop under the current operating condition is calculated. Based on the exponential reaching law, a variable structure PID feedback regulation is performed. According to the voltage tracking error between the set target voltage and the high voltage feedback quantity, the dynamic proportional gain is calculated in real time, and the feedback control output is obtained by combining the integral term. A dual-channel vector synthesis output is performed, in which the feedforward compensation duty cycle and the feedback control output are vector-superimposed, and after amplitude limiting protection processing, the final output PWM duty cycle is generated to the inverter full bridge, and a PWM drive signal is generated accordingly to control the acceleration voltage of the neutron tube.

2. The fuzzy PID stable control method for neutron tube accelerating voltage according to claim 1, characterized in that, Constructing a multi-dimensional state-aware feedback loop includes: using the parallel ADC channel of a DSP or FPGA controller to synchronously acquire data at a preset sampling rate.

3. The fuzzy PID stable control method for neutron tube accelerating voltage according to claim 2, characterized in that, Constructing a multi-dimensional state-aware feedback loop also includes: The high voltage feedback quantity is obtained by acquiring real-time voltage through a high voltage divider arm and performing sliding window filtering. The real-time current flowing through the target circuit is collected as the load interference quantity; The ambient temperature around the power device and the voltage multiplier cylinder is collected as the thermal state quantity.

4. The fuzzy PID stable control method for neutron tube accelerating voltage according to claim 3, characterized in that, The real-time current flowing through the target circuit is collected by connecting a sampling resistor in series in the target circuit or by using a current sensor.

5. The fuzzy PID stable control method for neutron tube accelerating voltage according to claim 1, characterized in that, Dynamic impedance feedforward compensation based on the physical inverse model of a voltage multiplier circuit includes: The temperature-corrected equivalent load factor is calculated based on the following formula: ； In the formula, The equivalent loading factor, The load interference amount, The inverter drive frequency, This is the nominal value of the voltage multiplier capacitor. The thermal decay coefficient of the capacitor. The thermal state quantity is denoted as .

6. The fuzzy PID stable control method for neutron tube accelerating voltage according to claim 5, characterized in that, Dynamic impedance feedforward compensation based on the physical inverse model of a voltage multiplier circuit also includes: The feedforward compensation duty cycle is calculated according to the following formula: ； In the formula, For feedforward compensation duty cycle, The geometric constant of the voltage multiplier series is . This is the normalized gain coefficient.

7. The fuzzy PID stable control method for neutron tube accelerating voltage according to claim 1, characterized in that, The dynamic proportional gain satisfies the following relationship: ； In the formula, For dynamic proportional gain, To maintain the gain in steady state, For transient burst gain, This is the variable structure sensitivity factor. The voltage tracking error is given, and , For the set target voltage, This refers to the high-voltage feedback quantity.

8. The fuzzy PID stable control method for neutron tube accelerating voltage according to claim 7, characterized in that, The feedback control output satisfies the following relationship: ； In the formula, For feedback control output, This is the integral gain coefficient.

9. The fuzzy PID stable control method for neutron tube accelerating voltage according to claim 1, characterized in that, The output of dual-channel vector synthesis satisfies the following relationship: ； In the formula, To determine the PWM duty cycle that will ultimately be output to the inverter full-bridge, For the amplitude limiting function, The feedforward compensation duty cycle is... This is the feedback control output.

10. A fuzzy PID stable control system for neutron tube accelerating voltage, characterized in that, include: The system includes a memory and a processor, wherein the memory stores computer program instructions that, when executed by the processor, implement the fuzzy PID stable control method for the neutron tube accelerating voltage as described in any one of claims 1-9.

Images