An optical cavity constant pressure control system based on pump-valve coordination and water wading sensing and its active protection method

Abstract

This invention relates to a constant pressure control system for an optical cavity based on pump-valve coordination and water immersion sensing, and its active protection method. Belonging to the fields of precision instrument control and industrial automation, the system integrates an electronically controlled proportional valve and an air pump through a pump-valve coordinated pressure control unit. It dynamically adjusts the intake and exhaust flow rates based on an adaptive PID algorithm and a coordinated allocation function, achieving high-precision, rapid, bidirectional adjustment of the optical cavity pressure. Simultaneously, it employs a dual environmental sensing unit including visual perception and capacitive liquid level sensing, which executes constant pressure control and active anti-intrusion monitoring in parallel through an intelligent coordinated control unit. When visual detection identifies a risk of water intrusion, the system proactively limits the pump speed; when the capacitive sensor detects liquid water intrusion, it immediately executes an emergency shutdown. This solves the problem of difficulty in coordinating pressure control and safety assurance in traditional technologies, achieving a unified approach of long-term precise constant pressure and active contamination protection, significantly improving the system's reliability, safety, and intelligence level under complex operating conditions.

Description

Technical Field

[0001] This invention relates to the field of precision instrument control and industrial automation technology, specifically to an optical cavity constant pressure control system and its active protection method based on pump-valve coordination and water immersion sensing. Background Technology

[0002] In high-precision optical measurement and spectral analysis applications, the stability of the internal environment of an optical cavity is a core factor determining its measurement accuracy and equipment reliability. Among these, the precise constancy of the internal pressure and the cleanliness and dryness of the surrounding medium are two crucial and interrelated indicators. Even minute pressure fluctuations can directly introduce significant phase noise or frequency drift, leading to increased measurement errors. Conversely, the intrusion of moisture or even liquid water from the external environment can contaminate, corrode, or condense high-precision optical mirrors, causing a surge in optical losses, a decrease in the cavity's quality factor (Q value), and even irreversible permanent performance degradation and equipment damage.

[0003] Currently, in the field of optical cavity environmental control, pressure control and environmental protection are generally treated as two relatively independent issues, which has obvious technical limitations: In pressure control, mainstream solutions often employ simple PID control loops based on a single actuator. For example, pressure is stabilized solely by adjusting the electronically controlled proportional valve at the inlet, or by controlling the pumping speed of the vacuum pump at the outlet. This type of single-degree-of-freedom control architecture has inherent drawbacks: slow response and unidirectional adjustment (pressure can only be changed by increasing or decreasing a single flow rate). When faced with load changes or external disturbances, the system struggles to achieve rapid and precise bidirectional pressure regulation, easily exhibiting overshoot, steady-state error, or continuous oscillation, failing to meet the stringent requirements of long-term pressure stability for ultra-high precision applications.

[0004] In terms of environmental protection, existing technologies mainly rely on passive protection methods such as physical filtration (e.g., desiccants, particulate filters). These methods lack the ability to provide early warning of macroscopic environmental risks around the air intake source (e.g., ground water accumulation, or high-humidity air being drawn in due to coolant leaks). More importantly, they are completely incapable of responding to emergencies where liquid water has entered the pipeline through the air intake, lacking an active intervention mechanism that can detect intrusion within milliseconds and immediately implement safety measures to block it.

[0005] The functional separation between pressure control and environmental protection schemes mentioned above often leads to imbalances in existing systems when faced with complex or sudden operating conditions. For example, increasing the intake air volume to maintain pressure may unintentionally draw more moisture or even droplets from the environment into the cavity; conversely, conservative operation for safety reasons may sacrifice the dynamic performance of pressure control. This inherent contradiction between "constant pressure" and "cleanliness" restricts the reliability of high-precision equipment in a wider range of applications.

[0006] Therefore, there is an urgent need in this field for an innovative integrated control solution that not only needs to break through the performance bottleneck of traditional single-actuator pressure control through a more advanced control architecture to achieve high-precision and high-stability control of optical cavity pressure, but also needs to deeply integrate proactive environmental risk perception and intelligent intervention capabilities. Summary of the Invention

[0007] To address the problems of existing technologies, this invention provides an optical cavity constant pressure control system and its active protection method based on pump-valve coordination and water immersion sensing.

[0008] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution: Firstly, a constant pressure control system based on pump-valve coordination and water immersion sensing, including a pump-valve coordinated pressure control unit, a dual environmental sensing unit, a distributed intelligent coordinated control unit, and a self-tuning and adaptive PID module. The pump-valve coordinated pressure control unit includes a sealed optical cavity, an electronically controlled proportional valve connected to the air inlet of the optical cavity via a pipeline, an air pump connected to the air outlet of the optical cavity via a pipeline, and a high-precision pressure sensor for monitoring the pressure inside the cavity; the detection end of the pressure sensor is connected to the inside of the optical cavity. The dual environmental sensing unit includes a visual sensing module for macro-environmental early warning and a capacitive liquid level sensing module for real-time detection of micro-intrusion; the visual sensing module is deployed in the air intake area of ​​the equipment to collect images of the surrounding environment; the capacitive liquid level sensing module is set on the outer wall of the air intake pipe to detect sudden changes in the dielectric constant of the medium in the pipe. The distributed intelligent collaborative control unit includes an MCU main control board and a Linux main control board. The Linux main control board is communicatively connected to the visual perception module and integrates a target detection neural network for analyzing image data and identifying liquid water risk sources. The MCU main control board is electrically connected to the pressure sensor, the capacitive liquid level sensing module, the electronically controlled proportional valve, and the air pump. The MCU main control board is configured to run collaborative control decision logic, which includes pump-valve collaborative constant pressure control main logic based on pressure deviation, and active anti-intrusion intervention logic triggered by risk warning signals sent by the Linux main control board and / or intrusion detection signals from the capacitive liquid level sensing module. The MCU main control board and the Linux main control board interact with each other via a communication bus. The self-tuning and adaptive PID module is integrated into the MCU main control board. It is used to automatically tune the PID control parameters of the pump-valve coordinated pressure control unit during system initialization and to fine-tune the parameters during operation.

[0009] In one specific implementation of the first aspect, the proactive intrusion prevention intervention logic specifically includes: Level 1 intervention: When the Linux main control board detects a liquid water risk source within the preset range of the air inlet and issues a warning signal, the MCU main control board forcibly limits the current and target air pumping speed of the air pump to a preset safe speed range; Secondary intervention: When the capacitive liquid level sensing module detects a sudden change in the pipeline capacitance value exceeding the safety threshold, the MCU main control board immediately issues the highest priority interrupt command to stop the air pump and limit the opening degree of the electronically controlled proportional valve.

[0010] In one specific implementation of the first aspect, the pump-valve coordinated constant pressure control main logic follows the following strategy: When the cavity pressure is lower than the target value, the opening of the electronically controlled proportional valve is increased first to increase the air intake. If the valve opening reaches the upper limit but the pressure is still insufficient, the speed of the air pump is reduced simultaneously. When the cavity pressure is higher than the target value, the speed of the air pump is increased first to strengthen the pumping. If the pump speed reaches the upper limit but the pressure is still too high, the opening of the electronically controlled proportional valve is reduced simultaneously. During the steady-state maintenance phase, the electronically controlled proportional valve and the air pump are adjusted in a complementary manner based on minute changes in the pressure deviation. The control method is based on the following mathematical model: The gas dynamics equation for the cavity is: in, The pressure inside the cavity, This refers to the intake flow rate of the proportional valve. This refers to the air pump's pumping flow rate. and These are system constants; Controller output control quantity according to Control law calculation: in, ; The control output is distributed to the proportional valve and the air pump by a cooperative allocation function: in, The coordination coefficient is adaptively adjusted based on the pressure change rate. .

[0011] Secondly, a protection method for an optical cavity constant pressure control system based on pump-valve coordination and water immersion sensing, the method comprising the following steps: Step S1: System initialization and parameter self-tuning: After the system is powered on, a hardware self-test is performed, followed by the self-tuning process. By analyzing the pressure response curve, a set of initial PID control parameters for the pump-valve coordinated pressure control unit are obtained. Step S2: Parallel operation of constant pressure control and risk monitoring: The system enters the normal operation phase, and the pump and valve coordinated constant pressure control task of the main thread and the active anti-intrusion monitoring task of the background thread are executed in parallel. Step S3: Environmental risk identification and classification: Active intrusion prevention monitoring continues. Visual risk assessment: Environmental images are collected by the visual perception module and analyzed by the target detection neural network of the Linux main control board. If liquid water risk sources are identified in multiple consecutive frames and the confidence level exceeds the threshold, it is determined that there is a risk of water ingress and a first-level warning signal is triggered. Capacitive intrusion detection: The capacitance value of the pipeline is monitored in real time by a capacitive liquid level sensing module. If a sudden change in the capacitance value exceeding the safety threshold is detected and continues, it is determined that liquid water has intruded into the pipeline, triggering a secondary alarm signal. Step S4: Execution of graded safety intervention: The MCU main control board executes graded intervention based on the received signals; When a Level 1 warning signal is received, Level 1 intervention is executed: the speed of the air pump is forcibly limited to a preset safe speed range, and the PID parameters are adjusted to enhance system stability; When a level 2 alarm signal is received, level 2 intervention is executed: the highest priority interrupt command is immediately issued to stop the air pump, limit the opening of the electronically controlled proportional valve, and trigger an audible and visual alarm. Step S5: Status recovery: When the triggering conditions of the first-level alarm disappear and are confirmed, the system gradually releases the limitation on the speed of the air pump and restores normal constant pressure control; when the second-level alarm is handled manually on-site and the risk is confirmed to be eliminated, the system is re-initialized and enters normal working mode through a reset operation.

[0012] The beneficial effects of this invention are as follows: 1. By constructing a pump-valve coordinated pressure control unit and introducing an adaptive coordinated allocation algorithm, this invention establishes a new generation of optical cavity constant pressure control paradigm. Traditional single-actuator control schemes, due to inherent response delays and unidirectional adjustment, struggle to achieve rapid and precise bidirectional pressure regulation. This invention places the electronically controlled proportional valve (inlet) and the air pump (exhaust) within the same high-performance control closed loop, and uses a coordinated coefficient α(t) based on the pressure change rate dP(t) / dt for adaptive adjustment of the control quantity. This enables the system to intelligently calculate and execute composite commands to the valve and pump based on the real-time pressure deviation e(t) and its changing trend. For example, when rapid pressure increase is required, the system can simultaneously execute a composite action of "increasing valve opening" and "decreasing pump speed," significantly accelerating the dynamic response speed. During the steady-state maintenance phase, complementary micro-adjustments to both effectively suppress overshoot and oscillation. This "accelerator and brake coordination" control strategy fundamentally improves the dynamic performance and robustness of the control loop, achieving steady-state accuracy (such as better than ±0.5Pa) and disturbance rejection capability that is difficult for traditional architectures to achieve.

[0013] 2. By integrating dual environmental sensing units and an intelligent collaborative decision-making mechanism, this invention constructs a proactive safety defense line of "forward-looking early warning + real-time blocking," achieving intelligent unification of core functions and high-strength safety protection. The visual sensing module, based on a deep learning model, continuously scans the macroscopic environment for risks. It can trigger a first-level warning before high-humidity air is drawn in (e.g., when a puddle is detected on the ground), prompting the system to proactively limit the air pump speed to a safe range and adjust PID parameters to smoothly avoid risks. The capacitive liquid level sensing module provides millisecond-level micro-intrusion detection. Once liquid water enters the pipeline causing a sudden change in dielectric constant, a second-level alarm is immediately triggered, driving the system to execute the highest-priority emergency blocking (e.g., immediately stopping the pump and closing the valve). The intelligent collaborative control unit, as the decision-making core, allows the constant pressure control thread and the safety intervention thread to operate in parallel. It can intelligently arbitrate when risks occur, maintaining pressure control as much as possible while ensuring absolute safety. This upgrades environmental protection from passive filtration to proactive sensing and intelligent intervention, significantly reducing the risk of performance degradation of the optical cavity due to contamination and ensuring the long-term operational reliability and data consistency of the system. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the system framework of the present invention.

[0015] Figure 2 This is a schematic diagram of the system framework of an embodiment of the present invention.

[0016] Figures 1 to 2The components include: 1. Capacitive liquid level sensing module; 2. Visual sensing module; 3. Electro-controlled proportional valve; 4. Pressure sensor; 5. Optical cavity; 6. Air pump; 7. MCU main control board; 8. Linux main control board. Detailed Implementation

[0017] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0018] like Figures 1 to 2 The invention illustrates an optical cavity constant pressure control system based on pump-valve coordination and water immersion sensing, and its active protection method.

[0019] The specific hardware configuration and control process of the system This embodiment describes in detail the hardware connection, software configuration, and complete workflow of the system described in this invention.

[0020] 1. System hardware composition and connection relationships refer to Figure 1 The system hardware architecture diagram shown below illustrates the specific hardware configuration of this system: Pump and valve coordinated pressure control unit: Sealed optical cavity 5: This is the core component to be controlled. Optical cavity 5 has two interfaces: an air inlet and an air outlet.

[0021] Electro-controlled proportional valve 3: Preferably a piezoelectrically driven proportional valve with high response speed. Its inlet end is connected to the clean air source through a Φ6mm PTFE pipe, and its outlet end is connected to the inlet port of the optical cavity 5 through a Φ6mm pipe. The control signal of the proportional valve is a 0-5V analog voltage, output by the DA conversion channel of the MCU main control board 7. The voltage value is linearly related to the valve opening.

[0022] Air pump 6: A low-vibration, oil-free diaphragm pump is selected. Its inlet is connected to the air outlet of the optical cavity 5 through a Φ6mm pipe, and its outlet is open to the atmosphere. The speed of air pump 6 is controlled by the PWM signal output by the MCU main control board 7, and the duty cycle is proportional to the speed.

[0023] High-precision pressure sensor 4: An absolute pressure sensor with a standard atmospheric pressure range and an accuracy better than 0.1%FS is selected. Its sensing end is installed on the air inlet pipe between the electronically controlled proportional valve 3 and the optical cavity 5 via a three-way connector to achieve real-time monitoring of the cavity pressure. Pressure sensor 4 outputs a digital signal and communicates with the MCU main control board 7 via an SPI interface.

[0024] Dual environmental sensing unit: Visual Perception Module 2: This module uses an industrial-grade network camera with a resolution of 1920x1080. It is mounted on the bottom of the equipment rack (on top of the vehicle, not the bottom) with a bracket, and the lens is tilted downwards at approximately 30 degrees to ensure a clear view covering a 1-meter radius area around the air intake. The camera is connected to the gigabit Ethernet port of the Linux main control board 8 via an Ethernet cable.

[0025] Capacitive liquid level sensing module 1: This module uses a ring-shaped capacitive sensor, directly fitted into the air inlet pipe. It provides digital switching signals to the MCU main control board 7 via the I / O port, and can also output analog signals via the ADC channel to monitor capacitance changes. Its sensitivity is adjustable; when the medium in the pipe changes from air to liquid water, the capacitance value will change abruptly by more than 50% of the reference value, triggering an alarm.

[0026] Distributed intelligent collaborative control unit: MCU Main Control Board 7: Employs a high-performance 32-bit microcontroller based on the ARM Cortex-M7 core, running the FreeRTOS real-time operating system. This main control board is responsible for: Data from pressure sensor 4 is read via the SPI interface.

[0027] The analog signal of the capacitive liquid level sensing module 1 is read through the ADC interface, and its digital alarm signal is received through the interrupt pin.

[0028] The proportional valve 3 is controlled via the DA channel output.

[0029] The speed of air pump 6 is controlled by the PWM channel.

[0030] Bidirectional data communication is performed with the Linux main control board 8 via UART serial port (baud rate 115200).

[0031] The core algorithms include the main logic of pump-valve coordinated constant pressure control, the active anti-intrusion intervention logic, and the self-tuning and adaptive PID module.

[0032] Linux Main Control Board 8: This board uses an embedded computing module equipped with a quad-core ARM processor and runs the Linux operating system. This main control board is responsible for: The video stream is received from the visual perception module 2 via the Ethernet port.

[0033] Run the visual analysis module, which integrates a target detection neural network based on the YOLOv5s model (trained on a custom dataset and specifically designed to identify puddles, splashes, etc.) to perform real-time analysis of the video stream (processing speed of approximately 15 frames per second).

[0034] The identification results (including risk source category and confidence level) are sent to the MCU main control board 7 via UART serial port.

[0035] 2. System Workflow After the system is powered on, it will operate in the following stages: Phase 1: System Startup and Parameter Self-Tuning Hardware self-test: The MCU main control board 7 sequentially checks the communication and response of the pressure sensor 4, capacitive liquid level sensing module 1, electronically controlled proportional valve 3, and air pump 6 to ensure they are normal. The Linux main control board 8 checks whether the camera 2 is online. If any component malfunctions, an audible and visual alarm is triggered, and the system waits for further processing.

[0036] Parameter self-tuning: After passing the self-test, the system enters the self-tuning process. The MCU main control board 7 controls the air pump 6 and the proportional valve 3 to stabilize the pressure of the optical cavity 5 at the commonly used operating point (500 mbar). Then, a small step change in the opening of the proportional valve 3 (e.g., an increase of 5%) is applied, and the response curve of the pressure sensor is recorded. Based on the characteristic parameters such as the rise time and overshoot of the response curve, a set of initial PID parameters (Kp0, Ki0, Kd0) is calculated using the Ziegler-Nichols method or a similar engineering tuning method. During this stage, the coordination coefficient α(t) is fixed at 0.5.

[0037] Phase Two: Normal Operation and Parallel Task Execution After initialization, the system enters the normal operation phase, and the MCU main control board 7 and the Linux main control board 8 execute the following tasks in parallel: High-priority constant voltage control task of MCU main control board 7 (period 1ms): Pressure sampling and error calculation: Read the current pressure value P(t) and calculate the error e(t) between it and the target pressure Pset.

[0038] PID calculation: based on the adaptive PID algorithm Calculate control quantity . , , It will make online fine-tuning based on the magnitude of the error and the stability of the system.

[0039] Pump and valve coordinated output: Calculate the current pressure change rate. According to the formula Calculate the coordination coefficient, where the empirical parameter λ is set to 10. Then, allocate the control quantity as valve flow rate. and pump flow This is then converted into specific valve control voltage and air pump PWM duty cycle output.

[0040] The visual analysis task on the Linux mainboard 8 (cycle approximately 66ms): continuously acquires camera images and performs real-time analysis using the YOLO model.

[0041] Active intrusion prevention monitoring task of MCU main control board 7 (cycle 10ms): continuously monitor the visual warning signal from Linux main control board 8 and the alarm signal from capacitive liquid level sensing module 1.

[0042] Phase Three: Implementation of Proactive Protective Interventions Triggering and execution of Level 1 intervention (prospective early warning and deceleration): Triggering conditions: When the Linux main control board 8 identifies risk sources such as "puddles" or "overflow" for 5 consecutive frames, and the confidence level of each frame is higher than 0.8, it sends a level 1 warning signal (including risk type) to the MCU main control board 7 via serial port.

[0043] Action executed: After receiving the signal, the MCU main control board 7 forcibly limits the upper limit of the air pump 6's speed to 40% of the rated speed. At the same time, to avoid control instability caused by possible changes in intake air humidity, the integral coefficient Ki(t) in the PID algorithm is temporarily adjusted to 70% of the normal value. Pressure control is mainly maintained by finely adjusting the electronically controlled proportional valve 3.

[0044] Triggering and execution of Level 2 intervention (real-time intrusion emergency blocking): Triggering condition: When the capacitive liquid level sensing module 1 detects that the capacitance value increases by more than 50% of the reference value within 3 sampling cycles (30ms), it immediately sends a secondary alarm signal to the MCU main control board 7 through the hardware interrupt line.

[0045] Action executed: Upon receiving the interrupt, the MCU main control board 7 immediately stops outputting PWM signals to the air pump 6, causing it to stop; simultaneously, it sets the control voltage of the electronically controlled proportional valve 3 to 0, closing it. It also triggers the audible and visual alarm, displaying emergency alarm information on the human-machine interface. The system enters a safety lockout state, and all automatic controls are suspended.

[0046] Phase 4: Condition Recovery Level 1 Warning Recovery: If the visual analysis module does not report any risk source for 15 consecutive seconds, and this is confirmed by the operator on the interface, the MCU main control board 7 will deactivate the Level 1 warning. Subsequently, the upper limit of the air pump speed will be gradually increased by 5% of the rated speed every 2 seconds until it is fully restored to the normal speed range. The PID parameters will also be restored synchronously.

[0047] Level 2 alarm recovery: The operator must inspect the air intake pipe on-site to confirm that it is free of water and dry, then press and hold the reset button for 5 seconds. The system will then re-execute the hardware self-test and parameter self-tuning process before entering normal operating mode.

[0048] Example 1: System Hardware Architecture and Signal Connections refer to Figure 2 The system hardware architecture diagram shown illustrates that this system adopts a distributed control architecture, and the specific hardware connections are as follows: Perception and Execution Layer: The visual perception module, or camera, is used to capture environmental images. It connects to the upper-layer Linux main control board via a network interface (such as Ethernet) to transmit video stream data to the Linux main control board for analysis.

[0049] The micro-intrusion detection module, also known as a capacitive liquid level sensor, is installed on the outer wall of the air inlet pipe. It is directly connected to the analog / digital input port of the MCU main control board via a signal line, and reports the detected capacitance signal (reflecting changes in the medium inside the pipe) in real time.

[0050] An electronically controlled proportional valve is connected in series between the clean air source and the air inlet of the optical cavity. Its opening degree is controlled by an analog voltage signal from the MCU main control board to adjust the air intake flow.

[0051] The high-precision pressure sensor's sensing end is connected to the inside of the optical cavity to monitor the pressure within the cavity. It connects to the MCU main control board via a digital communication interface (such as SPI or I2C) to provide real-time pressure feedback.

[0052] The pump is connected to the air outlet of the optical cavity, and its rotation speed is controlled by a PWM (pulse width modulation) signal from the MCU main control board to adjust the air flow rate.

[0053] Control and Decision-Making Level: As an advanced computing unit, the Linux mainboard is primarily responsible for handling visual analysis tasks that require high computing power. It receives data from cameras via the network, runs embedded target detection neural network algorithms, identifies liquid water risk sources, and sends the identification results (such as warning signals) to the MCU mainboard via a serial communication link (such as UART).

[0054] The MCU main control board, as the real-time control core, undertakes the main closed-loop control and safety logic functions. It receives warning information from the Linux main control board via serial port and directly acquires signals from the level and pressure sensors. Based on this information, the MCU main control board executes a PID control algorithm to generate coordinated control commands, driving the proportional valve and pump to achieve precise and stable pressure control within the optical cavity, and performs tiered safety interventions when a risk of water ingress is detected.

[0055] Workflow: During system operation, the MCU main control board forms a fast-response pressure control closed loop, simultaneously receiving visual warnings from the Linux main control board and direct alarms from the liquid level sensor. When there is no risk, the system focuses on precise pressure regulation; when a risk occurs, the MCU main control board dynamically adjusts or cuts off the actions of the actuators (pumps, valves) according to the preset two-level intervention logic, achieving active protection.

[0056] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A constant pressure optical cavity control system based on pump-valve coordination and wading sensing, characterized in that, This includes a pump-valve coordinated pressure control unit, a dual environmental sensing unit, a distributed intelligent coordinated control unit, and a self-tuning and adaptive PID module; The pump-valve coordinated pressure control unit includes a sealed optical cavity (5), an electronically controlled proportional valve (3) connected to the air inlet of the optical cavity (5) via a pipeline, an air pump (6) connected to the air outlet of the optical cavity (5) via a pipeline, and a high-precision pressure sensor (4) for monitoring the pressure inside the cavity; the detection end of the pressure sensor (4) is connected to the inside of the optical cavity (5). The dual environmental sensing unit includes a visual sensing module (2) for macro-environmental early warning and a capacitive liquid level sensing module (1) for real-time detection of micro-intrusion; the visual sensing module (2) is deployed in the air intake area of ​​the equipment to collect images of the surrounding environment; the capacitive liquid level sensing module (1) is set on the outer wall of the air intake pipe to detect sudden changes in the dielectric constant of the medium in the pipe. The distributed intelligent collaborative control unit includes an MCU main control board (7) and a Linux main control board (8); the Linux main control board (8) is communicatively connected to the visual perception module (2) and integrates a target detection neural network for analyzing image data and identifying liquid water risk sources; the MCU main control board (7) is electrically connected to the pressure sensor (4), the capacitive liquid level sensing module (1), the electronically controlled proportional valve (3), and the air pump (6); the MCU main control board (7) is configured to run collaborative control decision logic, which includes pump-valve collaborative constant pressure control main logic based on pressure deviation, and active anti-intrusion intervention logic triggered by risk warning signals sent by the Linux main control board (8) and / or intrusion detection signals from the capacitive liquid level sensing module (1); the MCU main control board (7) and the Linux main control board (8) interact with each other via a communication bus; The self-tuning and adaptive PID module is integrated in the MCU main control board (7) and is used to automatically tune the PID control parameters of the pump-valve coordinated pressure control unit during system initialization and to fine-tune the parameters during operation.

2. The optical cavity constant pressure control system based on pump-valve coordination and wading sensing according to claim 1, characterized in that: The proactive intrusion prevention intervention logic specifically includes: Level 1 intervention: When the Linux main control board (8) detects a liquid water risk source within the preset range of the air inlet and issues a warning signal, the MCU main control board (7) forcibly limits the current and target pumping speed of the air pump (6) to a preset safe speed range; Secondary intervention: When the capacitive liquid level sensing module (1) detects a sudden change in the pipeline capacitance value exceeding the safety threshold, the MCU main control board (7) immediately issues the highest priority interrupt command to stop the operation of the air pump (6) and limit the opening degree of the electronically controlled proportional valve (3).

3. The optical cavity constant pressure control system based on pump-valve coordination and wading sensing according to claim 1, characterized in that: The pump-valve coordinated constant pressure control main logic follows the following strategy: When the cavity pressure is lower than the target value, the opening of the electronically controlled proportional valve (3) is increased first to increase the air intake. If the valve opening reaches the upper limit but the pressure is still insufficient, the speed of the air pump (6) is reduced simultaneously. When the cavity pressure is higher than the target value, the speed of the air pump (6) is increased first to strengthen the pumping. If the pump speed reaches the upper limit but the pressure is still too high, the opening of the electronically controlled proportional valve (3) is reduced simultaneously. During the steady-state maintenance phase, the electronically controlled proportional valve (3) and the air pump (6) are adjusted in a complementary manner according to the slight changes in pressure deviation; The control method is based on the following mathematical model: The gas dynamics equation for the cavity is: in, The pressure inside the cavity, This refers to the intake flow rate of the proportional valve. This refers to the air pump's pumping flow rate. and These are system constants; Controller output control quantity according to Control law calculation: in, ; The control output is distributed to the proportional valve and the air pump by a cooperative allocation function: in, The coordination coefficient is adaptively adjusted based on the pressure change rate. 。 4. A protection method for an optical cavity constant pressure control system based on pump-valve coordination and wading sensing, characterized in that: The method includes the following steps: Step S1: System initialization and parameter self-tuning: After the system is powered on, a hardware self-test is performed, followed by the self-tuning process. By analyzing the pressure response curve, a set of initial PID control parameters for the pump-valve coordinated pressure control unit are obtained. Step S2: Parallel operation of constant pressure control and risk monitoring: The system enters the normal operation phase, and the pump and valve coordinated constant pressure control task of the main thread and the active anti-intrusion monitoring task of the background thread are executed in parallel. Step S3: Environmental risk identification and classification: Active intrusion prevention monitoring continues. Visual risk assessment: The visual perception module (2) collects environmental images, and the Linux main control board (8) uses a target detection neural network to analyze them. If multiple consecutive frames identify liquid water risk sources and the confidence level exceeds the threshold, it is determined that there is a risk of water ingress and a first-level warning signal is triggered. Capacitive intrusion judgment: The capacitance value of the pipeline is monitored in real time by the capacitive liquid level sensing module (1). If the capacitance value is detected to exceed the safety threshold and continues to do so, it is determined that liquid water has intruded into the pipeline and a secondary alarm signal is triggered. Step S4: Execution of graded safety intervention: The MCU main control board (7) executes graded intervention based on the received signals; When a Level 1 warning signal is received, Level 1 intervention is performed: the speed of the air pump (6) is forcibly limited to the preset safe speed range, and the PID parameters are adjusted to enhance system stability; When a level 2 alarm signal is received, level 2 intervention is executed: the highest priority interrupt command is immediately issued to stop the operation of the air pump (6), and the opening degree of the electronically controlled proportional valve (3) is limited, while triggering an audible and visual alarm; Step S5: Status recovery: When the triggering conditions of the first-level alarm disappear and are confirmed, the system gradually releases the limitation on the speed of the air pump and restores normal constant pressure control; when the second-level alarm is handled manually on-site and the risk is confirmed to be eliminated, the system is re-initialized and enters normal working mode through a reset operation.

Images