Lead bismuth valve control device and control method

Abstract

The application relates to the technical field of valve control and discloses a lead-bismuth valve control device and a control method, the device comprising a moving unit and a control unit, the moving unit being detachably connected with a valve core through a connecting sleeve; the control unit comprising a data acquisition module, a data storage module and a controller, the controller estimating the opening and closing state of the valve through a mapping model based on the real-time driving device movement data and the stored historical correlation data, and outputting a control signal to drive the valve to move; after the valve reaches the target position, the moving unit can be automatically detached from the valve core, and dynamic correction and model updating can be carried out based on multiple data; the application also correspondingly provides a control method, and through the comprehensive scheme of redundant position monitoring, dynamic torque control and closed-loop feedback calibration, the problems of insufficient control precision, poor dynamic adaptability, large driving interference and low operation and maintenance efficiency of the traditional valve are solved.

Description

Technical Field

[0001] This invention relates to the field of valve control technology, specifically to a lead-bismuth valve control device and control method. Background Technology

[0002] In industries such as nuclear energy, chemical engineering, and pharmaceuticals, high-precision flow and pressure control of special media such as liquid lead and bismuth is often required. These media typically exhibit characteristics such as high temperature, high pressure, and strong corrosiveness, placing extremely stringent demands on the accuracy, reliability, and lifespan of valves and their control systems.

[0003] Existing valve control devices mostly employ simple closed-loop control, which has the following significant drawbacks: First, insufficient control accuracy: it is difficult to eliminate errors caused by transmission chain backlash, wear, and thermal deformation, making it impossible to achieve angular control accuracy at the ±0.1° level; Second, poor dynamic adaptability: when the viscosity of the medium or the system load changes abruptly, traditional PID control responds slowly, easily generating overshoot or oscillation, leading to control instability; Third, drive interference problems: the vibration and torque fluctuations of the drive motor are continuously transmitted to the valve, affecting the accurate measurement of the valve's state, especially in situations where the valve needs to maintain a stable opening; Fourth, low level of intelligence: lacking self-learning and self-correction capabilities, it cannot adaptively adjust parameters according to equipment wear and changes in operating conditions, and maintenance relies on manual experience, resulting in low efficiency and insufficient reliability.

[0004] Therefore, it is necessary to design a lead-bismuth valve control device to improve the above-mentioned problems. Summary of the Invention

[0005] To address the problems of the prior art, the present invention provides a lead-bismuth valve control device, comprising an actuating unit and a control unit;

[0006] The actuation unit includes a valve driving device and a transmission device;

[0007] The transmission device includes a transmission shaft, and the output end of the transmission shaft is connected to a connecting sleeve.

[0008] The output end of the valve drive device is connected to the input end of the drive shaft to drive the drive shaft; the output end of the drive shaft is detachably connected to the valve core of the lead-bismuth valve through a connecting sleeve.

[0009] The control unit includes:

[0010] The data acquisition module is used to acquire motion data of the valve drive device, motion data of the transmission device, and dynamic contact data between the transmission shaft and the connecting sleeve;

[0011] The data storage module stores at least historical correlation data between the motion relationship between the valve drive device and the transmission device, and between the valve opening degree and the drive angle of the valve drive device.

[0012] The controller is connected to the valve drive device, the data acquisition module, and the data storage module, respectively. The controller is configured to: calculate and obtain the current estimated opening and closing state of the valve based on the real-time acquired motion data of the valve drive device and the historical correlation data through a pre-established mapping model; output a control signal to the valve drive device according to the difference between the estimated opening and closing state and the target command, thereby driving the transmission device and the valve core to move; output a disengagement command after the valve reaches the target position, controlling the action unit to disengage from the valve core; and dynamically correct and update the historical correlation data and the mapping model based on the multi-data acquired in real time by the data acquisition module.

[0013] Furthermore, the valve driving device is a rotary driving device, and the motion data of the valve driving device includes at least the rotation angle, rotational speed, and output torque; the motion data of the transmission device includes at least the rotation angle and vibration data.

[0014] The motion data acquisition method for the transmission device and valve drive device is as follows:

[0015] If the transmission device and the valve driving device move synchronously, the motion data of the transmission device and the motion data of the valve driving device are consistent.

[0016] If the transmission device and the valve drive device move asynchronously, the motion data of the transmission device is obtained based on the correlation between the asynchronous movements of the valve drive device and the transmission device.

[0017] Furthermore, the dynamic contact data between the drive shaft and the connecting sleeve includes at least the contact pressure, pressure change gradient, and effective contact time.

[0018] Furthermore, the controller is configured to acquire the opening / closing state of the valve through the following steps:

[0019] Obtain the real-time drive angle of the valve drive device;

[0020] The historical correlation data is retrieved from the data storage module. The historical correlation data records the mapping relationship between the drive angle and the actual opening degree of the valve.

[0021] The real-time drive angle is input into the mapping relationship to calculate the estimated opening and closing state of the current valve;

[0022] Based on the distribution of the real-time driving angle in the mapping relationship data sequence, the confidence interval for the estimated opening and closing state is determined;

[0023] If the confidence interval is higher than a preset threshold, the estimated opening / closing state is determined to be valid; if it is lower than the preset threshold, a real-time correction procedure is triggered, and the mapping relationship is updated.

[0024] Furthermore, the control unit also includes a torque dynamic control module; the torque dynamic control module integrates model predictive control, sliding mode control and fuzzy-neural network fusion control algorithms, and is configured to execute a three-stage torque control strategy of "preloading-buffering-holding" to achieve a torque control accuracy of ±0.5 N·m.

[0025] Furthermore, the control unit also integrates a backup power protection module, which is composed of a backup lithium battery and a mechanical energy storage spring. The backup lithium battery and the mechanical energy storage spring are connected through a power coupler. The backup power protection module is also equipped with a power monitoring unit and a fast switching circuit. The power monitoring unit collects the main power voltage signal in real time. When the main power voltage is detected to be lower than the preset voltage value, the fast switching circuit is immediately triggered to provide emergency power in the event of main power failure, ensuring that the lead-bismuth valve completes the emergency closing action and the disengagement action of the actuating unit from the valve core.

[0026] Furthermore, the control unit also integrates a redundant position monitoring system, which consists of a laser displacement sensor, a Hall encoder, and a rotary transformer. This system is used to collect and feedback the position information of the lead-bismuth valve core in real time, providing redundant and reliable data source support for the valve position closed-loop control. The laser displacement sensor's measuring end faces the valve core body to acquire the absolute position reference signal of the valve core. The Hall encoder is connected to the input end of the transmission device via a flange structure to capture dynamic position change data during valve actuation and transmission in real time. The rotary transformer is mounted on the output end of the transmission device near the connecting sleeve via a sealed mounting base, and is used to output continuous analog position signals under high temperature and strong electromagnetic interference conditions.

[0027] Furthermore, one end of the connecting sleeve is provided with two symmetrically arranged arc-shaped teeth, and the length error of the arc-shaped teeth is controlled within ±0.1mm; the output end of the drive shaft is provided with a tooth groove that meshes with the arc-shaped teeth, one end of the connecting sleeve meshes with the output end of the drive shaft, the meshing surface is coated with a solid lubricating coating, and it has a built-in double sealing structure including a metal sealing ring and an elastic sealing ring, which can withstand a pressure of not less than 10MPa; the other end of the connecting sleeve is rigidly connected to the valve stem through a flange.

[0028] A control method based on the lead-bismuth valve control device includes the following steps:

[0029] Step 1: Monitor external control commands, local manual signals, fault warning signals, and power outage emergency signals in real time;

[0030] Step 2: Based on the real-time motion data of the valve drive device and the stored historical correlation data, calculate the estimated opening and closing state of the valve through a mapping model and evaluate its confidence level;

[0031] Step 3: Based on the difference between the target command and the estimated opening / closing state, output a control signal to drive the valve to move, and achieve closed-loop control through feedback from the redundant position monitoring system;

[0032] Step 4: After confirming that the valve has reached the target position, control the transmission device to rotate to the predetermined angle so that the actuating unit disengages from the valve core;

[0033] Step 5: Based on the real-time collected torque, angle, pressure, and temperature data, the control parameters are dynamically corrected and the historical correlation data model is updated using sliding mode control and extended Kalman filter algorithm.

[0034] Step 6: When a fault or power failure signal is triggered, activate the backup power and perform an emergency reset according to the preset safety strategy;

[0035] Step 7: Record all operational data and upload it to the operation and maintenance platform for health assessment and preventive maintenance decisions.

[0036] Furthermore, in step 4, the predetermined angle of rotation of the transmission device is 60°±0.1°; when re-engaging, if the angle deviation exceeds ±1°, the angle calibration procedure is automatically triggered.

[0037] The beneficial effects of this invention are:

[0038] (1) This invention achieves a breakthrough in accuracy from two dimensions: error source suppression and real-time compensation by combining redundant position monitoring, composite material structure and closed-loop feedback calibration. It achieves an angle control accuracy of ±0.1° and a torque control accuracy of ±0.5 N·m, solving the problem of insufficient control accuracy of traditional devices. Through the fusion of multiple algorithms in the torque dynamic control module and the three-stage strategy of "preload-buffer-hold", it effectively copes with dynamic working conditions such as load changes, solving the problem of slow response and instability of traditional PID control. Through automatic disengagement design, vibration suppression structure and torque smooth control, it eliminates the interference of the drive device on valve status measurement from two dimensions: physical isolation and active suppression. Through dynamic parameter correction, full-process data recording and predictive maintenance, it constructs a self-learning and self-correcting intelligent operation and maintenance system, reducing manual dependence and improving operation and maintenance efficiency and reliability.

[0039] (2) This invention constructs a comprehensive safety guarantee system through backup power guarantee module, fault emergency mechanism and predictive maintenance design, realizes 10ms level emergency power supply switching, significantly improves equipment life and greatly reduces the risk of unexpected shutdown. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the lead-bismuth valve control device of the present invention;

[0041] Figure 2 A schematic diagram showing the connection between the drive shaft and the valve drive device of this invention;

[0042] Figure 3 Schematic diagram of the connection between the connecting sleeve and the valve core in this invention;

[0043] Figure 4 Flowchart of the control method for the lead-bismuth valve control device of the present invention.

[0044] Figure label:

[0045] In the diagram: 1-drive shaft, 2-connecting sleeve. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Please see Figure 1-4 The present invention provides a lead-bismuth valve control device, including an actuation unit and a control unit;

[0048] The actuation unit includes a valve driving device and a transmission device;

[0049] The transmission device includes a transmission shaft 1, and a connecting sleeve 2 is connected to the output end of the transmission shaft 1.

[0050] The output end of the valve drive device is connected to the input end of the drive shaft 1 to drive the drive shaft 1; the output end of the drive shaft 1 is detachably connected to the valve core of the lead-bismuth valve through the connecting sleeve 2.

[0051] The control unit includes:

[0052] The data acquisition module is used to acquire motion data of the valve drive device, motion data of the transmission device, and dynamic contact data between the transmission shaft 1 and the connecting sleeve 2;

[0053] The data storage module stores at least historical correlation data between the motion relationship between the valve drive device and the transmission device, and between the valve opening degree and the drive angle of the valve drive device.

[0054] The controller is connected to the valve drive device, the data acquisition module, and the data storage module, respectively. The controller is configured to: calculate and obtain the current estimated opening and closing state of the valve based on the real-time acquired motion data of the valve drive device and the historical correlation data through a pre-established mapping model; output a control signal to the valve drive device according to the difference between the estimated opening and closing state and the target command, thereby driving the transmission device and the valve core to move; output a disengagement command after the valve reaches the target position, controlling the action unit to disengage from the valve core; and dynamically correct and update the historical correlation data and the mapping model based on the multi-data information acquired in real time by the data acquisition module.

[0055] It should be noted that the dynamic contact data is collected by sensors placed at the contact interface to collect the dynamic contact data between the drive shaft and the connecting sleeve. The sensors at the contact interface are pressure sensors that collect the contact pressure.

[0056] Furthermore, the valve driving device is a rotary driving device, and the motion data of the valve driving device includes at least the rotation angle, rotational speed, and output torque; the motion data of the transmission device includes at least the rotation angle and vibration data.

[0057] The motion data acquisition method for the transmission device and valve drive device is as follows:

[0058] If the transmission device and the valve driving device move synchronously, the motion data of the transmission device and the motion data of the valve driving device are consistent.

[0059] If the transmission device and the valve drive device move asynchronously, the motion data of the transmission device is obtained based on the correlation between the asynchronous movements of the valve drive device and the transmission device.

[0060] Furthermore, the dynamic contact data between the drive shaft 1 and the connecting sleeve 2 includes at least the contact pressure, pressure change gradient, and effective contact time.

[0061] Furthermore, the controller is configured to acquire the opening / closing state of the valve through the following steps:

[0062] Obtain the real-time drive angle of the valve drive device;

[0063] The historical correlation data is retrieved from the data storage module. The historical correlation data records the mapping relationship between the drive angle and the actual opening degree of the valve.

[0064] The real-time drive angle is input into the mapping relationship to calculate the estimated opening and closing state of the current valve;

[0065] Based on the distribution of the real-time driving angle in the mapping relationship data sequence, the confidence interval for the estimated opening and closing state is determined;

[0066] If the confidence interval is higher than a preset threshold (e.g., 95%), the estimated opening and closing state is determined to be valid; if it is lower than the preset threshold, a real-time correction procedure is triggered, and the mapping relationship is updated.

[0067] Furthermore, the control unit also includes a torque dynamic control module; the torque dynamic control module integrates model predictive control, sliding mode control and fuzzy-neural network fusion control algorithms, and is configured to execute a three-stage torque control strategy of "preloading-buffering-holding" to achieve a torque control accuracy of ±0.5 N·m;

[0068] It should be noted that the torque dynamic control module is equipped with a magnetoelastic torque sensor with a sampling frequency of 1kHz and an Ethernet real-time bus (delay ≤1ms). Energy recovery is achieved through a bidirectional DC-DC converter and segmented buffer control, which suppresses torque fluctuations. The torque dynamic control module optimizes torque output throughout the entire process, ensuring smooth, precise operation and adaptability to dynamic loads, ultimately achieving a torque control accuracy of ±0.5 N·m. The specific implementation process is described below:

[0069] The system employs Model Predictive Control (MPC) to establish a model of the motor's torque-current-speed-temperature relationship. Data is collected every 10ms, and the torque trend is predicted for the next 50ms. The Pulse Width Modulation (PWM) duty cycle is adjusted in advance to control the torque within the target range (e.g., 30 N·m ± 2 N·m during engagement). A maximum current of 10A and a maximum torque of 50 N·m are embedded to prevent overshoot. Sliding mode variable structure control achieves precise system control by actively constructing a sliding surface and forcibly switching control quantities, adapting to the real-time requirements of dynamic torque adjustment. The sliding surface is designed as s = ce + g (where e is the torque error, g is the error change rate, s is the sliding surface state value, and c is the adjustment coefficient, which is set to 0.5-2 according to the operating conditions to balance response speed and stability). A discontinuous control law forces the system to move along the sliding surface to cope with motor parameter drift or sudden load changes (such as valve jamming). Boundary layer linear control weakens high-frequency chattering.

[0070] The adaptive control employs Model Reference Adaptive Control (MRAC) to ensure that the actual torque tracks the ideal response curve. The Sequential Turbocharging Controller (STC) uses recursive least squares to identify parameters such as motor resistance online and updates the proportional-integral control (PI) parameters in real time to compensate for the increased resistance caused by high temperatures. The fuzzy-neural network fusion control algorithm achieves control through three steps: fuzzification, rule-based reasoning, and defuzzification. This algorithm is used to quickly respond to nonlinear and uncertain disturbances in torque control (such as valve jamming and load fluctuations caused by sudden temperature changes). The torque error and its rate of change are divided into five fuzzy sets, and coarse adjustments are made according to rules such as "a large positive error and a small positive rate of change result in a large negative control output." The back propagation neural network (BP) optimizes the fuzzy factor online, automatically amplifying the error weight at low temperatures (-50℃) to improve adjustment sensitivity and achieve a control accuracy of ±0.5 N·m. Hardware co-optimization employs a magnetoelastic torque sensor with a 1kHz sampling frequency to capture millisecond-level torque fluctuations, and an Ethernet (EtherCAT) bus for control automation technology reduces data transmission latency to less than 1ms. At the moment of disengagement, excess electrical energy is recovered to the supercapacitor via a bidirectional DC-DC converter. The final 5° of engagement / disengagement stroke adopts an "S-shaped speed curve" to achieve a smooth transition of torque from 20N·m to 0, suppressing impact.

[0071] First preloading stage: Preloading stage – achieving smooth contact and avoiding rigid impact.

[0072] (1) Control objective: To achieve “flexible contact” between the arc-shaped teeth of the drive shaft 1 and the arc-shaped groove of the connecting sleeve 2, eliminate rigid collisions during startup, and prevent tooth surface deformation, cracking, or abnormal noise;

[0073] (2) The control logic and implementation process are as follows:

[0074] Initial torque output: The control module initially outputs a low-amplitude torque, which is usually set to 30%-50% of the final target meshing torque (e.g., 10-15 N·m). The specific value is preset based on historical working condition data.

[0075] Slow-speed smooth loading: The output torque is gradually increased at a low rate controlled within 5 N·m / s.

[0076] High-precision real-time monitoring: During this process, the torque value is monitored in real time by a magnetoelastic torque sensor with a sampling frequency of 1kHz, and high-speed data feedback is achieved by using an EtherCAT real-time bus with a transmission delay of ≤1ms.

[0077] Contact point determination: When the sensor detects a slight step increase in the torque signal, it is determined to be a "tooth surface contact signal", indicating that the tooth profile has been smoothly connected and the preloading stage is completed.

[0078] (3) Applicable scenarios: Low-temperature conditions (e.g., -50℃). Under this environment, the rigidity of the materials of the drive shaft 1 and the connecting sleeve 2 increases while the toughness decreases. The flexible contact strategy in this stage can effectively avoid tooth breakage caused by rigid collision.

[0079] Second buffer phase: Buffer phase – dynamic adjustment to compensate for fluctuations in operating conditions.

[0080] (1) Control objective: After the tooth surface contacts, respond to dynamic disturbances such as sudden load changes and temperature drift in real time, smooth torque fluctuations, and prevent stress concentration from causing overload damage to the tooth surface or sealing structure.

[0081] (2) Control logic and implementation process:

[0082] Signal purification: First, the extended Kalman filter algorithm is used to reduce noise in the torque sensor signal, improving the signal-to-noise ratio by more than 30dB, thereby accurately identifying the real load changes.

[0083] Feedforward predictive control: Model predictive control is employed, based on historical data stored in the data storage module, to establish a dynamic correlation model of "torque-temperature-load". This model predicts the torque change trend for the next 50 ms every 10 ms. Based on this prediction, the controller adjusts the power output in advance through the bidirectional DC-DC converter, thereby achieving proactive suppression of torque fluctuations.

[0084] Robust control against strong disturbances: When encountering sudden strong disturbances (such as valve jamming causing a sudden increase in load), the sliding mode variable structure control algorithm immediately intervenes. This algorithm, through its discontinuous control law, forces the system state trajectory to quickly slide towards the preset stability surface, so that the torque value can quickly return to the target range, avoiding erroneous shutdown due to momentary overshoot triggering the torque limiter (30-50 N·m threshold).

[0085] The dynamic adjustment of the second buffer stage can effectively protect the double sealing structure of the connecting sleeve 2, avoid the sudden increase in torque causing the stress at the sealing interface to exceed the material strength limit of 470MPa, thereby preventing leakage of high-pressure media (such as 10MPa helium gas).

[0086] Third Maintenance Phase: Maintenance Phase – Stabilize output and ensure proper execution of actions.

[0087] Control objective: To maintain a constant and appropriate torque output, ensure that drive shaft 1 accurately completes a rotation of 60°±0.1°, and achieve "soft positioning", ultimately guaranteeing the accuracy of disengagement or engagement.

[0088] Control logic and implementation process: Constant torque maintenance: When the torque reaches the preset target value (e.g., 30 N·m when engaged, 25 N·m when disengaged), the system enters the holding phase.

[0089] Dynamic parameter compensation: Adaptive control algorithm is adopted to identify and compensate for the impact of environmental changes (such as thermal expansion of drive shaft 1 caused by high temperature up to 500℃, resulting in increased load) on the system in real time, and automatically fine-tune the output torque to ensure uniform and stable rotation speed.

[0090] High-precision position synchronization and soft landing: A redundant position monitoring system integrating a laser displacement sensor, a Hall encoder, and a rotary transformer (with a resolution of 0.001°) tracks the angle of drive shaft 1 in real time. When drive shaft 1 is detected to have rotated to a target angle close to 60° (e.g., 59.8°), the controller commands the torque to slowly decrease to a low holding threshold (e.g., 5 N·m), achieving "soft landing" and effectively avoiding angle overshoot caused by inertia.

[0091] Closed-loop operation and linkage: After the action is completed, the torque output immediately returns to zero to prevent deformation of the tooth surface due to long-term stress. This stage is linked with the angle calibration program: If the redundant position monitoring system detects an angle deviation exceeding ±1°, the torque output during the holding stage will be immediately paused. The system will first trigger the automatic calibration program, controlling the electric actuator to fine-tune the angle to the target range of 60°±0.1°, and then complete the subsequent torque holding and zeroing actions.

[0092] By employing the aforementioned three-stage torque control strategy, the maximum impact torque during meshing is reduced from the traditional 45 N·m to below 25 N·m, significantly reducing tooth wear and sealing structure aging, and extending the device's service life. Through dynamic adjustment and temperature compensation, the device can operate stably within a wide range of -50℃ to 500℃ and 0-10 MPa, meeting the stringent requirements of the pharmaceutical and chemical industries. The three-stage torque control strategy, combined with high-precision sensing and intelligent algorithms, ensures that the disengagement / engagement angle deviation is ≤ ±0.1°, fundamentally avoiding power transmission interruptions or residual measurement interference caused by angle deviations.

[0093] Furthermore, the control unit also integrates a backup power protection module, which is composed of a backup lithium battery and a mechanical energy storage spring. The backup lithium battery and the mechanical energy storage spring are connected through a power coupler. The backup power protection module is also equipped with a power monitoring unit and a fast switching circuit. The power monitoring unit collects the main power supply voltage signal in real time. When the main power supply voltage is detected to be lower than a preset voltage value (e.g., 80% of the rated voltage), the fast switching circuit is immediately triggered to provide emergency power in the event of main power failure, ensuring that the lead-bismuth valve completes the emergency closing action and the disengagement action of the actuating unit from the valve core.

[0094] It should be noted that the control unit also integrates a backup power protection module, which is composed of a backup lithium battery and a mechanical energy storage spring. This module is used to provide emergency power support in the event of a main power failure, ensuring that the lead-bismuth valve can complete the emergency closure and the disengagement of the actuating unit from the valve core. The working mechanism of the mechanical energy storage spring is as follows: when the main power is normal, the transmission device rotates to drive the pawl to compress the spring and store energy; when the main power fails, the pawl unlocks and the spring releases energy.

[0095] The backup lithium battery is a high-rate lithium-sulfur battery with a rated capacity of not less than 5000mAh and an operating temperature range of -50℃ to 85℃. When the main power supply is normal, it maintains a float charge state through a DC-DC converter (the charging voltage is stable at 3.7V±0.05V) to ensure that the battery is always fully charged and ready for use. The mechanical energy storage spring is made of piano wire and has an initial energy storage torque of not less than 8N·m after prestressing. It is unidirectionally engaged with the input end of the transmission device through a pawl mechanism. When the main power supply is normal, it rotates synchronously with the transmission device to complete energy storage. When the main power supply fails, the pawl unlocks and releases the elastic potential energy.

[0096] The backup power protection module is also equipped with a power monitoring unit and a fast switching circuit. The power monitoring unit collects the main power supply voltage signal in real time at a sampling frequency of not less than 1kHz. When the main power supply voltage is detected to be lower than 18V (80% of the rated operating voltage), the fast switching circuit is immediately triggered. The fast switching circuit is constructed using a metal-oxide-semiconductor field-effect transistor (MOSFET) power switch, and the switching response time is no more than 10ms, which can realize uninterrupted power supply switching between the main power supply and the backup power supply.

[0097] After the main power fails, the backup lithium battery and the mechanical energy storage spring jointly output power through a power coupler. The backup lithium battery prioritizes providing power to the valve drive device, which drives the transmission device to move the valve core in the closing direction. The mechanical energy storage spring simultaneously releases torque to compensate for the problem of insufficient instantaneous output power of the lithium battery. When the valve reaches the fully closed position, the backup power continuously outputs stable power for no less than 3 seconds, driving the transmission device to rotate to the preset disengagement angle (60°±0.1°), ensuring that the action unit and the valve core are completely disengaged from physical contact, avoiding the risk of valve jamming or media leakage caused by the failure of the main power.

[0098] Furthermore, the control unit also integrates a redundant position monitoring system, which consists of a laser displacement sensor, a Hall encoder, and a rotary transformer. This system is used to collect and feedback the position information of the lead-bismuth valve core in real time, providing redundant and reliable data source support for the valve position closed-loop control. The measuring end of the laser displacement sensor faces the valve core body to obtain the absolute position reference signal of the valve core. The Hall encoder is connected to the input end of the transmission device through a flange structure to capture dynamic position change data during valve driving and transmission in real time. The rotary transformer is installed at the output end of the transmission device near the connecting sleeve 2 through a sealed mounting base to output continuous analog position signals under high temperature and strong electromagnetic interference conditions.

[0099] It should be noted that the control unit also integrates a redundant position monitoring system, which consists of a laser displacement sensor, a Hall encoder, and a rotary transformer. This system is used to collect and feedback the position information of the lead-bismuth valve core in real time, providing redundant and highly reliable data source support for the valve position closed-loop control. The laser displacement sensor employs a non-contact optical measurement principle, with its measuring end facing the valve stem end face or valve core body rigidly connected to the valve core. It is configured with a position measurement resolution of no less than 0.001° to acquire the absolute position reference signal of the valve core, eliminating position detection deviations caused by transmission device clearances and wear. The Hall encoder is fixed to the output shaft of the valve drive device or the input end of the transmission device via a flange structure, and is set to a position sampling frequency of no less than 1kHz. This is used to capture dynamic position change data during valve drive and transmission processes in real time, providing data support for high-frequency response adjustment of the closed-loop control. The rotary transformer is fixed to the transmission device near the connecting sleeve 2 via a sealed mounting base. The output terminal has a wide operating temperature range of -50℃ to 500℃ and an electromagnetic interference resistance of not less than 100V / m. It is used to output continuous analog position signals under high temperature and strong electromagnetic interference conditions to compensate for the adaptability defects of laser displacement sensors and Hall encoders in extreme environments. The signal output terminals of the laser displacement sensor, Hall encoder and rotary transformer are all electrically connected to the controller through shielded cables. The position data collected by the three are processed by the controller's built-in data fusion algorithm (preferably extended Kalman filter algorithm) for noise removal, data complementarity verification and redundancy fusion to form a valve position feedback signal with an accuracy of not less than ±0.03°. Based on the deviation between the feedback signal and the target position command, the controller dynamically adjusts the output parameters of the valve drive device to construct a closed-loop control link of "command issuance - position detection - deviation analysis - parameter correction - execution feedback" to ensure that the valve position control accuracy is not less than ±0.1°, which meets the precision position control requirements under high temperature and high pressure conditions of lead-bismuth media.

[0100] Furthermore, one end of the connecting sleeve 2 is provided with two symmetrically arranged arc-shaped teeth, and the length error of the arc-shaped teeth is controlled within ±0.1mm; the output end of the transmission shaft 1 is provided with a tooth groove that meshes with the arc-shaped teeth, one end of the connecting sleeve 2 meshes with the output end of the transmission shaft 1, the meshing surface is coated with a solid lubricating coating, and it has a built-in double sealing structure including a metal sealing ring and an elastic sealing ring, which can withstand a pressure of not less than 10MPa. The other end of the connecting sleeve 2 is rigidly connected to the valve stem through a flange.

[0101] It should be noted that the meshing structure between the drive shaft 1 and the connecting sleeve 2 is made of high-strength alloy material to ensure that the hardness of the tooth surface is not less than HRC50 at a high temperature of 300℃. Temperature field simulation has verified that the local temperature rise is ≤150℃, avoiding thermal deformation from affecting transmission accuracy. The metal graphite sealing ring in the double sealing structure is made of nickel-based high-temperature alloy (Inconel 718), and the fluororubber elastic ring is made of perfluoroether rubber (FFKM), with a leakage rate ≤1×10⁻⁶ in a 10MPa high-pressure helium environment. -6 The pressure field simulation shows a uniform stress distribution at the sealing interface, with a peak stress ≤400MPa. Furthermore, the flange connection adopts the ISO5211 standard interface, and an anti-loosening locking nut is added to the valve stem end. The preload torque is controlled within the range of 50-70 N·m, and stiffness simulation verifies that the deformation at the connection is ≤0.05mm. Considering fluid characteristics, the meshing clearance is designed to be 0.02-0.05mm, combined with a streamlined lubricant channel. Fluid simulation shows that helium flow rate fluctuation is <±5%, effectively suppressing pressure pulsation. The control method includes a stepper motor drive module, which adjusts the gear meshing displacement in real time through a closed-loop feedback system, achieving a positional accuracy of ±0.01mm. Combined with the controller's control logic, it achieves dynamic flow balance in the pharmaceutical mixing system.

[0102] A control method for the lead-bismuth valve control device, such as Figure 4 The steps shown are as follows:

[0103] Step 1: Monitor external control commands, local manual signals, fault warning signals, and power outage emergency signals in real time;

[0104] Step 2: Based on the real-time motion data of the valve drive device and the stored historical correlation data, calculate the estimated opening and closing state of the valve through a mapping model and evaluate its confidence level; the evaluation method is based on the standard deviation calculated by the distribution of historical data, and the confidence interval is ±2σ.

[0105] Step 3: Based on the difference between the target command and the estimated opening / closing state, output a control signal to drive the valve to move, and achieve closed-loop control through feedback from the redundant position monitoring system;

[0106] Step 4: After confirming that the valve has reached the target position, control the transmission device to rotate to the predetermined angle so that the actuating unit disengages from the valve core;

[0107] Step 5: Based on the real-time collected torque, angle, pressure, and temperature data, the control parameters are dynamically corrected and the historical correlation data model is updated using sliding mode control and extended Kalman filter algorithm.

[0108] Step 6: When a fault or power failure signal is triggered, activate the backup power and perform an emergency reset according to the preset safety strategy;

[0109] Step 7: Record all operational data and upload it to the operation and maintenance platform for health assessment and preventive maintenance decisions.

[0110] It should be noted that in step 1, the external control command comes from the remote control system and is the valve opening / closing target command received through the remote communication interface of the Industrial Communication Protocol (Modbus); the local manual signal is the manual operation command (such as emergency stop, manual adjustment) input by the field personnel through the human-machine interface; the fault warning signal is the warning signal triggered by abnormal data detected by the sensor (such as torque exceeding the threshold of 35 N·m, temperature exceeding 500℃, leakage rate exceeding the standard); the power failure emergency signal is the power interruption signal detected by the main power monitoring module.

[0111] In step 2, the real-time drive angle of the valve drive device is obtained from the data acquisition module, and historical correlation data between the valve opening degree and the drive angle is retrieved from the data storage module (e.g., "drive angle 90° corresponds to opening degree 100%" in historical data). The real-time drive angle is input into the historical mapping model of "drive angle - valve opening degree" stored in the data storage module (e.g., a neural network model or linear regression model trained on historical data) to calculate the estimated opening state of the current valve (e.g., "drive angle 45° corresponds to estimated opening degree 50%"). Based on the distribution of the real-time drive angle in the historical correlation data sequence (e.g., the opening degree corresponding to drive angle 45° in historical data is all within the range of 49%-51%), the confidence interval of the estimated opening state is determined (e.g., 98%). If the confidence interval is higher than the preset threshold (e.g., 95%), the estimated state is determined to be valid, and the next step of action control is entered; if it is lower than the threshold, a real-time correction program is triggered (e.g., the estimated value is corrected by combining the actual position data of the redundant position monitoring system), and the mapping model parameters are updated.

[0112] In step 3, based on the difference between the target command and the estimated opening / closing state, the controller outputs a control signal to drive the valve to move, achieving closed-loop control. The difference calculation involves comparing the external target command (e.g., "opening / closing degree 60%)" with the estimated opening / closing state (e.g., "opening / closing degree 50%)" to obtain a difference value (e.g., a difference of 10%). The signal output is based on this difference value; the controller outputs a 4-20mA control signal to the valve drive device (e.g., a 12mA signal corresponds to a 10% difference), driving the transmission device to move the valve core. Real-time feedback involves the redundant position monitoring system collecting the actual valve position data in real time and feeding it back to the controller. Dynamic adjustment: the controller adjusts the control signal based on the feedback data (e.g., adjusting the output signal to 14mA when the actual position reaches 55%) until the actual valve position matches the target command (error ≤ ±0.1° corresponding to the opening / closing degree error). In step 5, based on real-time operating data, the controller dynamically corrects the control parameters and model to improve system adaptability. Specifically, this is achieved by the data acquisition module collecting multi-dimensional data such as torque, angle, pressure, and temperature in real time. The controller uses a sliding mode control algorithm to suppress sudden load changes. To mitigate parameter fluctuations, an extended Kalman filter algorithm is used to reduce data noise and eliminate measurement errors. Based on the processed data, the algorithm parameters of the torque dynamic control module (such as the weight coefficients of the fuzzy neural network) and the correlation parameters of the mapping model are corrected. The corrected parameters and new operating data are stored in the data storage module to update the historical correlation data model and improve the accuracy of subsequent state estimation and action control. In step 6, when a fault warning signal or a power outage emergency signal is triggered, the system activates the emergency mechanism. Specifically, if a fault such as torque exceeding the threshold or leakage rate exceeding the standard is detected, the controller immediately outputs a pause command to stop the valve action. At the same time, the processing strategy in the fault record (such as starting sliding mode control to force torque reduction when torque overshoot) is retrieved and the emergency operation is executed. If the fault cannot be resolved on its own, an alarm message is sent to the operation and maintenance platform. After the main power supply fails, the backup power protection module switches power supply within 10ms, and the controller drives the valve action according to the preset safety strategy (such as "prioritize closing the valve and disconnecting") to ensure that the medium does not leak and the equipment is not damaged. After the power supply is restored, the system automatically records the operating data during the power outage and uploads it to the operation and maintenance platform. In step 7, the valve's action parameters (action time, rotation angle, torque change), fault information (fault type, occurrence time, handling result), and operating parameters (temperature, pressure, leakage rate) are recorded through the data storage module. The recorded data is then uploaded to the operation and maintenance platform via the Modbus protocol remote communication interface. Based on the uploaded data, the operation and maintenance platform performs equipment health assessments (such as predicting tooth surface wear through torque change trends) and makes preventative maintenance decisions (such as setting a "remind to replace connecting sleeve 2 when tooth surface wear reaches a threshold") to reduce equipment failure rate and operation and maintenance costs.

[0113] Furthermore, in step 4, the predetermined angle of rotation of the transmission device is 60°±0.1°; during re-engagement, if the angle deviation exceeds ±1°, the angle calibration program is automatically triggered; it should be noted that after confirming that the valve position meets the standard, the controller outputs a disengagement command, controlling the transmission device to rotate to the predetermined angle of 60°±0.1°, so that the arc-shaped teeth of the transmission shaft 1 and the connecting sleeve 2 disengage, and the action unit and the valve core are completely separated from physical contact, eliminating the interference of the drive device on the measurement system; when the valve needs to be controlled again, the controller drives the transmission device to rotate in the opposite direction to attempt to engage with the connecting sleeve 2; if the angle sensor detects that the engagement angle deviation exceeds ±1°, the angle calibration program is automatically triggered—the drive device fine-tunes the rotation angle until the deviation is ≤±0.1°, achieving precise engagement;

[0114] It should be noted that traditional valve control devices suffer from the following problems: transmission chain backlash leads to idling errors, mechanical wear increases tooth flank clearance, and thermal deformation causes measurement drift, making it difficult to achieve high angle control accuracy. This application achieves a breakthrough in accuracy by combining redundant position monitoring, composite material structure, and closed-loop feedback calibration, addressing both error source suppression and real-time compensation. It achieves an angle control accuracy of ±0.1° and a torque control accuracy of ±0.5 N·m, solving the problem of "insufficient control accuracy." Specifically, in terms of redundant position monitoring, the system adopts a three-in-one structure of "laser displacement sensor, Hall encoder, and rotary transformer," effectively complementing each other for different error sources. The laser displacement sensor provides an absolute position reference with a resolution of 0.001°, directly avoiding the "idling error" problem caused by transmission chain backlash. Traditional single encoders are prone to "deviation between commanded angle and actual position" due to transmission backlash, while the laser sensor, through non-contact absolute measurement, can directly capture the actual position of the valve core, compressing the backlash error from the traditional ±0.5° to within ±0.05°. The rotary transformer possesses wide temperature stability from -50℃ to 500℃, resisting thermal deformation interference under high-temperature conditions. Traditional sensors are prone to measurement drift due to material thermal expansion and contraction above 300℃, with errors reaching ±0.3°. The rotary transformer, through its high-temperature resistant structural design and Kalman filtering algorithm, controls the measurement error caused by thermal deformation to within ±0.02°. After data from the three types of sensors are processed by a fusion algorithm, random errors are further eliminated, ultimately providing a measurement accuracy of ±0.03° for angle control and laying the hardware support for achieving a control accuracy of ±0.1°. Regarding composite materials and precision structures, the drive shaft 1 uses a composite material of 316L stainless steel and titanium alloy. The low coefficient of thermal expansion of titanium alloy (only half that of carbon steel) significantly reduces thermal deformation. Traditional carbon steel drive shaft 1 experiences thermal deformation of up to 0.15mm at 300℃, resulting in a transmission angle deviation of ±0.2°. In contrast, the composite material drive shaft 1 controls thermal deformation within 0.03mm, corresponding to an angle deviation of only ±0.04°. The length error of the arc-shaped teeth of the connecting sleeve 2 is strictly controlled within ±0.1mm, and the meshing surface is coated with a solid lubricating coating, effectively reducing the tooth backlash caused by long-term wear. Traditionally, the tooth backlash of the connecting sleeve 2 can expand from 0.05mm to 0.2mm after wear, corresponding to an angular deviation of ±0.3°. This application, through high-precision machining and a wear-resistant coating, maintains the tooth backlash within 0.08mm after wear, with an angular deviation ≤±0.06°. Regarding closed-loop feedback calibration, the controller constructs a "command-measurement-deviation-correction" closed-loop link, comparing the target angle with the actual angle of redundant monitoring in real time. When deviations caused by transmission chain backlash, wear, or thermal deformation are detected, the compensation amount is immediately calculated through a mapping model, adjusting the output angle of the drive device to correct the deviation to within ±0.1° in real time.For example, in a chemical raw material proportioning scenario, the transmission chain develops a wear gap of 0.08mm due to long-term operation, corresponding to an angle deviation of 0.09°. The closed-loop system can detect the deviation within 10ms and output a compensation command, so that the final angle control accuracy is stabilized at 89.98°-90.02°, which fully meets the ±0.1° requirement.

[0115] Traditional proportional-integral-derivative (PID) control, due to its linear regulation characteristics, is prone to problems such as slow response and overshoot oscillation when the viscosity of the medium changes or the system load changes abruptly. This application achieves stable control under dynamic operating conditions and solves the problem of "poor dynamic adaptability" by "multi-algorithm fusion and three-stage strategy of torque dynamic control module". Specifically, in terms of multi-algorithm fusion, the torque dynamic control module integrates model predictive control, sliding mode control, and fuzzy neural network algorithm to form a hierarchical response mechanism for different dynamic disturbances. Model predictive control can predict load changes in advance and avoid response lag. When the viscosity of the medium suddenly increases (e.g., from 10 cP to 50 cP) and the load increases, the MPC algorithm predicts the load trend for the next 50 ms every 10 ms using real-time collected torque data and adjusts the drive torque output in advance (e.g., pre-increasing from 20 N·m to 28 N·m). Compared to the "lag adjustment" of traditional PID (which requires waiting for the deviation to appear before adjustment), the response speed is improved by 3 times, effectively avoiding overshoot caused by untimely adjustment. Sliding mode control suppresses sudden disturbances with its strong robustness. When the system experiences valve jamming (load suddenly increases by 50%), traditional PID is prone to oscillation due to fixed parameters, with overshoot reaching 20%. However, sliding mode control quickly pulls the torque back to the target range through "on / off adjustment," controlling the overshoot within 5% and without oscillation. The fuzzy neural network algorithm accurately compensates for nonlinear disturbances. The viscosity of the medium exhibits nonlinear characteristics with temperature changes (e.g., viscosity decreases by 8% for every 10°C increase in temperature). Traditional PID cannot dynamically adapt, while the fuzzy neural network algorithm learns the correlation between temperature, viscosity, and torque, correcting control parameters in real time to keep the torque output deviation always within ±0.5 N·m, avoiding control instability caused by nonlinear disturbances. In terms of the "preload-buffer-hold" three-stage strategy, the system achieves a smooth transition through segmented control for scenarios with sudden load changes. During the preloading phase, the system applies a low-speed load of 30%-50% of the target torque (e.g., if the target torque is 30 N·m, the initial output is 12 N·m) to allow the transmission system to gradually adapt to load changes, avoiding the shock overshoot caused by traditional "full torque start-up". During the buffering phase, extended Kalman filtering reduces noise in the torque data, accurately identifying load change trends (e.g., distinguishing between "instantaneous stall" and "continuous load increase"), and adjusting the torque increase rate accordingly to avoid adjustment oscillations caused by misjudgment. During the holding phase, adaptive compensation for load fluctuations ensures that when medium viscosity fluctuations cause torque variations between 25-27 N·m, the system automatically fine-tunes the output to maintain torque stability, ensuring smooth valve operation without sudden changes in speed.

[0116] In traditional devices, vibration and torque fluctuations from the drive motor are transmitted to the valve via a fixed connection, severely affecting measurement accuracy. This application addresses the drive interference problem from two dimensions: "automatic disengagement design + vibration suppression structure + torque stabilization control," specifically through "automatic disengagement design + vibration suppression structure + torque stabilization control." The automatic disengagement design physically isolates the interference source. After the valve reaches the target position, the controller automatically controls the transmission device to rotate 60°±0.1°, completely disengaging the actuating unit (including the drive device and transmission shaft 1) from the valve core. This design completely cuts off the vibration transmission path; the high-frequency vibration of the drive motor (the amplitude of which can reach 0.1mm in traditional devices) is completely unable to be transmitted to the valve core after disengagement, keeping the valve stable. In precision measurement scenarios requiring a constant valve opening (such as density detection), flow fluctuations caused by vibration can be avoided, reducing measurement errors by 40%. Simultaneously, the influence of torque fluctuations is eliminated. Torque fluctuations in the drive unit (traditional unit fluctuation range ±2 N·m) no longer act on the valve core after disengagement. The valve opening is maintained solely by its own structure, eliminating the need for the drive unit to continuously output torque to "maintain position," thus completely resolving the opening drift problem caused by torque fluctuations. Regarding vibration suppression and smooth torque control, the transmission device employs an elastic buffer structure. A metal elastic gasket is installed at the connection between the transmission shaft 1 and the drive unit, absorbing over 30% of the motor vibration, reducing the vibration amplitude transmitted to the valve core from 0.1 mm to below 0.03 mm. The torque dynamic control module achieves stable output by replacing the traditional large-step adjustment with small-step fine-tuning, keeping the torque output change rate within 5 N·m / s. This avoids shock vibrations caused by sudden increases and decreases in torque, further reducing interference from the drive device on the valve. In practical applications, taking the pharmaceutical industry's liquid concentration measurement scenario as an example, traditional devices cause valve opening fluctuations of ±0.5° due to drive interference, corresponding to flow fluctuations of ±2%, severely affecting the accuracy of concentration measurement. This application, through automatic disengagement and vibration suppression, controls valve opening fluctuations to ±0.1° and flow fluctuations ≤0.5%, providing stable medium conditions for concentration measurement and reducing the measurement error from the traditional ±1.5% to ±0.5%.

[0117] Traditional devices lack adaptive capabilities, and operation and maintenance rely entirely on manual experience. This application constructs an intelligent operation and maintenance system through "dynamic parameter correction, full-process data recording, and predictive maintenance," achieving parameter self-adaptation and efficient management. Specifically, in terms of self-learning and self-correction, the controller continuously optimizes control parameters based on real-time data without manual intervention. Dynamic parameter correction uses sliding mode control and extended Kalman filtering algorithms to collect torque, angle, temperature, and pressure data in real time, analyzing equipment wear and operating condition trends. For example, after wear on the second tooth surface of the connecting sleeve, the meshing torque increases from 30 N·m to 32 N·m. The system can automatically identify this change and adjust the preload torque from 12 N·m to 13 N·m to maintain meshing accuracy, avoiding the cumbersome process of manual disassembly, inspection, and parameter adjustment required in traditional devices. The mapping model is updated periodically by integrating newly collected "drive angle - actual opening degree" data into the historical correlation model, optimizing the accuracy of state estimation. After 1000 hours of equipment operation, due to wear of transmission components, the original mapping relationship of "90° drive angle corresponds to 100% opening degree" showed a 0.5° deviation. The system automatically updates the model through self-learning, ensuring that the deviation between the estimated opening degree and the actual value is always ≤0.1°, eliminating the need for manual recalibration. Regarding full-process data recording, the data storage module records valve operation data in real time throughout its entire lifecycle, covering action parameters (action time, rotation angle, torque change curve), operating parameters (temperature, pressure, leakage rate), and fault information (fault type, occurrence time, and handling result). All data is uploaded to the maintenance platform via the Modbus protocol, replacing traditional manual recording (prone to omissions and large errors). For example, when a torque overshoot fault occurs, the platform can retrieve the torque change curve from 10 minutes prior to the fault, quickly locating causes such as "sudden load change" or "sensor malfunction," reducing troubleshooting time from the traditional 2 hours to 15 minutes. In terms of predictive maintenance, the maintenance platform achieves "on-demand maintenance" based on data analysis, assessing the equipment's health status through algorithmic models. For example, the wear degree of the connecting sleeve tooth surface 2 can be predicted by the torque fluctuation trend (when the torque fluctuation increases from ±0.3 N·m to ±0.6 N·m, the wear is judged to be close to the threshold), and the aging state of the seal can be judged by the change in leakage rate (the leakage rate increases from 1×10). -7 scc / s increased to 8×10 - 7When the speed is scc / s, a reminder to replace the seals is given. Under the traditional regular maintenance mode, problems may occur such as "replacing the seals before they are aged" (wasting costs) or "not replacing them after they are aged" (causing leakage). Predictive maintenance reduces maintenance costs by 25% and avoids equipment failures caused by human experience misjudgment, improving operation and maintenance reliability by 40%. In terms of actual operation and maintenance results, in the management practice of multiple valves in a large chemical industrial park, traditional operation and maintenance requires a dedicated person to regularly inspect and calibrate, with each person managing 10 devices. The intelligent system of this application can achieve the management of 50 devices per person, and the failure of equipment due to parameter mismatch is reduced from 3 times per month to 1 time per quarter, greatly reducing reliance on manual labor and significantly improving operation and maintenance efficiency and reliability.

[0118] It is worth noting that this invention is applicable to precision measurement scenarios in chemical, scientific research, and pharmaceutical industries, especially for high-temperature (-50℃ to 500℃), high-pressure (not exceeding 10MPa), and corrosive media (such as helium) environments. It enables high-precision valve control, automatic disconnection, and intelligent operation and maintenance, solving the problems of insufficient control precision, poor dynamic adaptability, drive interference, and low operation and maintenance efficiency in traditional valve control solutions. Through "redundant position monitoring + torque dynamic control + closed-loop feedback," control precision is significantly improved, achieving an angle control accuracy of ±0.1° and a torque control accuracy of ±0.5N·m. This eliminates interference from the drive device on the precision measurement system, meeting the high-precision process requirements of the chemical and pharmaceutical fields. Specifically, at the redundant position monitoring level, a three-in-one structure of "laser displacement sensor + Hall encoder + rotary transformer" is adopted to construct a multi-dimensional, highly reliable position sensing system. The laser displacement sensor provides an absolute position reference with an ultra-high resolution of 0.001°, ensuring fundamental measurement accuracy. The Hall encoder, with its fast response characteristics, captures dynamic position changes in real time, achieving millisecond-level data updates. The rotary transformer is designed for extreme environments, maintaining stable performance over a wide temperature range of -50℃ to 500℃, effectively resisting electromagnetic interference and preventing accuracy collapse due to the failure of a single sensor. After the data from these three types of sensors are fused using a Kalman filter algorithm, not only is measurement noise eliminated, but the data complementarity also improves the accuracy of position determination, providing the hardware foundation and algorithmic support for achieving an angle control accuracy of ±0.1°.

[0119] The multi-algorithm fusion and three-stage strategy of the torque dynamic control module are key to achieving a torque accuracy of ±0.5 N·m. The module integrates model predictive control, sliding mode control, and fuzzy-neural network fusion algorithms, forming a dynamic adaptability for different operating conditions: the MPC algorithm predicts the torque change trend for the next 50 ms every 10 ms, adjusting output parameters in advance to avoid accuracy deviations caused by lag; sliding mode control, through strong robustness, quickly suppresses torque fluctuations caused by sudden load changes, valve jamming, and other emergencies; and the fuzzy-neural network algorithm accurately compensates for errors caused by nonlinear factors such as temperature drift and material wear. Combined with the "preload-buffer-hold" three-stage strategy, the preload stage achieves flexible contact with low-amplitude torque, the buffer stage uses extended Kalman filtering for noise reduction, and the holding stage adaptively compensates for environmental interference, avoiding rigid impacts and torque overshoot throughout the process, ensuring the stability and accuracy of torque output.

[0120] The closed-loop feedback mechanism maximizes the effectiveness of the two technologies mentioned above. The controller compares the target command with the actual data fed back from redundant position monitoring and torque sensors in real time, and dynamically adjusts the control signal through a mapping model, forming a closed-loop link of "command issuance - data acquisition - deviation analysis - parameter correction". For example, in the scenario of chemical raw material proportioning, a small deviation in the valve opening angle will directly affect the material mixing accuracy. The closed-loop feedback can immediately start adjustment when an angle deviation of ±0.05° is detected, ensuring that the valve action is highly consistent with the target command. This collaborative mechanism not only achieves high-precision control of angle and torque, but also completely cuts off the torque and vibration transmission path of the drive device through the automatic disengagement design of the action unit and valve core, avoiding the reading deviation of the weighing sensor and density detector caused by the traditional fixed connection method, and providing an interference-free operating environment for the precision measurement system.

[0121] This invention addresses special operating conditions such as high temperature, high pressure, and corrosive media. Through an innovative design of a composite material drive shaft 1 and a double-sealing structure for the connecting sleeve 2, it achieves stable operation within a temperature range of -50℃ to 500℃ and a pressure range of ≤10MPa, with a leakage rate controlled to ≤1×10⁻⁶. -6The SCC / S design solves the problem of poor adaptability to operating conditions in traditional devices. Specifically, the drive shaft 1 uses a composite material of 316L stainless steel and titanium alloy, combining the advantages of high strength and corrosion resistance. 316L stainless steel has excellent corrosion resistance, resisting the erosion of corrosive media such as helium, acids, and alkalis, while the titanium alloy significantly improves the high-temperature stability of the drive shaft 1, ensuring that the tooth surface hardness remains ≥HRC50 even at 300℃, avoiding power transmission deviations caused by thermal deformation. This composite material design allows the drive shaft 1 to adapt to a wide temperature range, from low-temperature freezing conditions to high-temperature reaction environments, while resisting wear and erosion of components by corrosive media, extending the service life of the equipment in harsh environments. The double-sealing structure of the connecting sleeve 2 is the core of achieving high-pressure sealing and low leakage rate. The metal sealing ring is made of Inconel 718 material, which has excellent high-temperature resistance, corrosion resistance, and pressure resistance, maintaining a stable sealing shape under high-temperature and high-pressure environments. The elastic sealing ring uses perfluoroether rubber, designed for highly corrosive media, and does not react with acids, alkalis, or organic solvents commonly used in the chemical and pharmaceutical fields. The dual-sealing structure is complementary, with the metal sealing ring bearing the main pressure load and the elastic sealing ring filling the tiny gaps. Combined with the rigid connection of the ISO5211 standard flange and a preload torque of 50-70 N·m, this ensures that the device maintains good sealing performance under high pressure conditions ≤10 MPa, with the leakage rate controlled to ≤1×10⁻⁶. -6 The leakage rate is significantly lower than that of traditional single-seal structures, with a leakage rate of only 5 × 10⁻⁶ scc / s. In practical applications, such as high-temperature sterilization processes in the pharmaceutical industry, valves need to operate frequently under 300°C and 8MPa pressure. Traditional devices are prone to problems such as seal aging and drive shaft deformation, leading to media leakage and decreased control accuracy. However, the device of this invention, through composite materials and a double-seal design, can operate stably for a long time under such conditions, with a leakage rate of only 5 × 10⁻⁶ scc / s. -7 SCC / s ensures no leakage of sterilization media and no disruption of the process environment. In corrosive media transmission scenarios in the chemical industry, traditional carbon steel drive shafts are easily corroded, resulting in rough surfaces and affecting power transmission accuracy. However, the composite material drive shaft of this invention can resist corrosion for a long time, maintain smooth transmission, and ensure the stability of valve control.

[0122] This invention significantly improves the intelligence and automation level of the control device through remote control, automatic disengagement / engagement, and self-learning and self-correction functions. It enables valve control under complex operating conditions without on-site personnel, while adapting to equipment wear and changing operating conditions, greatly reducing the cost and safety risks of manual intervention. Specifically, the remote control function is implemented based on a Modbus protocol remote communication interface. Operators can send target commands such as opening / closing degree and action speed to the valve control device through the maintenance platform, and view key information such as equipment operating status, torque changes, and position data in real time. In sterile production workshops in the pharmaceutical industry, traditional manual on-site operation requires overcoming the sterile environment barrier, which not only affects the cleanliness of the production environment but also introduces operational delays and errors. Remote control enables non-contact operation, ensuring that the sterile environment is not compromised while improving operational efficiency, reducing valve adjustment response time from minutes to seconds.

[0123] The automatic disengagement / engagement function completely revolutionizes the traditional manual disengagement operation. Once the valve reaches the target position, the controller automatically outputs a disengagement command, controlling the transmission device to rotate 60°±0.1°, disengaging the actuating unit from the valve core. During re-engagement, if an angular deviation exceeding ±1° is detected, a calibration program is automatically triggered to ensure precise engagement. This function is particularly important in high-pressure reactor scenarios. Traditional manual disengagement requires operators to work near the high-pressure equipment, posing safety risks such as media leakage and pressure surges. Automatic disengagement / engagement, however, can be completed while the equipment is running, eliminating the need for personnel to approach and fundamentally avoiding operational risks. It also avoids engagement deviations and incomplete disengagement issues caused by manual operation.

[0124] The self-learning and self-correcting capabilities of this control device enable it to dynamically adapt. Based on real-time collected data such as torque, angle, temperature, and pressure, the controller dynamically corrects control parameters and mapping models, and updates historical correlation data through sliding mode control and extended Kalman filtering algorithms. As equipment operating time increases, transmission components will experience normal wear. Traditional devices require periodic shutdowns for calibration, affecting production continuity; however, the control device of this invention can automatically compensate for accuracy deviations caused by wear through its self-learning function. For example, when wear on the tooth surface of the connecting sleeve 2 causes changes in meshing torque, the control device can dynamically adjust the torque control parameters to maintain stable control accuracy, eliminating the need for frequent manual calibration and reducing maintenance costs and downtime losses.

[0125] This invention constructs a comprehensive safety assurance system through a backup power guarantee module, a fault emergency mechanism, and predictive maintenance design, achieving 10ms emergency power supply switching, significantly extending equipment lifespan, and greatly reducing the risk of unexpected downtime. Specifically, the backup power guarantee module consists of a backup lithium battery and a mechanical energy storage spring, forming a dual emergency power supply mechanism. When the main power supply is normal, the lithium battery is in a float charge state, and the mechanical energy storage spring remains pre-compressed; after the main power supply fails, the module completes the power supply switch within 10ms, providing power through the lithium battery and the mechanical energy storage spring to ensure that the valve completes the emergency closing and disengagement actions. In the scenario of chemical raw material transportation pipelines, a sudden interruption of the main power supply may cause the valve to fail to close, leading to serious accidents such as media leakage, fire, and explosion; while the backup power guarantee module can initiate emergency operation at the moment of power failure, quickly cutting off the media transmission path, preventing the accident from escalating, and providing a critical guarantee for safe production. The fault emergency mechanism has preset precise handling strategies for common faults such as torque overshoot, leakage, and abnormal temperature. When the torque sensor detects that the torque exceeds the safety threshold of 35 N·m, the system immediately triggers sliding mode control to force torque reduction and simultaneously sends an alarm message to the operation and maintenance platform. If the leakage rate is detected to be excessive, the controller initiates an emergency shutdown procedure to ensure that the medium does not leak. Compared with the passive shutdown of traditional devices when they fail, the proactive emergency handling of this invention can minimize the impact of failures. For example, when the torque exceeds the limit, rapid torque reduction can prevent damage to the tooth surface and deformation of the drive shaft 1, reducing equipment maintenance costs and downtime. The predictive maintenance function is based on full-process data recording and analysis. The system stores information such as valve action parameters, fault records, and operating data. The operation and maintenance platform evaluates the health status of the equipment through data analysis. For example, it can predict the wear degree of the tooth surface of the connecting sleeve 2 through torque fluctuation trends, judge the aging status of the seals through changes in leakage rate, and issue preventive maintenance reminders. Traditional devices adopt a periodic maintenance mode, which has the problems of over-maintenance or untimely maintenance. Over-maintenance increases costs, while untimely maintenance can easily lead to sudden failures. Predictive maintenance can achieve "on-demand maintenance," replacing components when they are about to reach the wear threshold, avoiding unexpected downtime, significantly improving equipment lifespan, and greatly reducing the risk of unexpected downtime.

[0126] This invention constructs a data-driven operation and maintenance (O&M) system through full-process data recording, remote uploading, and centralized management on an O&M platform. This avoids blind repairs, reduces O&M costs by more than 25%, and significantly improves O&M efficiency. Specifically, the full-process data recording function covers all aspects of valve operation, including key data such as action time, rotation angle, torque changes, temperature and pressure, fault type, and emergency handling results. All data is stored in real time and uploaded to the O&M platform via the Modbus protocol. Traditional O&M relies on manual recording and on-site inspection, making it difficult to guarantee data accuracy and completeness, and tracing faults. Full-process data recording provides complete traceability for O&M. When equipment malfunctions, staff can retrieve historical data through the platform to quickly locate the cause of the fault. For example, by analyzing torque change curves, they can determine whether it is due to sudden load changes or sensor failure, shortening troubleshooting time. The centralized management function of the O&M platform supports unified monitoring and management of multiple devices. Staff can view the operating status of all valve control devices, issue maintenance commands, and receive alarm information on the same platform, eliminating the need for individual on-site inspections. In large chemical industrial parks, dozens or even hundreds of valve devices are often deployed. Traditional manual inspections consume a significant amount of manpower and time and have blind spots. Centralized management, however, enables real-time monitoring of equipment status, timely alarms for abnormalities, and improves inspection efficiency by over 50%. It also reduces on-site work time for inspection personnel and lowers operational risks. Health assessment and preventative maintenance functions shift maintenance from "passive repair" to "proactive prevention." Based on uploaded data, the maintenance platform uses algorithmic models to assess equipment health status, generate health reports, clarify key information such as component wear levels and remaining service life, and develop targeted maintenance plans. For example, when the leakage rate of the sealing component of connecting sleeve 2 increases from 1×10⁻⁶... -7 scc / s increased to 8×10 -7 When the system reaches scc / s, the platform reminds staff to replace the seals during the next production break to avoid downtime caused by seal failure. This precise maintenance model avoids the blind repairs of traditional periodic maintenance, reduces unnecessary component replacements and downtime, lowers maintenance costs by more than 25%, and improves the continuity and stability of equipment operation.

[0127] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A lead bismuth valve control device, characterized by, The action unit and the control unit are included; The action unit includes a valve driving device and a transmission device; The transmission device includes a transmission shaft, and a connecting sleeve is connected to an output end of the transmission shaft; An output end of the valve driving device is connected to an input end of the transmission shaft, and is used to drive the transmission shaft to drive; an output end of the transmission shaft is detachably connected to a valve core of the lead bismuth valve through the connecting sleeve; The control unit includes: A data acquisition module is used to acquire motion data of the valve driving device, motion data of the transmission device, and dynamic contact data between the transmission shaft and the connecting sleeve; A data storage module at least stores motion correlation between the valve driving device and the transmission device, and historical correlation data between valve opening and closing degrees and driving angles of the valve driving device; A controller is connected to the valve driving device, the data acquisition module, and the data storage module, and is configured to: based on the real-time acquired motion data of the valve driving device and the historical correlation data, calculate and acquire a current estimated opening and closing state of the valve through a pre-established mapping model, output a control signal to the valve driving device according to a difference between the estimated opening and closing state and a target instruction, and drive the transmission device and the valve core to act; after the valve reaches a target position, output a disengagement instruction to control the action unit to disengage from physical contact with the valve core; and based on multi-element data acquired by the data acquisition module in real time, dynamically correct and update the historical correlation data and the mapping model; The motion data of the valve driving device at least includes a rotation angle, a rotation speed, and an output torque; and the motion data of the transmission device at least includes a rotation angle and vibration data; The motion data acquisition method of the transmission device and the valve driving device is as follows: If the transmission device and the valve driving device move synchronously, the motion data of the transmission device is consistent with the motion data of the valve driving device; If the transmission device and the valve driving device move asynchronously, the motion data of the transmission device is acquired based on an asynchronous motion correlation between the valve driving device and the transmission device; The dynamic contact data between the transmission shaft and the connecting sleeve at least includes contact pressure, pressure change gradient, and effective contact time; The controller is configured to acquire the opening and closing state of the valve by the following steps: Acquire a real-time driving angle of the valve driving device; Retrieve the historical correlation data from the data storage module, which records a mapping relationship between the driving angle and the actual opening and closing degree of the valve; Input the real-time driving angle into the mapping relationship to calculate an estimated opening and closing state of the current valve; Determine a confidence interval of the estimated opening and closing state based on a distribution of the real-time driving angle in the mapping relationship data sequence; If the confidence interval is higher than a preset threshold, it is determined that the estimated opening and closing state is valid; if it is lower than the preset threshold, a real-time correction program is triggered, and the mapping relationship is updated.

2. The lead bismuth valve control apparatus according to claim 1, characterized in that, The control unit further comprises a torque dynamic control module; the torque dynamic control module is integrated with a model predictive control, a sliding mode control and a fuzzy-neural network fusion control algorithm, and is configured to execute a "preload-buffer-hold" three-stage torque control strategy to achieve a torque control accuracy of ±0.5 N·m.

3. The lead bismuth valve control apparatus of claim 1, wherein The control unit is further integrated with a backup power guarantee module, which is composed of a backup lithium battery and a mechanical energy storage spring in cooperation, and the backup lithium battery and the mechanical energy storage spring are connected through a power coupler, wherein the backup power guarantee module is further provided with a power supply monitoring unit and a quick switching circuit, the power supply monitoring unit collects the main power voltage signal in real time, and when it is detected that the main power voltage is lower than the preset voltage value, the quick switching circuit is triggered immediately to provide emergency power in the main power failure scenario, ensuring that the lead bismuth valve completes the emergency closing and the disengagement action of the action unit and the valve core.

4. The lead bismuth valve control apparatus of claim 1, wherein The control unit is further integrated with a redundant position monitoring system, which is composed of a laser displacement sensor, a Hall encoder and a rotary transformer, for real-time acquisition and feedback of the position information of the lead bismuth valve core, providing a redundant and reliable data source support for valve position closed-loop control; The measurement end of the laser displacement sensor is opposite to the valve core body, for obtaining the absolute position reference signal of the valve core, the Hall encoder is connected with the input end of the transmission device through a flange structure, for real-time capturing of dynamic position change data in the valve driving and transmission process; the rotary transformer is installed on the output end of the transmission device close to the connecting sleeve through a sealed mounting seat, for outputting continuous analog position signals under high temperature and strong electromagnetic interference conditions.

5. The lead bismuth valve control apparatus of claim 1, wherein, One end of the connecting sleeve is provided with two symmetrical arc teeth, and the length error of the arc teeth is controlled within ±0.1 mm; the output end of the transmission shaft is provided with a tooth groove engaged with the arc teeth, one end of the connecting sleeve is engaged with the output end of the transmission shaft, the engagement surface is coated with a solid lubricating coating, and a double sealing structure including a metal sealing ring and an elastic sealing ring is built-in, and the resistance to pressure is not less than 10 MPa, and the other end of the connecting sleeve is rigidly connected with the valve stem through a flange on the outside.

6. A control method for the Pb-Bi valve control device according to any one of claims 1 to 5, characterized by, The method comprises the following steps: Step 1: real-time monitoring of external control instructions, local manual signals, fault warning signals and power failure emergency signals; Step 2: based on the real-time motion data of the valve driving device and the stored historical correlation data, the estimated opening and closing state of the valve is calculated through a mapping model, and the confidence is evaluated; Step 3: according to the difference between the target instruction and the estimated opening and closing state, a control signal is output to drive the valve action, and a closed-loop control is realized through the feedback of the redundant position monitoring system; Step 4: after confirming that the valve reaches the target position, the transmission device is rotated to a predetermined angle to make the action unit and the valve core disengage; Step 5: based on the real-time collected torque, angle, pressure and temperature data, the sliding mode control and extended Kalman filter algorithm are used to dynamically correct the control parameters and update the historical correlation data model; Step 6: when the fault or power failure signal is triggered, the backup power is enabled and the emergency reset is executed according to the preset safety strategy. Step 7: Record all running data and upload to the operation and maintenance platform, and make health assessment and preventive maintenance decision.

7. The control method of the Pb-Bi valve control apparatus according to claim 6, wherein In step 4, the predetermined angle of rotation of the transmission device is 60°±0.1°; when re-engaging, if the angle deviation exceeds ±1°, the angle calibration program is automatically triggered.

Images