Deepwater dynamic seal self-adaptive underwater propeller intelligent control system

Abstract

The invention relates to the technical field of underwater robots and deep sea equipment, and discloses a deepwater dynamic seal self-adaptive underwater propeller intelligent control system. Comprising an environment sensing module, a data preprocessing module, a sensor health diagnosis module, a dynamic sealing compensation module, an adaptive adjustment module, an execution driving module, a feedback optimization module and a man-machine interaction interface module. A pressure difference sensor, a temperature sensor, a vibration sensor and insulation monitoring equipment form a distributed acquisition station of an environment sensing module, a sealing state real-time sensing and fusion diagnosis model based on digital twinning is constructed in a sensor health diagnosis module, and a parameter degradation prediction model is adopted to calculate a sensor health index. And according to a calculation numerical value range, on-line monitoring and diagnosis are carried out on early faults of micro leakage, abrasion and aging of the dynamic seal, and the system is converted from alarming after faults to early warning before failure.

Description

Technical Field

[0001] This invention relates to the field of underwater robots and deep-sea equipment technology, specifically to an intelligent control system for a deep-sea dynamic seal adaptive underwater propulsion system. Background Technology

[0002] With the deepening of marine resource development and deep-sea exploration activities, underwater thrusters, as the core power components of equipment such as ROVs (Remotely Operated Vehicles) and AUVs (Autonomous Underwater Vehicles), are crucial for long-term reliable operation. The deep-sea environment (usually referring to water depths exceeding 1000 meters) poses extreme challenges to the dynamic sealing system of thrusters: high-pressure seawater can easily cause seal failure, leading to leakage, motor short circuits, thrust reduction, or even system damage. Existing technologies mainly have the following shortcomings: (1) Traditional thrusters generally use passive mechanical seals, which cannot detect minute changes in the sealing state in real time. Alarms are often triggered only when leakage has caused serious failures, resulting in delayed warnings; (2) Existing control systems are mostly open-loop or simple closed-loop speed / thrust control, which do not take the sealing health status as the core control parameter and cannot actively adjust the operating strategy to prevent catastrophic failures when the sealing performance deteriorates; (3) There is a lack of intelligent diagnosis and life prediction capabilities for sealing failures based on the fusion of multi-source information (such as pressure, temperature, vibration, and insulation resistance). Maintenance relies on periodic disassembly and inspection, which is inefficient and costly. Summary of the Invention

[0003] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides an intelligent control system for deep-water dynamic seal adaptive underwater thrusters. It has the advantages of real-time status perception, intelligent diagnosis and early warning, and adaptive fault-tolerant control, and solves the problems of low reliability, delayed fault warning, and high risk of sudden failure in existing deep-water thrusters due to the unknown sealing status and the disconnect between control strategy and sealing health status.

[0004] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: an intelligent control system for a deep-sea dynamic seal adaptive underwater propulsion system, comprising an environmental perception module, a data preprocessing module, a sensor health diagnosis module, a dynamic seal compensation module, an adaptive adjustment module, an execution drive module, a feedback optimization module, and a human-machine interface module; The environmental sensing module collects temperature, salinity, flow rate, and pressure data in real time through distributed acquisition stations; The data preprocessing module collects data and performs preprocessing such as multi-sensor spatiotemporal synchronization calibration, outlier filtering and removal, and normalized feature extraction to eliminate noise and dimensional differences in the original data. The sensor health diagnosis module diagnoses the sensor status based on historical benchmark data and real-time measurement deviations, and calculates the sensor health index by combining the built-in multi-parameter degradation prediction model. ,when When the value is less than 0.7, the sensor enters a scrap warning state, triggering a switchover of backup facilities; The dynamic sealing compensation module integrates a set of fluid dynamics equations, establishes an oil film thickness-pressure-rotation speed-flow velocity coupled model, and performs feedforward compensation calculations based on real-time flow velocity data and thruster rotation speed commands to dynamically adjust the sealing cavity pressure. The adaptive parameter adjustment module is based on historical data of the system's dynamic response and incorporates an improved MPC algorithm framework. It uses a standard flow field with a velocity <0.5 m / s as the performance baseline and calculates key dynamic coupling values ​​of the system in real time, which are then used as the proportional coefficient for the PID controller. Integral coefficient And dynamic adjustment of feedforward gain; The execution drive module receives and executes instructions from the adaptive parameter adjustment module and the dynamic sealing compensation module to drive the propeller motor, buoyancy pump and proportional pressure valve. It contains a backup facility switching circuit and bus, which can seamlessly switch to the built-in backup facility switching unit when the main sensor or actuator fails. The closed-loop feedback and optimization module performs multi-source data fusion processing on the system output and compares it with the expected target. The optimization result is fed back to the adaptive parameter adjustment module and the dynamic sealing compensation module to realize online fine-tuning and learning of model parameters. The human-machine interface module provides a visual monitoring platform that centrally displays the status of each module in the system, environmental data, alarm information and performance curves, and allows operators to switch modes, preset parameters and remotely intervene.

[0005] Preferably, the environmental perception module constructs a three-dimensional environmental field model through four distributed acquisition stations (front, rear, left, and right) to provide an environmental state baseline for the system.

[0006] Preferably, the forward distributed acquisition station uses an integrated temperature, salinity, and depth sensor and a three-dimensional acoustic Doppler current profiler to monitor the flow field and environmental data directly in front of the thruster in real time.

[0007] Preferably, the post-distributed acquisition station uses an integrated temperature, salinity, and depth sensor and a particle image velocimetry system probe to monitor the wake field and vortex disturbance data in real time.

[0008] Preferably, the left and right distributed acquisition stations respectively monitor the corrosion environment and boundary layer data on the left and right sides in real time through a micro electrochemical sensor array and a laser Doppler velocimeter.

[0009] Preferably, the sensor health diagnosis module calculates the sensor health index. The calculation formula is as follows: In the formula, Indicates the amount of temperature drift. This represents the short-term standard deviation of the sensor output. This indicates the maximum allowable temperature drift threshold. This represents the maximum permissible standard deviation threshold. Indicates the environmental stress factor. , , These represent the weights of temperature stability, output stability, and environmental tolerance in the health assessment, respectively. This formula is then used to evaluate the sensor's condition: When When the value is less than 0.7, the sensor enters a scrap warning state, triggering the backup facility switching unit in the sensor health diagnosis module to start the backup sensor and report the warning event to the management module.

[0010] Preferably, the dynamic sealing compensation module has a built-in working condition identification unit and a compensation calculation unit, and its working process is as follows: S1.1 The operating condition identification unit identifies whether the current propeller is in a stable, accelerating, or turbulent operating condition by using real-time flow velocity pulsation intensity, propeller speed change rate, and sealed cavity pressure spectrum characteristic data. S1.2 The compensation calculation unit identifies the current working condition based on the working condition identification unit, and performs feedforward compensation calculation based on real-time flow velocity data and thruster speed command. The calculation formula is as follows: In the formula, Indicates dynamic sealing compensation pressure. Indicates hydrostatic pressure; This represents the velocity component perpendicular to the sealing surface. Indicates the density of seawater. Indicates the real-time rotational speed of the thruster. Indicates the design reference speed. This represents the ocean current disturbance gain coefficient. This represents the speed compensation coefficient. The real-time calculated value is applied to the sealing cavity through a proportional pressure valve to dynamically adjust the pressure in the sealing cavity, so as to stabilize the oil film thickness within the design range of ±0.05mm.

[0011] Preferably, the adaptive parameter adjustment module integrates a baseline comparison unit, a key value calculation unit, and a parameter mapping unit, and its workflow is as follows: S2.1 In the baseline comparison unit: Real-time acquisition of the current system response data provided by the data preprocessing module, comparison with the standard operating condition baseline stored in the unit, calculation of the deviation ratio of the current actual value relative to the standard baseline, generation of dynamic deviation vector, and output as a primary quantitative indicator of the degree of environmental disturbance. S2.2 In the key value calculation unit, the dynamic deviation vector output by S2.1 and the real-time environmental data can be combined to calculate a single scalar value representing the current dynamic complexity of the system. ; S2.3, Parameter Mapping Unit Based on Adjust the PID controller parameters according to the calculated value range.

[0012] Preferably, the key value calculation unit calculates the dynamically coupled key values ​​of the system based on the preprocessed data. The calculation formula is as follows: In the formula, Indicates the rate of change of attitude angle. This represents the norm of multi-attitude angular coupling error. Indicates turbulence intensity. , , α, β, and γ represent the upper limits of each range, respectively, and represent the attitude dynamic weight, coupling error weight, and turbulence disturbance weight, respectively.

[0013] Preferably, the parameter mapping unit is based on Adjust the PID controller parameters as follows: (1) When When the value is less than 0.4, the system is determined to be in a quasi-steady state or a micro-disturbance condition. The parameter adjustment strategy is to maintain the PID controller as the core control unit and use the standard proportional coefficient. Standard integral coefficient With standard differential coefficients The standard baseline parameters are used for operation, focusing only on the integral coefficients of the PID controller parameters. A correction ranging from 0.95 to 1.05 was made. (2) When 0.4≤ When the value is less than 0.7, the current system is determined to have moderate coupling and disturbance. The parameter adjustment strategy is as follows: use the PID controller as the main control unit and introduce a feedforward compensation element; dynamically adjust the PID parameters and increase the proportional coefficient. Enhance response speed to moderately reduce differential coefficients Suppressing moderately coupled oscillations; while the integral coefficient An integral separation strategy is adopted; (3) When When the value is ≥0.7, the current system is determined to be in a strongly nonlinear and strongly turbulent condition. The parameter adjustment strategy is to switch the main controller to the improved model predictive control algorithm and the original PID controller becomes an auxiliary correction loop.

[0014] Compared with the prior art, the present invention provides an intelligent control system for a deep-water dynamic seal adaptive underwater thruster, which has the following beneficial effects: 1. This invention utilizes a multi-parameter sensor array integrated within the propulsion chamber. This array comprises a distributed data acquisition station for the environmental sensing module, consisting of differential pressure sensors, temperature sensors, vibration sensors, and insulation monitoring equipment. Furthermore, a real-time sensing and fusion diagnostic model of the sealing condition based on digital twins is constructed within the sensor health diagnosis module. A parameter degradation prediction model is then used to calculate the sensor health index. Based on the calculated numerical range, the system performs online monitoring and diagnosis of micro-leakage, wear, and early aging faults of dynamic seals, realizing the transformation of the system from alarm after a fault to early warning before failure.

[0015] 2. This invention embeds an adaptive mapping algorithm between the seal health status and the thruster operating parameters into the control system, enabling it to dynamically and smoothly adjust the maximum allowable output torque, speed threshold, and operating mode (such as reduced power operation or intermittent operation) of the thruster based on the real-time diagnosed seal health level. This proactively implements protective operating strategies when the seal performance degrades, significantly extending the effective service life of the seals in deep water conditions and avoiding sudden complete failure.

[0016] 3. This invention establishes a seal life prediction model based on historical and real-time operational data, and links it with the mission planning module. This model can provide maintenance suggestions for autonomous decision-making of the vehicle (such as "recommend maintenance after mission" or "return immediately") and support online replanning of mission paths and speeds. This achieves the beneficial effect of improving the autonomy and mission reliability of the entire underwater system in long-endurance and complex missions. Attached Figure Description

[0017] Figure 1 This is a system flowchart of the present invention. Detailed Implementation

[0018] 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.

[0019] Please see Figure 1A deep-sea dynamic seal adaptive underwater propulsion intelligent control system includes an environmental perception module, a data preprocessing module, a sensor health diagnosis module, a dynamic seal compensation module, an adaptive adjustment module, an execution drive module, a feedback optimization module, and a human-machine interface module. The environmental sensing module collects temperature, salinity, flow rate, and pressure data in real time through distributed data acquisition stations; The data preprocessing module collects data and performs preprocessing such as multi-sensor spatiotemporal synchronization calibration, outlier filtering and removal, and normalization feature extraction to eliminate noise and dimensional differences in the original data and improve data reliability. The sensor health diagnosis module diagnoses the sensor's condition based on historical benchmark data and real-time measurement deviations, and calculates the sensor health index using a built-in multi-parameter degradation prediction model. ,when When the value is less than 0.7, the sensor enters a scrap warning state, triggering the switching of backup facilities. This model integrates the temperature-salinity coupled corrosion rate equation, the time-drift characteristics of piezoresistive elements and the dynamic error accumulation effect to achieve full life cycle health assessment. The dynamic sealing compensation module integrates a set of fluid dynamics equations to establish a coupled model of oil film thickness, pressure, rotational speed, and flow velocity. Based on real-time flow velocity data and thruster rotational speed commands, it performs feedforward compensation calculations. Its calculation formula is used to dynamically adjust the sealing cavity pressure to stabilize the oil film thickness, thus realizing the transition from a fixed threshold to dynamic feedforward compensation. The adaptive parameter adjustment module is based on the historical data of the system dynamic response and is equipped with an improved MPC (Model Predictive Control) algorithm framework. It uses the "system step response under standard flow field (flow velocity <0.5m / s)" as the performance baseline, establishes an "environmental disturbance intensity-controller parameter" mapping table, and calculates the system dynamic coupling key value in real time through the system dynamic coupling key value calculation formula to dynamically adjust the proportional coefficient (Kp), integral coefficient (Ki) and feedforward gain of the PID controller. The execution drive module receives and executes instructions from the adaptive parameter adjustment module and the dynamic seal compensation module to drive the propeller motor, buoyancy pump and proportional pressure valve. It contains a backup facility switching circuit and bus to ensure seamless switching to the backup facility switching unit when the main sensor or actuator fails. The closed-loop feedback and optimization module performs multi-source data fusion processing on the system output (actual thrust, attitude, and sealing status), compares it with the expected target, and feeds back the optimization results to the adaptive parameter adjustment module and the dynamic sealing compensation module to realize online fine-tuning and learning of model parameters. The human-machine interface module provides a visual monitoring platform that centrally displays the status of each module in the system, environmental data, alarm information and performance curves, and allows operators to switch modes, preset parameters and remotely intervene.

[0020] The environmental perception module constructs a three-dimensional environmental field model through four distributed acquisition stations at the front, rear, left, and right, providing the system with an environmental state baseline.

[0021] The forward distributed acquisition station uses an integrated temperature, salinity, and depth sensor (CTD) and a three-dimensional acoustic Doppler current profiler (ADCP) to monitor the flow field and environmental data in front of the thruster in real time, including temperature, salinity, pressure, three-dimensional velocity vector, and turbulence intensity index.

[0022] The distributed acquisition station uses an integrated temperature, salinity, and depth sensor (CTD) and a particle image velocimetry (PIV) probe to monitor wake field and vortex disturbance data in real time, including wake temperature gradient, salinity distribution, pressure pulsation, and vortex shedding frequency.

[0023] The left and right distributed acquisition stations monitor the corrosion environment and boundary layer data on the left and right sides in real time through a micro electrochemical sensor array and a laser Doppler velocimeter (LDV), including local corrosion potential, pH value, dissolved oxygen concentration and boundary layer flow velocity profile.

[0024] The advantages are: by dividing the environmental sensing module into four distributed acquisition stations (front, rear, left, and right), the front acquisition station collects flow field temperature, salinity, depth, and three-dimensional flow velocity data; the rear acquisition station collects wake vortex and pressure pulsation data; and the left / right acquisition stations collect corrosion potential and boundary layer flow velocity data. The data from the four acquisition stations are used to construct a three-dimensional environmental field model that includes three-dimensional flow field distribution, wake disturbance characteristics, and corrosion environment gradient. This enables the system to perceive the environmental status around the entire system, accurately quantify multi-source disturbances, and provide early warnings under extreme conditions. Ultimately, it provides high-precision environmental input parameters and operating condition identification benchmark data for the dynamic sealing compensation module and the adaptive parameter adjustment module.

[0025] The sensor health diagnosis module calculates the sensor health index. The calculation formula is as follows: In the formula, This represents the temperature drift, which is the difference between the sensor's current temperature reading and the median of the sensor's normal operating temperature range determined by historical data from the environmental sensing module. This represents the short-term standard deviation of the sensor output (reflecting the fluctuation of the readings, its value is calculated from the filtered sensor data of the most recent N sampling periods output by the statistical data preprocessing module). This indicates the maximum permissible temperature drift threshold, determined by the sensor calibration manual and failure analysis data. This represents the maximum permissible standard deviation threshold, determined by the static test noise level of the sensor under nominal conditions and the system's fault tolerance requirements. Represents the environmental stress factor (derived from temperature). ,salinity Corrosion potential The data was calculated comprehensively. This reflects the severity of the environmental degradation, among which... , , These are the design tolerance limits for the corresponding parameters. For reference corrosion potential, , , These are the weighting coefficients. , The higher the value, the more severe the environmental degradation. , , These represent the weights of temperature stability, output stability, and environmental tolerance in the health assessment, respectively. This formula is then used to evaluate the sensor's condition: When When the value is less than 0.7, the sensor enters a scrap warning state, triggering the backup facility switching unit in the sensor health diagnosis module to start the backup sensor and report the warning event to the management module.

[0026] The advantage is that it calculates the sensor's health index. It is used to assess the health status of sensors throughout their entire life cycle (normal / early warning / discard). When Lr < 0.7, it triggers a sensor discard warning, automatically starts the switch to a backup sensor and reports the warning event. This achieves the beneficial effects of identifying the risk of progressive sensor failure in advance, avoiding control deviations caused by sensor data distortion, and ensuring the continuous and stable operation of the system in extreme corrosive environments.

[0027] The dynamic sealing compensation module has a built-in operating condition identification unit (identifying whether the current operating condition is stable, accelerating, or turbulent) and a compensation calculation unit (performing compensation calculations). Its workflow is as follows: S1.1 The operating condition identification unit identifies whether the current propeller is in a stable, accelerating, or turbulent operating condition by using real-time flow velocity pulsation intensity, propeller speed change rate, and sealed cavity pressure spectrum characteristic data. S1.2 The compensation calculation unit identifies the current working condition based on the working condition identification unit, and performs feedforward compensation calculation based on real-time flow velocity data and thruster speed command. The calculation formula is as follows: In the formula, Indicates dynamic sealing compensation pressure. Indicates hydrostatic pressure; The velocity component perpendicular to the sealing surface is provided by the environmental sensing module. This represents the density of seawater (calculated in real time from data from temperature, salinity, and depth sensors). This indicates the real-time rotational speed of the thruster (feedback from the drive module). Indicates the design reference speed. The gain coefficient representing the ocean current disturbance (a core parameter of the control model) is determined by the geometry of the sealing surface and fluid-structure interaction simulation. This represents the speed compensation coefficient (derived from the dynamic characteristics of the sealing system). The real-time calculated value is applied to the sealing cavity through a proportional pressure valve to dynamically adjust the pressure in the sealing cavity, so as to stabilize the oil film thickness within the design range of ±0.05mm.

[0028] The advantages are: by dividing the dynamic sealing compensation module into a condition identification unit (identifying whether the current condition is stable, accelerated, or turbulent) and a compensation calculation unit (performing compensation calculations), the condition identification unit identifies the real-time operating condition of the propeller based on the flow velocity pulsation intensity, speed change rate, and pressure spectrum characteristics. The compensation calculation unit is used to match the sealing pressure compensation strategy under different operating conditions. When the condition is stable, a basic compensation coefficient is used; when the condition is accelerated, the speed-related compensation weight is increased; and when the condition is turbulent, the flow velocity coupling compensation is strengthened. Combined with the dynamic sealing compensation pressure calculation formula, the sealing cavity pressure is adjusted in real time, ultimately achieving the beneficial effects of stabilizing the oil film thickness within the design range of ±0.05mm, eliminating the hysteresis of fixed threshold compensation, and improving the reliability and durability of deep-water dynamic seals.

[0029] The adaptive parameter adjustment module incorporates a baseline comparison unit (comparing the deviation of the current response with the standard baseline), a key value calculation unit, and a parameter mapping unit. Its workflow is as follows: S2.1 In the baseline comparison unit: Real-time acquisition of the system's current response data (such as step response curve, thrust build-up time, attitude stabilization time) provided by the data preprocessing module, and comparison with the standard operating condition baseline (the ideal response model measured under a stable flow field and nominal load, including standard dead time, standard response time, and allowable overshoot) stored in the unit, calculate the deviation ratio of the current actual value relative to the standard baseline, generate a dynamic deviation vector, and output it as a primary quantitative indicator of the degree of environmental disturbance. S2.2 In the key value calculation unit, the dynamic deviation vector output by S2.1 and the real-time environmental data can be combined to calculate a single scalar value representing the current dynamic complexity of the system. ; S2.3, Parameter Mapping Unit Based on Adjust the PID controller parameters according to the calculated value range.

[0030] The key value calculation unit calculates the dynamic coupling key values ​​of the system based on the preprocessed data. The calculation formula is as follows: In the formula, The rate of change of attitude angle (from the data preprocessing module) reflects the intensity of the motion. This represents the multi-attitude angular coupling error norm (calculated from the current attitude and the decoupled target attitude). This represents the turbulence intensity (the high-frequency component of the flow velocity provided by the environmental sensing module). , , α, β, and γ represent the upper limits of each range, respectively, and represent the attitude dynamic weight, coupling error weight, and turbulence disturbance weight, respectively.

[0031] Parameter mapping unit according to Adjust the PID controller parameters as follows: (1) When When the value is less than 0.4, the system is determined to be in a quasi-steady state or a micro-disturbance condition. The parameter adjustment strategy is to maintain the PID controller as the core control unit and use the standard proportional coefficient. Standard integral coefficient With standard differential coefficients Run using standard baseline parameters; only the integral coefficients in the PID controller parameters are considered. A correction within the range of 0.95 to 1.05 (micro-correction) is performed to suppress steady-state errors caused by small disturbances, ensuring that the overshoot control target is ≤5% and the response time deviation is ≤±0.1s; (2) When 0.4≤ When the value is less than 0.7, the current system is determined to have moderate coupling and disturbance. The parameter adjustment strategy is as follows: use the PID controller as the main control unit and introduce a feedforward compensation element; dynamically adjust the PID parameters and increase the proportional coefficient. Enhance response speed to moderately reduce differential coefficients Suppressing moderately coupled oscillations; while the integral coefficient By adopting an integral separation strategy (cutting off the integral when the absolute value of the deviation is greater than the threshold), the overshoot is controlled to be ≤10%, and the attitude coupling error norm is reduced by more than 30%. (3) When When the value is ≥0.7, the current system is determined to be under strong nonlinear and strong turbulence conditions. The parameter adjustment strategy is as follows: switch the main controller to the improved model predictive control (MPC) algorithm, and the original PID controller becomes an auxiliary correction loop; the prediction time domain and control time domain of MPC are dynamically compressed according to the turbulence intensity; and significantly adjust the feedforward gain (based on the joint output of the dynamic sealing compensation module and the fluid dynamics model) to actively cope with the large lag and strong disturbances caused by strong turbulence (flow velocity >1.5m / s), and limit the overshoot to below 20%. The advantages are: the system response deviation is quantified through the baseline comparison unit, and the dynamic coupling key value is extracted through the key value calculation unit. The parameter mapping unit performs hierarchical parameter control. This workflow can realize a closed loop from deviation perception to complex quantification, and then to hierarchical parameter adaptive adjustment, so as to ensure the robustness and dynamic accuracy of the control system in complex nonlinear underwater environments. When implemented, the thruster thrust response delay is shortened to within 0.2s under all operating conditions and the overshoot is controlled to below 20% under strong turbulence conditions, ultimately achieving the beneficial effect of improving attitude stability accuracy.

[0032] 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 deep-water dynamic seal adaptive intelligent control system for underwater propulsion, characterized in that, It includes an environmental perception module, a data preprocessing module, a sensor health diagnosis module, a dynamic sealing compensation module, an adaptive adjustment module, an execution drive module, a feedback optimization module, and a human-machine interaction interface module; The environmental sensing module collects temperature, salinity, flow rate, and pressure data in real time through distributed acquisition stations; The data preprocessing module collects data and performs preprocessing such as multi-sensor spatiotemporal synchronization calibration, outlier filtering and removal, and normalized feature extraction to eliminate noise and dimensional differences in the original data. The sensor health diagnosis module diagnoses the sensor status based on historical benchmark data and real-time measurement deviations, and calculates the sensor health index by combining the built-in multi-parameter degradation prediction model. ,when When the value is less than 0.7, the sensor enters a scrap warning state, triggering a switchover of backup facilities; The dynamic sealing compensation module integrates a set of fluid dynamics equations, establishes an oil film thickness-pressure-rotation speed-flow velocity coupled model, and performs feedforward compensation calculations based on real-time flow velocity data and thruster rotation speed commands to dynamically adjust the sealing cavity pressure. The adaptive parameter adjustment module is based on historical data of the system's dynamic response and incorporates an improved MPC algorithm framework. It uses a standard flow field with a velocity <0.5 m / s as the performance baseline and calculates key dynamic coupling values ​​of the system in real time, which are then used as the proportional coefficient for the PID controller. Integral coefficient And dynamic adjustment of feedforward gain; The execution drive module receives and executes instructions from the adaptive parameter adjustment module and the dynamic sealing compensation module to drive the propeller motor, buoyancy pump and proportional pressure valve. It contains a backup facility switching circuit and bus, which can seamlessly switch to the built-in backup facility switching unit when the main sensor or actuator fails. The closed-loop feedback and optimization module performs multi-source data fusion processing on the system output and compares it with the expected target. The optimization result is fed back to the adaptive parameter adjustment module and the dynamic sealing compensation module to realize online fine-tuning and learning of model parameters. The human-machine interface module provides a visual monitoring platform that centrally displays the status of each module in the system, environmental data, alarm information and performance curves, and allows operators to switch modes, preset parameters and remotely intervene.

2. The intelligent control system for a deep-water dynamic seal adaptive underwater propulsion system according to claim 1, characterized in that: The environmental perception module constructs a three-dimensional environmental field model through four distributed acquisition stations (front, rear, left, and right) to provide an environmental state baseline for the system.

3. The intelligent control system for a deep-water dynamic seal adaptive underwater propulsion system according to claim 2, characterized in that: The forward distributed acquisition station uses an integrated temperature, salinity, and depth sensor and a three-dimensional acoustic Doppler velocity profiler to monitor the flow field and environmental data directly in front of the thruster in real time.

4. The intelligent control system for a deep-water dynamic seal adaptive underwater propulsion system according to claim 2, characterized in that: The post-distributed acquisition station uses an integrated temperature, salinity, and depth sensor and a particle image velocimetry system probe to monitor wake field and vortex disturbance data in real time.

5. The intelligent control system for a deep-water dynamic seal adaptive underwater propulsion system according to claim 2, characterized in that: The left and right distributed acquisition stations respectively use a micro electrochemical sensor array and a laser Doppler velocimeter to monitor the corrosion environment and boundary layer data on the left and right sides in real time.

6. The intelligent control system for a deep-water dynamic seal adaptive underwater propulsion system according to claim 1, characterized in that: The sensor health diagnosis module calculates the sensor health index. The calculation formula is as follows: In the formula, Indicates the amount of temperature drift. This represents the short-term standard deviation of the sensor output. This indicates the maximum allowable temperature drift threshold. This represents the maximum permissible standard deviation threshold. Indicates the environmental stress factor. , , These represent the weights of temperature stability, output stability, and environmental tolerance in the health assessment, respectively. This formula is then used to evaluate the sensor's condition: When When the value is less than 0.7, the sensor enters a scrap warning state, triggering the backup facility switching unit in the sensor health diagnosis module to start the backup sensor and report the warning event to the management module.

7. The intelligent control system for a deep-water dynamic seal adaptive underwater propulsion system according to claim 1, characterized in that: The dynamic sealing compensation module has a built-in working condition identification unit and a compensation calculation unit, and its workflow is as follows: S1.1 The operating condition identification unit identifies whether the current propeller is in a stable, accelerating, or turbulent operating condition by using real-time flow velocity pulsation intensity, propeller speed change rate, and sealed cavity pressure spectrum characteristic data. S1.2 The compensation calculation unit identifies the current working condition based on the working condition identification unit, and performs feedforward compensation calculation based on real-time flow velocity data and thruster speed command. The calculation formula is as follows: In the formula, Indicates dynamic sealing compensation pressure. Indicates hydrostatic pressure; This represents the velocity component perpendicular to the sealing surface. Indicates the density of seawater. Indicates the real-time rotational speed of the thruster. Indicates the design reference speed. This represents the ocean current disturbance gain coefficient. This represents the speed compensation coefficient. The real-time calculated value is applied to the sealing cavity through a proportional pressure valve to dynamically adjust the pressure in the sealing cavity, so as to stabilize the oil film thickness within the design range of ±0.05mm.

8. The intelligent control system for a deep-water dynamic seal adaptive underwater propulsion system according to claim 1, characterized in that: The adaptive parameter adjustment module integrates a baseline comparison unit, a key value calculation unit, and a parameter mapping unit. Its workflow is as follows: S2.1 In the baseline comparison unit: Real-time acquisition of the current system response data provided by the data preprocessing module, comparison with the standard operating condition baseline stored in the unit, calculation of the deviation ratio of the current actual value relative to the standard baseline, generation of dynamic deviation vector, and output as a primary quantitative indicator of the degree of environmental disturbance. S2.2 In the key value calculation unit, the dynamic deviation vector output by S2.1 and the real-time environmental data can be combined to calculate a single scalar value representing the current dynamic complexity of the system. ; S2.3, Parameter Mapping Unit Based on Adjust the PID controller parameters according to the calculated value range.

9. The intelligent control system for a deep-water dynamic seal adaptive underwater propulsion system according to claim 8, characterized in that: The key value calculation unit calculates the dynamically coupled key values ​​of the system based on the preprocessed data. The calculation formula is as follows: In the formula, Indicates the rate of change of attitude angle. This represents the norm of multi-attitude angular coupling error. Indicates turbulence intensity. , , α, β, and γ represent the upper limits of each range, respectively, and represent the attitude dynamic weight, coupling error weight, and turbulence disturbance weight, respectively.

10. The intelligent control system for a deep-water dynamic seal adaptive underwater propulsion system according to claim 8, characterized in that: The parameter mapping unit according to Adjust the PID controller parameters as follows: (1) When When the value is less than 0.4, the system is determined to be in a quasi-steady state or a micro-disturbance condition. The parameter adjustment strategy is to maintain the PID controller as the core control unit and use the standard proportional coefficient. Standard integral coefficient With standard differential coefficients The standard baseline parameters are used for operation, focusing only on the integral coefficients of the PID controller parameters. A correction ranging from 0.95 to 1.05 was made. (2) When 0.4≤ When the value is less than 0.7, the current system is determined to have moderate coupling and disturbance. The parameter adjustment strategy is as follows: use the PID controller as the main control unit and introduce a feedforward compensation element; dynamically adjust the PID parameters and increase the proportional coefficient. Enhance response speed to moderately reduce differential coefficients Suppressing moderately coupled oscillations; while the integral coefficient An integral separation strategy is adopted; (3) When When the value is ≥0.7, the current system is determined to be in a strongly nonlinear and strongly turbulent condition. The parameter adjustment strategy is to switch the main controller to the improved model predictive control algorithm and the original PID controller becomes an auxiliary correction loop.

Images