Dredge concentration control system based on adaptive algorithm

Abstract

The application relates to the technical field of control systems, and particularly discloses a dredging ship mud concentration control system based on an adaptive algorithm, which comprises a data acquisition module, which is used for collecting internal working condition data and external dynamic environment data of dredging operation in real time; a data processing module, which is in communication connection with the data acquisition module, is used for carrying out space-time alignment processing on the collected data and calculating an external environment correction factor based on a nonlinear model. The adaptive control module is integrated with the data acquisition, space-time alignment processing and external environment correction factor calculation, so that dynamic optimization and accurate risk management of the dredging ship mud concentration control are realized; the system can dynamically adjust risk threshold values and execution parameters according to real-time environmental parameters such as water flow and tides, so that the accuracy and timeliness of pipe blockage risk identification are remarkably improved; and the system inhibits measurement mutation caused by ship body transverse movement through motion decoupling and inertia filtering, so that the system stability is improved.

Description

Adaptive Algorithm-Based Dredging Vessel Mud Concentration Control System Technical Field

[0001] This invention relates to the field of dredging control technology, and in particular to a dredging vessel mud concentration control system based on an adaptive algorithm. Background Technology

[0002] In dredging operations, precise control of mud concentration is crucial for ensuring operational efficiency and safety. Traditional dredging vessel mud concentration control systems often rely on preset fixed thresholds and static models, such as using constant pressure gradients or vibration amplitude thresholds to assess pipe blockage risks, and setting control parameters for actuators such as mud pumps and cutterheads based on historical experience.

[0003] However, the dredging operation environment is highly dynamic. External factors such as changes in water flow velocity and direction, as well as tidal phase fluctuations, can significantly affect slurry flow characteristics and pipeline pressure distribution. Internal operating conditions, such as hull lateral movement and slurry flow fluctuations, can also introduce real-time disturbances. Fixed threshold models cannot adapt to these dynamic changes, leading to false alarms or missed alarms in harsh environments, while being overly conservative and limiting operational efficiency in stable environments. This conflict between rigid control strategies and complex and variable operating environments results in traditional systems exhibiting lag and inaccuracy in risk identification and control response, making it difficult to achieve optimal control.

[0004] Specifically, existing technologies lack the ability to collaboratively perceive and correct external environment and internal operating conditions in real time. For example, changes in water flow and tides can alter the flow resistance of mud in pipes, but traditional systems do not quantify these factors as correction factors and dynamically adjust risk thresholds. As a result, under conditions such as high-speed water flow or high tide, the system may not be able to improve risk sensitivity in time, leading to pipe blockage.

[0005] Conversely, in calm environments, excessively high thresholds lead to wasted production capacity. Furthermore, the lack of effective decoupling and suppression mechanisms for issues such as measurement interference caused by hull lateral movement and transient changes in physical parameters due to mud pump acceleration further exacerbates control oscillations and malfunctions. Summary of the Invention

[0006] This invention aims to at least partially solve one of the technical problems in related technologies. Therefore, the objective of this invention is to propose a dredging vessel mud concentration control system based on an adaptive algorithm to improve operational efficiency and system reliability.

[0007] To achieve the above objectives, a first aspect of the present invention proposes a dredging vessel mud concentration control system based on an adaptive algorithm, comprising:

[0008] The data acquisition module is used to collect internal working condition data and external dynamic environmental data of dredging operations in real time.

[0009] The data processing module is communicatively connected to the data acquisition module and is used to perform spatiotemporal alignment processing on the acquired data and calculate the external environment correction factor based on a nonlinear model.

[0010] An adaptive control module, which is communicatively connected to the data processing module, is used to dynamically adjust the risk level of mud concentration control according to the external environment correction factor and generate collaborative control commands.

[0011] The actuator module is communicatively connected to the adaptive control module and is used to perform mud concentration adjustment actions in response to the cooperative control command.

[0012] The process by which the adaptive control module dynamically adjusts the risk level of mud concentration control includes:

[0013] Obtain a preset risk identification threshold under static conditions, wherein the risk identification threshold includes a pressure gradient threshold and a vibration amplitude threshold;

[0014] The risk identification threshold is corrected by division using the external environment correction factor to obtain a dynamic threshold.

[0015] The real-time collected internal operating condition data is compared with the dynamic threshold to determine the current risk level, which corresponds to different pipe blockage probability intervals.

[0016] The adaptive control module generates execution parameters, including mud pump speed, suction port opening, and cutter speed, based on the risk level and external environment type.

[0017] To achieve the above objectives, a second aspect of the present invention proposes a method for controlling the mud concentration of a dredging vessel based on an adaptive algorithm, comprising the following steps:

[0018] First, the mud concentration and flow rate are collected in real time by ultrasonic concentration sensors and electromagnetic flow meters installed behind the water intake. At the same time, the pressure gradient of the pipeline is collected by pressure gradient sensors installed along the mud discharge pipe, and the water flow velocity and direction are collected by Doppler current meter installed at the bow of the ship. The tidal phase and rate are also obtained by connecting to the ship's GPS.

[0019] Next, the collected data is spatiotemporally aligned. The transmission time delay is calculated based on the slurry flow velocity in the pipeline and the sensor distance. The external environment data is then aligned with the delay, and the water flow direction is converted into the angle relative to the pipeline axis. Subsequently, the water flow correction factor is calculated through a three-dimensional nonlinear model, and the tidal correction factor is calculated through a multi-dimensional coupling model.

[0020] Then, the static risk threshold is dynamically corrected using a correction factor to obtain a new dynamic threshold. The risk level is determined by comparing the real-time operating data with the dynamic threshold.

[0021] Finally, based on the risk level and the type of external environment, control commands for the mud pump speed, suction port opening and cutter speed are generated through linkage calculation formula, and the actuator is driven to complete the concentration adjustment.

[0022] To achieve the above objectives, a third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein when the computer program is executed by the processor, it implements the above-described method for controlling dredging vessel mud concentration based on an adaptive algorithm.

[0023] The dredging vessel mud concentration control system based on adaptive algorithms in this invention integrates data acquisition, spatiotemporal alignment processing, external environment correction factor calculation, and adaptive control modules to achieve dynamic optimization and precise risk management of dredging vessel mud concentration control. The system can dynamically adjust risk thresholds and execution parameters according to real-time environmental parameters such as water flow and tides, significantly improving the accuracy and timeliness of pipe blockage risk identification.

[0024] Furthermore, the system suppresses measurement abrupt changes caused by hull lateral movement through motion decoupling and inertial filtering, enhancing system stability. Meanwhile, the oscillation suppression unit and self-learning iteration mechanism further ensure the smoothness and long-term adaptability of the control process, thereby effectively reducing the probability of pipe blockage, improving operational efficiency and system reliability in complex operating environments, while reducing the need for manual intervention. Attached Figure Description

[0025] Figure 1 is a schematic diagram of the implementation of the dredging vessel mud concentration control system based on adaptive algorithm provided by the present invention;

[0026] Figure 2 is a three-dimensional surface diagram showing the variation of the flow correction factor with the angle between flow velocity and direction in the dredging vessel mud concentration control system based on adaptive algorithm provided by the present invention.

[0027] Figure 3 is a schematic diagram of the dynamic threshold change trend under the action of different correction factors in the dredging vessel mud concentration control system based on adaptive algorithm provided by the present invention.

[0028] Figure 4 is a schematic diagram of the response curves of risk level and mud pump speed, suction port opening and cutter speed in the dredging vessel mud concentration control system based on adaptive algorithm provided by the present invention.

[0029] Figure 5 is a decision diagram of weighted similarity and matching threshold in the dredging vessel mud concentration control system based on adaptive algorithm provided by the present invention.

[0030] Figure 6 is an interference projection diagram of the hull lateral displacement vector in the Doppler direction in the dredging vessel mud concentration control system based on adaptive algorithm provided by the present invention.

[0031] Figure 7 is a schematic diagram comparing false alarms of the dredging vessel mud concentration control system based on adaptive algorithm provided by the present invention with and without oscillation suppression mechanism;

[0032] Figure 8 is a schematic diagram of the electronic device provided by the present invention. Detailed Implementation

[0033] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0034] The following description, with reference to the accompanying drawings, describes an adaptive algorithm-based dredging vessel mud concentration control system, method, and electronic equipment according to embodiments of the present invention.

[0035] Example 1:

[0036] Figure 1 is a schematic diagram illustrating the implementation of a dredging vessel mud concentration control system based on an adaptive algorithm according to an embodiment of the present invention, specifically including the following structure:

[0037] In this embodiment, the system includes a data acquisition module, a data processing module, an adaptive control module, and an actuator module. The overall architecture of this system is based on a modular design, aiming to improve operational efficiency and safety by dynamically adjusting control parameters in real time based on changes in the internal and external environment.

[0038] The system is structured as follows: The data acquisition module collects internal operating condition data and external dynamic environmental data during dredging operations. Internal operating condition data includes the dredging vessel's own operational parameters, such as mud concentration, flow rate, and pipeline pressure gradient. External dynamic environmental data covers water flow velocity, flow direction, tidal phase, and tidal rate. The data processing module communicates with the data acquisition module to perform spatiotemporal alignment processing on the collected data and calculate external environmental correction factors based on a nonlinear model. The adaptive control module communicates with the data processing module to dynamically adjust the risk level of mud concentration control based on the external environmental correction factors and generate coordinated control commands. The actuator module communicates with the adaptive control module to respond to coordinated control commands and execute mud concentration adjustment actions, including adjusting key parameters such as mud pump speed, suction port opening, and cutter head speed.

[0039] For example, the data acquisition module includes an internal operating condition acquisition unit and an external environment acquisition unit. The internal operating condition acquisition unit, located behind the intake, includes an ultrasonic concentration sensor and an electromagnetic flowmeter, used to collect real-time mud concentration data. and mud flow rate In addition, pressure gradient sensors are installed along the sludge discharge pipe to collect the pipeline pressure gradient. The external environment acquisition unit includes a Doppler current meter installed on the bow of the dredging vessel to collect water flow velocity data. and water flow direction The external environment acquisition unit also includes a tidal data parsing unit connected to the ship's GPS, used to obtain tidal phases. and tidal rate These data are sampled and transmitted at high frequency to ensure that the system can capture changes in the working environment in real time.

[0040] Optionally, the spatiotemporal alignment processing performed by the data processing module aims to resolve inconsistencies between internal and external data caused by differences in acquisition location and time. Spatiotemporal alignment processing includes adjustments based on the mud flow rate within the pipe. Physical distance from the sensor Calculate transmission time delay Its formula is ;

[0041] Among them, the mud flow rate inside the pipeline Real-time mud flow and pipe diameter The calculation yielded the following results: And the pipe diameter These are known design parameters, which are usually set according to the dredging vessel model and operational requirements.

[0042] After calculating the time delay, the system will process the data collected by the external environment acquisition unit according to... The timeline is delayed to align with data from the internal operating condition acquisition unit, ensuring synchronization of data comparison and analysis.

[0043] In addition, spatiotemporal alignment processing also includes adjusting the direction of water flow. Converted to the relative angle with respect to the pipe axis Its formula is ;

[0044] in, The ship's heading is obtained through the ship's navigation system; The direction of the pipeline axis is determined based on the structure and layout of the dredging vessel. This conversion ensures a direct correlation between the water flow direction data and the internal flow characteristics of the pipeline, improving the accuracy of subsequent correction factor calculations.

[0045] For example, external environment correction factors include water flow correction factors. The data processing module calculates using a three-dimensional nonlinear model. This model comprehensively considers the effects of flow velocity, direction, and pulsation; its formula can be expressed as:

[0046] ;

[0047] In the formula, , and Preset weighting coefficients are used to balance the influence of different factors, either by calibration with historical data or by setting them based on expert experience. Let be the flow velocity influence function, when the water flow velocity When it is greater than 1.5 meters per second, Follow The linear increase reflects the enhancing effect of high-speed water flow on mud flow; Let be the directional influence function, defined as This function reflects the influence of the angle between the water flow direction and the pipeline axis on the resistance of mud transport. The smaller the angle, the higher the function value, indicating that the flow direction is favorable for operation. The pulsation effect function is related to the water flow pulsation coefficient. Positive correlation; water flow pulsation coefficient Defined as the ratio of the difference between the maximum and minimum flow velocities per unit time to the average flow velocity, it is used to quantify flow stability; pulsation influence function. Linear or exponential forms are typically used to capture the contribution of pulsation to mud concentration fluctuations.

[0048] Figure 2 shows the variation trend of the flow correction factor under different flow velocities and directional angles. In Figure 2, the horizontal axis represents the flow velocity, ranging from 0.5 meters per second to 2.5 meters per second, the vertical axis represents the angle between the flow direction and the pipe axis, ranging from 0 degrees to 90 degrees, and the vertical axis represents the numerical value of the correction factor.

[0049] As can be observed from Figure 2, with the increase of water flow velocity, especially after exceeding 1.5 meters per second, the value of the correction factor begins to rise significantly, highlighting the enhancing effect of high flow velocity on the mud flow characteristics.

[0050] On the other hand, as the directional angle gradually increases, that is, when the water flow direction deviates from the pipe axis, the color in Figure 2 gradually darkens, and the corresponding correction factor value decreases significantly. This indicates that countercurrent or lateral flow will significantly reduce the transport efficiency. In the figure, this is shown as the area where the surface reaches its maximum value when the angle is close to 0 degrees, that is, the part with the brightest color and the highest correction factor.

[0051] The color change from dark blue to yellowish-green and then to light yellow represents a transition in the correction factor from low to high. This visual gradient helps to clearly identify high-risk and low-risk operating conditions. In control systems, a higher correction factor makes the system more sensitive to external disturbances, thus lowering the risk identification threshold and improving the sensitivity of pipe blockage warnings.

[0052] Optionally, the external environment correction factor also includes a tidal correction factor. The data processing module calculates through a multi-dimensional coupled model. This model integrates the effects of tidal phase, velocity, and water depth deviation, and its formula can be expressed as:

[0053] ;

[0054] In the formula, , and The preset weighting coefficients are determined through field experiments or simulation optimization. The phase influence function is given when the tidal phase... The value is greater than 1.0 during high tide and less than 1.0 during low tide, reflecting the effect of the tidal cycle on water depth and flow conditions; Let be the rate influence function, which varies with tidal rate. The increase reflects the auxiliary or hindering effect of tidal flow on mud transport; The water depth deviation influence function is calculated based on the ratio of the deviation between the real-time water depth and the baseline water depth during slack tide. This function is used to compensate for the indirect impact of water level changes caused by tides on pipeline pressure and concentration. A multi-dimensional coupling model ensures that the tidal correction factor can comprehensively capture environmental dynamics, providing reliable input for adaptive control.

[0055] Figure 3 shows the changing trend of dynamic threshold under different correction factors. The horizontal axis in the figure represents the variation range of the water flow correction factor (from 0.8 to 1.5) and the tidal correction factor (from 0.9 to 1.2), and the vertical axis represents the corresponding dynamic threshold.

[0056] In Figure 3, the blue solid line represents the dynamic threshold change trend under the influence of the flow correction factor. As the flow correction factor increases, the dynamic threshold gradually decreases. This indicates that at higher flow velocities, the system's sensitivity to pipe blockage risk increases, necessitating adjustment of the dynamic threshold to lower the risk threshold and thus improve the system's response sensitivity to risk. Specifically, as the flow correction factor increases from 0.8 to 1.5, the dynamic threshold decreases significantly, especially when the flow correction factor approaches 1.5, where the threshold reduction is particularly pronounced.

[0057] The red dashed line represents the trend of dynamic threshold change under the influence of the tidal correction factor. Unlike the flow correction factor, the change in the tidal correction factor has a more gradual impact on the dynamic threshold. Figure 3 shows that when the tidal correction factor increases from 0.9 to 1.2, the change in the dynamic threshold is very small. This indicates that tidal factors do not significantly affect the system's risk perception and threshold adjustment as much as flow factors. Especially when the tidal correction factor is close to 1.0, the dynamic threshold is almost unaffected and remains near a stable value.

[0058] As the flow correction factor increases, the dynamic threshold decreases. This change reflects the effect of the flow correction factor; that is, in high-speed flow environments, the system needs to improve its sensitivity to risks. Therefore, a lower dynamic threshold means that the system is more likely to issue risk warnings and prevent accidents such as pipe blockage. The significant decrease in Figure 3, especially when the flow correction factor is close to 1.5, indicates that as the flow velocity increases, the system's risk control becomes more refined and flexible.

[0059] In contrast, the tidal correction factor has a smaller impact. Since the effects of tides on mud flow and pipeline pressure are relatively stable, the dynamic threshold is less affected by the tidal correction factor, and the changes shown in Figure 3 are gradual and small. This indicates that the system can maintain a relatively constant risk threshold in an environment with relatively stable tidal changes, avoiding over-adjustment or false alarms caused by tidal variations.

[0060] For example, the process by which the adaptive control module dynamically adjusts the risk level of mud concentration control includes several steps:

[0061] First, the system acquires preset risk identification thresholds under static conditions, including pressure gradient thresholds and vibration amplitude thresholds, which are set based on historical operation data and safety standards.

[0062] Secondly, the risk identification threshold is corrected by division using an external environmental correction factor to obtain a dynamic threshold. The specific correction formula is as follows: ;in, For micro-feature thresholds in static environments, This is the corrected dynamic threshold.

[0063] when When the value is greater than 1, the dynamic threshold decreases, the system's sensitivity to risk increases, and it is suitable for harsh environments such as high-speed water flow or high tide to provide early warning of pipe blockage risks; when When the value is less than 1, the dynamic threshold increases, the system's tolerance for risk increases, it is suitable for calm environments, and unnecessary intervention is avoided;

[0064] The system then compares the real-time collected internal operating data with dynamic thresholds to determine the current risk level. Each risk level corresponds to a different probability range for pipe blockage, for example, integers from one to five, representing safe, low-risk, medium-risk, high-risk, and critical-risk, respectively. This level classification is based on statistical models or machine learning algorithms to ensure the accuracy of risk identification.

[0065] Optionally, the adaptive control module generates execution parameters including mud pump speed, suction port opening, and cutter head speed based on risk level and external environment type. Among these, the mud pump speed... Water intake opening and auger speed The calculation is performed using a linkage formula, and the specific formula can be expressed as follows:

[0066] ;

[0067] ;

[0068] ;

[0069] In the formula, , and These are the basic mud pump speed, the basic suction port opening, and the basic cutter speed, which are set according to the dredging vessel's design parameters and operational objectives. This represents the current risk level. The exponents and coefficients in the linkage formula are optimized through experimental data to ensure that the execution parameters can dynamically respond to environmental changes and risk conditions. For example, reducing the mud pump speed to slow down the flow rate during high-risk periods, or increasing the cutter speed to enhance digging efficiency during low-risk periods.

[0070] Figure 4 shows the response trends of risk level with mud pump speed, suction port opening, and cutter head speed. In Figure 4, the horizontal axis (X-axis) represents the risk level, ranging from 1 (safe) to 5 (critical risk), reflecting the control system's response to different risk states. The vertical axis (Y-axis) represents the magnitudes of the system's three control parameters: mud pump speed, suction port opening, and cutter head speed.

[0071] The solid blue line (mud pump speed curve) represents the trend of mud pump speed as the risk level increases. As the risk level increases from 1 to 5, the mud pump speed gradually increases. This indicates that at higher risk levels, the system increases the mud flow rate by increasing the mud pump speed, thereby more effectively addressing the risk of pipe blockage.

[0072] The green dashed line (suction port opening curve) represents the trend of the suction port opening. As the risk level increases, the suction port opening gradually increases, reflecting that under high-risk conditions, the system needs to increase the amount of mud sucked in to ensure the normal operation of the system and avoid equipment damage caused by insufficient suction.

[0073] The red dotted line (cutter speed curve) indicates the trend of cutter speed change. As the risk level increases, the cutter speed also increases, indicating that in high-risk environments, the cutter needs a higher speed to improve operating efficiency and ensure stable mud excavation and transportation.

[0074] In summary, increasing the mud pump speed at high-risk levels is to better address the risk of pipe blockage. As the risk level increases from 1 to 5, the mud pump speed gradually increases, ensuring the system can respond effectively during high-risk situations. The suction port opening also increases with the risk level, indicating that when facing a higher risk of pipe blockage, the system increases the amount of mud drawn in by increasing the suction port opening, maintaining stable system operation. Furthermore, increasing the cutter head speed helps improve operational efficiency, especially in high-risk environments, by enhancing the mud digging and conveying capabilities.

[0075] For example, the workflow of this system begins with real-time monitoring by the data acquisition module, wherein: the internal operating condition acquisition unit continuously collects mud concentration data through an ultrasonic concentration sensor and an electromagnetic flowmeter. With traffic Pressure gradient sensors monitor pipeline pressure gradients. The external environment acquisition unit obtains water flow velocity through a Doppler current meter. With direction The tidal data analysis unit extracts tidal phases from GPS data. With rate All data is transmitted to the data processing module via wireless or wired communication; the data processing module first performs spatiotemporal alignment processing and calculates the time delay. And adjust data synchronization, while changing the water flow direction to a relative angle. Subsequently, the module calculates the flow correction factor. With tidal correction factor We use a three-dimensional nonlinear model and a multi-dimensional coupled model to ensure that the correction factor accurately reflects the environmental dynamics.

[0076] Optionally, the data processing module transmits the correction factor to the adaptive control module. The adaptive control module dynamically adjusts the risk identification threshold based on the correction factor, generating a dynamic threshold. The system compares real-time internal operating data with dynamic thresholds to determine the risk level. .

[0077] Based on the risk level and the type of external environment, the module calculates execution parameters, including the mud pump speed, using a linkage formula. Water intake opening and auger speed These parameters are sent to the actuator module via control commands. The actuator module includes a mud pump driver, a suction inlet regulating valve, and a cutter motor. These components respond to the commands by performing regulatory actions, such as adjusting the mud pump speed to control the flow rate or modifying the suction inlet opening to optimize suction efficiency. The entire process operates in a cyclical manner, achieving closed-loop control.

[0078] For example, the system of this embodiment can be further extended in specific applications. For instance, the sensors in the data acquisition module can employ a redundant design to improve reliability, and the internal operating condition acquisition unit can add a temperature sensor to monitor mud temperature to compensate for the influence of temperature on concentration measurement. The external environment acquisition unit can integrate a weather radar to acquire wind speed and direction data, expanding the dimensions of the environmental correction factor. The spatiotemporal alignment processing of the data processing module can incorporate machine learning algorithms to optimize time delay calculations, especially under unsteady flow conditions. The risk level determination of the adaptive control module can incorporate fuzzy logic to handle uncertain data and improve the system's decision-making ability under edge conditions.

[0079] Optionally, the system implementation relies on the collaboration of hardware and software. The hardware platform includes an embedded processor, high-precision sensors, and communication interfaces, while the software layer implements data acquisition, processing, and control algorithms. This embodiment can adopt a distributed architecture, where the data acquisition module is deployed at key points on the dredging vessel, the data processing module and adaptive control module run on a central server, and the actuator module is connected via a fieldbus. The software algorithms can be implemented in C++ or Python, and an integrated real-time operating system ensures fast response. Furthermore, the system can be equipped with a human-machine interface to display real-time data, risk levels, and control parameters, facilitating operator monitoring and intervention.

[0080] For example, the advantages of this system lie in its adaptability and accuracy. Through dynamic threshold correction, the system can adapt to environmental changes such as water flow and tides, reducing false alarms and missed alarms, and improving operational safety. Linked execution parameter generation ensures coordinated control actions, avoiding equipment conflicts or efficiency losses. Spatiotemporal alignment processing and correction factor calculation enhance data consistency and improve model prediction accuracy. Overall, this system is suitable for various dredging scenarios, such as port dredging, river clearing, and seabed excavation, maintaining efficient and stable operation in complex environments.

[0081] Optionally, to verify the effectiveness of this embodiment, simulation tests and field trials can be conducted. Simulation tests use a Computational-Fluid-Dynamics model to simulate mud flow under different water flow and tidal conditions, verifying the rationality of the correction factor and dynamic threshold. Field trials deploy the system in actual dredging operations, collect data, and analyze the incidence of blockage events and system response time. The test results can be used to further optimize the weighting coefficients. , , , , , With basic parameters , , This ensures the system's generalization ability.

[0082] For example, this embodiment can also integrate safety mechanisms. For instance, when the risk level reaches a critical threshold, the system can trigger an emergency shutdown protocol, suspend operations, and issue an alarm. The data processing module can incorporate anomaly detection algorithms to identify sensor malfunctions or data loss and initiate backup procedures. The adaptive control module can record historical decisions and effects for subsequent analysis and improvement. These enhancements further improve the system's robustness and usability.

[0083] Optionally, regarding implementation details, the pipe diameter... The diameter can be adjusted according to the stage of operation; for example, a larger diameter can be selected to reduce resistance when transporting high-concentration mud. The ship's course... With respect to the pipe axis The direction data can be acquired in real time via an inertial measurement unit, ensuring accuracy. Water flow pulsation coefficient. The calculations are based on a sliding time window, for example, one minute, to calculate the maximum, minimum, and average velocities, ensuring the real-time nature of the pulsation effects. Tidal phase It can be resolved from publicly available tide tables or measured directly by local tide sensors.

[0084] For example, the computational model of this system can be further refined. For instance, the flow velocity influence function... exist For speeds greater than 1.5 meters per second, a piecewise linear form can be used to more accurately capture the nonlinear effects in the high-speed region. Directional influence function. An asymmetry coefficient can be introduced to handle the unequal effects of countercurrent and downstream flow. (Pulsation influence function) Based on turbulence theory, square root or logarithmic relationships can be used to better fit actual data. Phase influence function. It can be combined with a sine model to simulate the smooth transition of tidal cycles.

[0085] Optionally, the control strategy in this embodiment emphasizes real-time optimization. The adaptive control module can be embedded with a predictive controller, using external environmental correction factors to predict future risk trends and adjust execution parameters in advance. For example, during tidal transition periods, the system can reduce the mud pump speed in advance to prevent pressure fluctuations. The response time of the actuator module can be ensured by calibrating actuator dynamics, such as the acceleration curve of the mud pump driver and the torque control of the cutterhead motor, to avoid mechanical stress.

[0086] For example, this system offers added value in terms of energy efficiency. By dynamically adjusting execution parameters, the system can reduce equipment speed and energy consumption during low-risk periods, and optimize operation during high-risk periods to avoid downtime losses caused by pipe blockage. The low-power design of the data acquisition module extends sensor life, especially in underwater environments. Overall, this system supports green dredging principles, balancing operational efficiency with resource consumption.

[0087] Optionally, the deployment of this embodiment can be extended to multi-vessel collaborative operations. In large-scale dredging projects, multiple dredging vessels can share environmental data and control strategies via a network, and the data processing module can integrate cloud computing resources to process large-scale data. The adaptive control module can employ distributed decision-making algorithms to coordinate the actions of each vessel and avoid conflicts. This extension enhances the system's scalability and application scope.

[0088] For example, the maintenance and upgrades of this system are based on a modular design. The sensors in the data acquisition module can be calibrated periodically, and their accuracy adjusted using standard tools. The model parameters in the data processing module can be optimized through software updates, without requiring hardware replacement. The algorithm in the adaptive control module can be retrained based on new data to maintain its advanced capabilities. The actuator module's drives can be equipped with diagnostic functions to warn of potential faults.

[0089] Example 2:

[0090] This embodiment provides a self-learning iterative module, which, as an important component of the dredging vessel's mud concentration control system, aims to achieve continuous optimization and experience reuse of the system's control strategy by constructing a case library, calculating similarity, and updating model parameters based on feedback. Specifically, it includes the following:

[0091] In this embodiment, the core function of the self-learning iteration module is to improve the system's adaptability and efficiency in complex operating environments through historical data accumulation and intelligent matching. This module works in conjunction with the data acquisition module, data processing module, and adaptive control module in the system to construct a closed-loop learning system.

[0092] The self-learning iteration module is first used to build a case library, which stores historical cases including external environment correction factors, internal operating condition data, and corresponding control action parameters. External environment correction factors include water flow correction factors. and tidal correction factor These factors were calculated using a three-dimensional nonlinear model and a multi-dimensional coupled model within the data processing module. Internal operating data included real-time mud concentration. Real-time mud flow rate Pipeline pressure gradient And the dredging vessel's own operational parameters. Control parameters include mud pump speed. Water intake opening and auger speed These parameters are generated by the adaptive control module and executed by the actuator module. The case library is implemented in the form of a structured database, such as a relational database or a time-series database, to ensure efficient data storage and retrieval.

[0093] For example, the case library is built based on continuous data recordings of the system during normal operation. Each historical case corresponds to the operational status at a specific point in time, including a complete set of external environmental correction factors, internal operating condition data, and control action parameters.

[0094] For example, a historical case might involve water flow velocity. The water flow is 1.2 meters per second in the direction of water flow. 30 degrees, tidal phase Mid-high tide, tidal rate The flow correction factor calculated at 0.3 meters per second. and tidal correction factor Simultaneously record the mud concentration at that moment. 1200 kg per cubic meter, mud flow rate 0.5 cubic meters per second, pipeline pressure gradient The pressure is 5000 Pa per meter, and the corresponding control parameters include the mud pump speed. 1200 revolutions per minute, suction port opening 60% of the auger speed The speed is 800 revolutions per minute. The case library is also updated regularly, for example, by adding new cases and removing outdated data on an hourly or daily basis, in order to maintain the timeliness and representativeness of the cases in the library.

[0095] Optionally, the self-learning iterative module achieves intelligent matching by calculating the external environment similarity and internal working condition similarity between the current working condition and historical cases in the case library. The external environment similarity focuses on comparing the current external environment correction factor with the corresponding factor in historical cases, and its calculation is based on metrics such as Euclidean distance or cosine similarity.

[0096] Specifically, external environment similarity It can be calculated using the following formula:

[0097] ;

[0098] In the formula, and This indicates the current flow correction factor and tidal correction factor; and Indicates the corresponding factor in historical cases; and This represents the historical range of values ​​for these factors, used for normalization.

[0099] Internal operating condition similarity Then, the current internal operating data is compared with the corresponding data in historical cases. The calculation can be similar to using a distance metric, for example, by using the following formula:

[0100] ;

[0101] In the formula, , , This indicates the current mud concentration, flow rate, and pressure gradient; , , This represents the corresponding data from historical cases; , , This indicates the range of values ​​for these data.

[0102] For example, the self-learning iterative module combines the similarity of the external environment and the similarity of the internal operating conditions into a weighted matching similarity, which is used to decide whether to invoke historical control action parameters.

[0103] Weighted matching similarity Calculated using the following formula:

[0104] ;

[0105] In the formula, and For the preset weighting coefficients, satisfy The weighting coefficients can be adjusted according to operational needs; for example, in areas with drastic environmental changes, they can be set... Set to 0.7 to emphasize matching with the external environment; in scenarios with stable operating conditions, set... The value is set to 0.6 to emphasize internal operating condition consistency.

[0106] When weighted matching similarity When a preset threshold is reached, such as 0.8 or higher, the system considers the current operating condition to be highly similar to a historical case and directly calls the control action parameters stored in that historical case, including the mud pump speed. Water intake opening and auger speed These parameters are then applied to the actuator module. This mechanism avoids redundant calculations and improves system response speed, especially in jobs with high real-time requirements.

[0107] Figure 5 illustrates the relationship between the weighted similarity score and the matching threshold used by the system in intelligent matching judgment. The horizontal axis represents the weighted similarity score, which is a comprehensive similarity score calculated based on the fusion of multi-dimensional external environmental parameters and internal operating condition indicators, ranging from 0 to 1. The vertical axis represents the matching threshold, used to determine whether the current operating condition is sufficiently close to historical cases to trigger case reuse or issue an early warning.

[0108] The blue curve in Figure 5 exhibits a typical S-shaped decision-making pattern, indicating that the system requires a higher threshold to avoid false matches when the similarity is low. However, as the similarity gradually increases and exceeds a certain critical value, the matching threshold tends to stabilize, and the system is more inclined to automatically adopt historical strategies. For example, when the similarity is around 0.5, the matching threshold rises rapidly to above 0.5, indicating that the system is more cautious about moderately similar cases; while when the similarity is close to 0.9, the matching threshold approaches 1, indicating that the system highly trusts the matching accuracy.

[0109] Optionally, the self-learning iterative module periodically updates the calculation model parameters of the external environment correction factor based on execution performance feedback to optimize future decisions. Execution performance feedback is obtained by monitoring operational indicators after the execution of control action parameters, such as pipe blockage rate, slurry transport efficiency, or equipment energy consumption. Feedback data is quantitatively evaluated, for example, using performance scoring. Its calculation is based on a weighted average of multiple indicators, such as ,in, Indicates the probability of pipe blockage; Indicates mud flow efficiency; This represents normalized energy consumption; , , These are the weighting coefficients.

[0110] Based on performance evaluation, the system periodically reviews historical cases in the case library, such as weekly or monthly, and updates the calculation model parameters of the external environment correction factor, including the weighting coefficients in the three-dimensional nonlinear model. , , Weight coefficients in multidimensional coupling models , , The update process can employ gradient descent or a genetic algorithm to minimize the error in the performance score. For example, for the water flow correction factor... The computational model, parameters The update formula is ,in, The learning rate; To score the performance of the parameters The partial derivatives are obtained by regression analysis using historical data.

[0111] For example, the implementation of the self-learning iterative module can be enhanced by combining machine learning techniques. For instance, the construction of the case library can incorporate clustering algorithms such as K-means clustering to automatically group similar cases and reduce storage redundancy. Similarity calculation can employ deep learning models, such as convolutional neural networks, to process multimodal data, including sensor readings and image information. The threshold for weighted matching similarity can be dynamically adjusted and automatically optimized based on historical matching success rates; for example, the threshold can be lowered to expand the scope of calls when the recent matching success rate is high. Model parameter updates can integrate reinforcement learning frameworks, treating the operating environment as a state space, control actions as an action space, and performance scores as a reward function, achieving long-term policy optimization. These enhancements improve the module's intelligence level, enabling it to quickly converge to optimal control in unknown environments.

[0112] Optionally, the deployment of the self-learning iterative module considers the balance between computational resources and real-time performance. In resource-constrained embedded systems, a simplified version of the module can be used, for example, storing only the most recent one hundred historical cases and using linear similarity calculation. On high-performance servers, the module can scale to distributed storage and parallel computing to handle tens of thousands of cases. The case database can be stored encrypted to ensure job information security. The module is also equipped with a logging function to record all matching decisions and update operations, facilitating fault diagnosis and performance analysis. In addition, the module supports manual intervention, allowing operators to mark specific cases as preferred or disabled to incorporate expert experience.

[0113] For example, the self-learning iterative module demonstrates its value in specific operational scenarios. In estuarine areas with frequent tidal changes, the system quickly retrieves efficient control parameters for high tide periods by matching cases from a database, avoiding delays caused by recalculation. Under conditions of large fluctuations in mud concentration, the module retrieves historical parameters based on similar internal operating data to stabilize the delivery process. By periodically updating model parameters, the system gradually adapts to long-term environmental changes such as seasonal water flow patterns, reducing the need for manual parameter adjustments. Overall, this module enhances the system's autonomy and reliability, while reducing operational costs and risks.

[0114] Example 3:

[0115] Embodiment 3 of this invention provides a motion decoupling and inertial filtering unit. This unit, as a crucial component of the data processing module in the dredging vessel's mud concentration control system, aims to improve the system's stability and accuracy in dynamic operating environments by decoupling the interference of the vessel's lateral motion on water flow measurement and applying inertial filtering to suppress abrupt changes in the correction factor. It includes the following:

[0116] In this embodiment, the core function of the motion decoupling and inertial filtering unit is to solve the problem of Doppler current meter measurement distortion caused by hull lateral movement during dredging operations. During dredging operations, the hull frequently moves laterally using a lateral winch to adjust its working position. This motion introduces additional velocity components into the Doppler current meter's detection data, leading to distortion of the water flow velocity. and water flow direction The measured values ​​deviate from the actual environmental water flow. This unit decouples the influence of the ship's lateral motion in real time and uses inertial filtering to smooth the output correction factor, ensuring the water flow correction factor is accurate. The calculations are based on real-world environmental conditions, not ship motion artifacts. Water flow correction factor. These are key parameters calculated by the data processing module using a three-dimensional nonlinear model. They are used for subsequent dynamic threshold adjustment and execution parameter generation, and their accuracy is directly related to the reliability of system risk identification and control commands.

[0117] For example, the motion decoupling and inertial filtering unit first acquires the real-time transverse winch rotation speed fed back by the actuator. The transverse winch is a key piece of equipment on the dredging vessel for controlling the lateral movement of the hull; its rotation speed is monitored in real time by an encoder or sensor and transmitted to the data processing module. This rotation speed is denoted as... The unit is radians per second. Combining the ship's geometric parameters, including the winch installation location, the coordinates of the bow reference point, and the ship's dimensions, the element calculates the transverse linear velocity vector at the bow. This transverse linear velocity vector is denoted as... Its size is determined by the formula Calculate, where, The effective radius is determined based on the winch drum diameter and transmission ratio; the direction is defined according to the winch layout and the ship's coordinate system, and is usually perpendicular to the ship's longitudinal axis. The ship's coordinate system has the bow as its origin, with the longitudinal axis along the ship's direction and the transverse axis along the lateral movement direction.

[0118] Through geometric transformation The components in the global coordinate system can be represented as and ,in, The lateral direction angle is set based on the winch operating parameters.

[0119] Optionally, the unit calculates the projection component of the transverse linear velocity vector in the detection direction of the Doppler current meter. The Doppler current meter is mounted on the bow of the dredging vessel, and its detection direction is denoted as... This direction is determined by the current meter's installation angle and calibration parameters, and typically points forward or sideways to capture ambient water flow. The projection component is calculated based on the vector dot product, using the formula: ,in, This represents the velocity contribution of the lateral motion in the detection direction. This projection component reflects the direct interference of the hull's lateral movement on the current meter reading. Subsequently, the unit calculates this projection component in the water flow velocity... The proportion of motion interference is defined as the percentage of motion interference.

[0120] For example, motion interference is denoted as The calculation formula is: Motion disturbance quantifies the degree of influence of hull lateral movement on water flow measurement, with a value range between 0 and 1, where 0 indicates no interference and 1 indicates that the interference completely dominates the measurement.

[0121] For example, the motion decoupling and inertial filtering unit determines whether to trigger the inertial filtering mode based on the motion disturbance level. When the motion disturbance level... When the interference exceeds a preset threshold, the system enters inertial filtering mode. The interference threshold is denoted as... Its value is set through experimental calibration, for example, 0.1 or 0.2, which means that when the contribution of lateral motion exceeds 10% or 20% of the water flow velocity, the system considers the measurement reliability to be reduced and filtering needs to be enabled.

[0122] Figure 6 shows the interference projection of the ship's lateral displacement vector in the Doppler direction. The horizontal axis represents the angle (in degrees) between the ship's lateral displacement direction and the direction of the Doppler current meter, ranging from 0 degrees to 360 degrees, reflecting the relative angular change between the ship's lateral displacement and the current meter's measurement direction. The vertical axis represents the projection value (in m / s), indicating the projection intensity of the ship's lateral displacement velocity in the Doppler direction.

[0123] The blue curve in Figure 6 illustrates the interference projection of the ship's lateral displacement, reflecting the impact of lateral displacement on the current meter measurement at different angles. As can be seen from the figure, when the angle is close to 150° to 200°, the projection value reaches its lowest point, approximately -2.0, indicating that the direction of lateral displacement is almost opposite to the Doppler measurement direction, resulting in the largest interference component. Conversely, when the angle is close to 90° or 270°, the projection value approaches zero, indicating that the direction of lateral displacement is almost perpendicular to the measurement direction, minimizing its impact on current velocity measurement.

[0124] In inertial filtering mode, the final flow correction factor output by the unit is not directly calculated using the instantaneous value from the formula in Example 1, but rather by weighting the current instantaneous calculated value with the final output value from the previous instant. The instantaneous calculated value is denoted as... The result is obtained in real time through a three-dimensional nonlinear model; the final output value at the previous moment is denoted as... The data is stored in the system cache. The weighted calculation aims to smooth the correction factor sequence and suppress abrupt changes caused by rapid hull lateral movement.

[0125] Optionally, in the specific implementation of the weighted calculation, the weight coefficient of the instantaneous calculated value at the current moment is negatively correlated with the motion disturbance degree. The weight coefficient is denoted as... Its value is based on the motion disturbance degree. Dynamic adjustment. For example, weighting coefficients can be defined using linear functions, such as...

[0126] ;

[0127] In the formula, This is a scaling factor, usually set to one or adjusted according to the system response characteristics.

[0128] The final flow correction factor The output value is calculated using the following formula:

[0129] ;

[0130] In the formula, when the motion disturbance degree is high, When the noise level is low, the system relies more on historical values ​​to avoid instantaneous fluctuations; when the noise level is low, Increase the value, and the system will prioritize using the new value to maintain responsiveness.

[0131] This mechanism ensures that the correction factor remains stable during periods of intense hull motion and is updated promptly during periods of stillness or slow movement. For example, the implementation of motion decoupling and inertial filtering units can be further extended to enhance robustness. For instance, the calculation of the lateral velocity vector can integrate inertial measurement unit data to compensate for the effects of hull roll or pitch. The definition of motion disturbance can incorporate multi-directional projection; if the Doppler current meter has multiple probe beams, the projection components on each beam are calculated, and the maximum value is taken as the disturbance. The inertial filtering mode can incorporate a time decay factor to ensure that the system can gradually recover to real-time calculations after prolonged lateral movements. The weighting coefficients in the weighted calculation can also employ nonlinear functions, such as exponential functions.

[0132] ;

[0133] In the formula, This is the motion interference attenuation coefficient, used to more precisely control the smoothness.

[0134] Optionally, the workflow of this unit in system integration includes data acquisition, decoupling calculation, disturbance assessment, and filtering output. In the data acquisition phase, the unit obtains the transverse winch rotation speed from the actuator module and Doppler current meter readings and hull attitude data from the data acquisition module. In the decoupling calculation phase, the unit calculates the transverse linear velocity vector and its projected components based on a geometric model. In the disturbance assessment phase, the unit calculates the motion disturbance degree and compares it with a threshold to determine the filtering mode. In the filtering output phase, the unit selects to directly output the instantaneous value or a weighted value according to the mode and transmits the final flow correction factor to the adaptive control module. The entire process runs at a high-frequency loop, for example, ten times per second, to ensure real-time performance.

[0135] For example, the advantage of the motion decoupling and inertial filtering unit lies in its ability to significantly improve the reliability of the system under dynamic operations. When a dredging vessel makes frequent lateral adjustments, traditional systems may misinterpret the data due to measurement interference, such as mistaking the vessel's motion for changes in the ambient water flow, leading to unnecessary control adjustments or false alarms. This unit effectively isolates these artifacts through decoupling and filtering, ensuring the accuracy of the water flow correction factor. It accurately reflects environmental conditions, thereby improving the accuracy of risk level assessment. Simultaneously, inertial filtering suppresses abrupt changes in correction factors, avoiding frequent start-ups and shutdowns of actuators such as mud pumps and cutterheads, thus reducing mechanical wear and energy consumption.

[0136] Optionally, the implementation details of this unit include sensor calibration and parameter initialization. The transverse winch speed sensor needs to be calibrated periodically to ensure accurate speed measurement; hull geometry parameters such as effective radius and heading angle should be set by measurement during system deployment. The detection direction of the Doppler current meter needs to be determined through calibration experiments and aligned with the hull coordinate system. Motion disturbance threshold. The threshold can be adjusted according to the task type. For example, a lower threshold can be set for detailed tasks to improve sensitivity, while a higher threshold can be set for rough tasks to enhance stability. Parameters in the weighting coefficient calculation include scaling factors. or motion interference attenuation coefficient It can be optimized through historical data or tuned based on simulation tests.

[0137] For example, the motion decoupling and inertial filtering unit can collaborate with other modules to extend its functionality. For instance, combined with a self-learning iteration module, the unit can store historical motion disturbance data in a case library for environmental context evaluation during similarity matching. In conjunction with the oscillation suppression unit, it can temporarily adjust tolerance during inertial filtering to avoid false triggering of protection mechanisms due to filtering delays. Furthermore, the unit can be equipped with a diagnostic interface to output disturbance indexes in real time, allowing operators to monitor the impact of hull motion and intervene manually when necessary.

[0138] Example 4:

[0139] Embodiment 4 of this invention provides an oscillation suppression unit. This unit, as an important component of the adaptive control module in the dredging vessel's mud concentration control system, aims to resolve the conflict between the low dynamic threshold and the natural increase in physical parameters caused by the acceleration of the mud pump during the production recovery phase, thereby avoiding secondary false alarms and improving the system's stability during transient processes. It includes the following:

[0140] In this embodiment, the core function of the oscillation suppression unit is to balance the inherent contradiction between system risk sensitivity and equipment dynamic response. During dredging operations, when the system recovers from a low-capacity state, the adaptive control module may issue a mud pump acceleration command to improve mud transport efficiency. However, the increase in mud pump speed will naturally cause an increase in physical parameters such as pipeline pressure and flow rate. Since the system uses dynamic thresholds for risk identification, the normal increase of these parameters may be misjudged as abnormal risks, triggering false alarms or unnecessary control interventions. The oscillation suppression unit introduces a transient tolerance mechanism to temporarily adjust the execution threshold during mud pump acceleration, accommodating the natural changes in physical parameters. Simultaneously, it restores the original threshold after the speed stabilizes, ensuring the accuracy of risk identification and the smoothness of the control process. This unit works in conjunction with other components of the adaptive control module, such as the dynamic threshold calculation module and the execution parameter generation module, to form a complete anti-oscillation closed loop.

[0141] For example, the execution logic of the oscillation suppression unit begins with real-time monitoring of the mud pump speed control command issued by the adaptive control module. The mud pump speed control command is denoted as... This instruction is generated by the adaptive control module based on the risk level and external environment type, and transmitted to the actuator module via the communication interface. The oscillation suppression unit acquires the information through high-frequency sampling. The timing data, for example, is sampled ten times per second, and the rate of change of the speed command is calculated.

[0142] For example, the rate of change Its calculation uses the discrete difference method, and the formula is: ;in, Indicates the current moment. Indicates the previous moment, The sampling interval is defined as the rate of change. The rate of change quantifies the speed at which the mud pump speed command changes; a positive value indicates an acceleration command, and a negative value indicates a deceleration command.

[0143] Optionally, the oscillation suppression unit compares the calculated rate of change of the speed command with a preset acceleration judgment value to determine whether to trigger the transient tolerance mode. The acceleration judgment value is denoted as... This value is based on the characteristics of the mud pump and the operation history settings. For example, a change in rotational speed of 100 revolutions per second per minute indicates the system's sensitivity to the acceleration process.

[0144] When the rate of change of speed command Positive value and exceeding At this time, the system triggers transient tolerance mode. The triggering condition can be expressed as follows: The transient tolerance mode is a temporary state in which the system suspends some strict threshold checks and initiates a compensation mechanism to cope with the expected increase in physical parameters. After the mode is triggered, the oscillation suppression unit records the trigger time. And initialize the relevant parameters to prepare for subsequent calculations.

[0145] For example, in transient tolerance mode, the oscillation suppression unit calculates the theoretical increment of physical parameters expected due to the increase in rotational speed, based on the performance curve of the mud pump and the current mud density. The theoretical increment of physical parameters is denoted as... Its calculations are based on a dynamic model of the mud pump. The performance curve of the mud pump describes the relationship between rotational speed and output parameters such as pressure and flow rate, and is usually obtained by fitting experimental data, such as a polynomial function. ,in, This refers to physical parameters such as pipeline pressure; This indicates the speed of the mud pump.

[0146] The current mud density is denoted as The concentration is obtained through the ultrasonic concentration sensor in the data acquisition module. Theoretical increment. Through the formula: Calculate, where, To achieve the target rotational speed after acceleration, This is the base speed before acceleration. For multi-parameter systems, It can be represented as a vector, encompassing key physical quantities such as pressure gradient and flow rate. The calculation process also considers the influence of mud density; for example, high-density mud may lead to a larger pressure increment, therefore a density correction factor is introduced. The formula is expanded to: ,in, and Positive correlation.

[0147] Optionally, based on the theoretical increment of physical parameters, the oscillation suppression unit generates a transient compensation value that decays exponentially with time. This transient compensation value is denoted as... The calculation formula is as follows: ,in, The transient time decay constant is... For the current moment, This refers to the trigger time of the transient fault-tolerant mode. The transient time decay constant. The decay rate of the compensation value is controlled, and its value is set according to the system response characteristics. For example, 0.1 per second means that the compensation value decays to approximately 37% of its initial value within 10 seconds. Exponential decay ensures that the compensation value is higher in the initial acceleration phase to accommodate parameter increases, and then gradually decreases to avoid long-term interference risk identification. Transient compensation value. It is a dimensionless quantity or a quantity with physical units, depending on the type of parameter being compensated. For example, when applied to a pressure gradient threshold, its unit may be kilopascal per meter.

[0148] For example, the oscillation suppression unit superimposes the transient compensation value onto the dynamic threshold to form the final execution threshold used for real-time comparison. The dynamic threshold is denoted as... The threshold value is calculated by the adaptive control module based on external environmental correction factors, such as pressure gradient threshold or vibration amplitude threshold. The final execution threshold is denoted as... The calculation formula is as follows: The superposition operation continues during transient fault-tolerant mode to ensure that the final execution threshold is temporarily raised to match the natural increase in physical parameters. For example, during sludge pump acceleration, the pipeline pressure gradient may rise from 5000 Pa / m to 6000 Pa / m, while the dynamic threshold may be set to 5500 Pa / m; by superimposing compensation values, the final execution threshold may be temporarily adjusted to 6500 Pa / m, thus avoiding misinterpreting a normal rise as a risk of pipe blockage. The final execution threshold is used to compare internal operating data in real time, such as comparing the real-time pressure gradient with... Compare to determine the current risk level.

[0149] Optionally, the final execution threshold is temporarily raised during mud pump acceleration to accommodate the natural increase in physical parameters, and then falls back to the dynamic threshold after the speed stabilizes. Speed ​​stabilization is determined based on the rate of change of the speed command returning to 0 or below a certain small threshold, for example... .

[0150] When the system detects that the rotational speed is stable, the oscillation suppression unit gradually exits the transient fault-tolerant mode, and the transient compensation value... As the exponential decays, it naturally approaches 0, eventually reaching the execution threshold. Synchronous fallback to dynamic threshold The pullback process is smooth and continuous, avoiding control oscillations caused by sudden threshold changes. The entire mechanism ensures that the system can tolerate normal changes in physical parameters during capacity recovery without affecting the accuracy of long-term risk monitoring, thereby effectively preventing secondary false alarms.

[0151] Figure 7 shows a comparison of false alarm signals with and without the oscillation suppression mechanism enabled. The horizontal axis represents time in seconds, and the vertical axis represents the alarm signal strength, which has been normalized.

[0152] In Figure 7, the red dashed line represents the alarm curve without the oscillation suppression mechanism enabled, while the blue solid line represents the curve change after enabling the oscillation suppression mechanism. As can be seen from Figure 7, the red dashed line exhibits high-frequency fluctuations with significant peaks and large amplitudes, easily triggering false alarms. In contrast, the blue solid line is significantly more stable, with greatly reduced fluctuation amplitude, avoiding frequent triggering of invalid alarms. This comparison visually verifies the effectiveness of the transient compensation attenuation + slow-release response window mechanism in resisting disturbances, demonstrating that the system can maintain alarm accuracy and stability even in complex disturbance environments, thereby enhancing the practicality and robustness of the intelligent control system in dredging construction environments.

[0153] For example, the implementation of the oscillation suppression unit can be further extended to enhance adaptability and robustness. For instance, the calculation of the theoretical increments of physical parameters can be integrated with machine learning models to predict more accurate increment values ​​based on historical data. The decay pattern of the transient compensation value can be adjusted to a piecewise exponential form, decaying rapidly in the initial acceleration phase to quickly restore sensitivity, and decaying slowly in the later stages to maintain stability. The formation of the final execution threshold can introduce multi-parameter coupling, such as simultaneously compensating for pressure gradients and flow thresholds, to ensure comprehensive coverage. Furthermore, the unit can be equipped with adaptive learning capabilities to dynamically adjust the acceleration judgment value based on past false alarm records. or transient time decay constant To optimize performance.

[0154] Optionally, the oscillation suppression unit's workflow in system integration includes monitoring, triggering, calculation, compensation, and recovery phases. In the monitoring phase, the unit tracks the mud pump speed control command and its rate of change in real time. In the triggering phase, the unit evaluates the rate of change and decides whether to enter transient fault-tolerant mode. In the calculation phase, the unit calculates the theoretical increment and compensation value based on the mud pump model and mud density. In the compensation phase, the unit superimposes the compensation value to a dynamic threshold to form the final execution threshold. In the recovery phase, the unit gradually cancels the compensation after the speed stabilizes. The entire process operates in a cyclical manner, synchronized with the main cycle of the adaptive control module, for example, executing once every 0.1 seconds.

[0155] For example, the advantage of the oscillation suppression unit lies in its ability to significantly reduce the false alarm rate during the production recovery phase, thereby improving operational efficiency and equipment lifespan. In traditional systems, the conflict between low dynamic thresholds and mud pump acceleration often leads to frequent false alarms, causing unnecessary downtime or control adjustments, increasing operating costs and mechanical wear. This unit, however, intelligently distinguishes between normal parameter increases and genuine risks through a transient tolerance mechanism, reducing the number of interventions while maintaining the overall reliability of risk identification.

[0156] Optionally, the implementation details of the oscillation suppression unit include parameter initialization and calibration. Acceleration determination value. This needs to be set according to the mud pump model and operating scenario, for example, by analyzing common acceleration ranges through historical data. Transient time decay constant. Optimization can be achieved through simulation testing to ensure that the compensation duration matches the dynamic response of the mud pump. Mud pump performance curves. It should be updated regularly to reflect equipment aging or changes in operating conditions. The unit is also equipped with a diagnostic interface that outputs real-time data such as compensation values ​​and thresholds for operators to monitor system status.

[0157] For example, the oscillation suppression unit can collaborate with other modules to extend its functionality. For instance, combined with a self-learning iterative module, the unit can record transient tolerance events to a case library for predictive optimization of similar future operating conditions. In conjunction with the motion decoupling and inertial filtering unit, it can adjust the compensation strategy during hull lateral movement to avoid the superposition of multiple dynamic disturbances. Furthermore, the unit can integrate safety redundancy, retaining basic risk checks during compensation to prevent real risks from being ignored.

[0158] Example 5:

[0159] Corresponding to the above system embodiments, the present invention also proposes a dredging vessel mud concentration control method based on an adaptive algorithm. The method first collects mud concentration and flow rate in real time by using an ultrasonic concentration sensor and an electromagnetic flow meter installed behind the suction port. At the same time, it collects the pipeline pressure gradient by using a pressure gradient sensor installed along the mud discharge pipe, and collects the water flow velocity and direction by using a Doppler current meter installed at the bow of the ship, and obtains the tidal phase and rate by connecting to the ship's GPS.

[0160] Next, the collected data is spatiotemporally aligned. The transmission time delay is calculated based on the slurry flow velocity in the pipeline and the sensor distance. The external environment data is then aligned with the delay, and the water flow direction is converted into the angle relative to the pipeline axis. Subsequently, the water flow correction factor is calculated through a three-dimensional nonlinear model, and the tidal correction factor is calculated through a multi-dimensional coupling model.

[0161] Then, the static risk threshold is dynamically corrected using a correction factor to obtain a new dynamic threshold. The risk level is determined by comparing the real-time operating data with the dynamic threshold.

[0162] Finally, based on the risk level and the type of external environment, control commands for the mud pump speed, suction port opening and cutter speed are generated through linkage calculation formula, and the actuator is driven to complete the concentration adjustment.

[0163] The advantages of this method are that it achieves accurate identification and early warning of pipe blockage risk through multi-source data fusion and dynamic environmental correction; it significantly improves operation efficiency and system stability through adaptive linkage control strategy; and it effectively suppresses control oscillations caused by sudden environmental changes and equipment start-up and shutdown, reducing false alarm rate and the need for manual intervention.

[0164] Example 6:

[0165] Corresponding to the above embodiments, the present invention also proposes an electronic device.

[0166] Figure 8 shows a schematic diagram of the structure of an electronic device according to the present invention. The electronic device 100 includes a processor 101 and a memory 103. The processor 101 and the memory 103 are connected, for example, via a bus 102. Optionally, the electronic device 100 may further include a transceiver 104. It should be noted that in practical applications, the transceiver 104 is not limited to one unit, and the structure of the electronic device 100 does not constitute a limitation on the embodiments of the present invention.

[0167] Processor 101 may be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in connection with this disclosure. Processor 101 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0168] Bus 102 may include a pathway for transmitting information between the aforementioned components. Bus 102 may be a PCI bus or an EISA bus, etc. Bus 102 may be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in Figure 8, but this does not mean that there is only one bus or one type of bus.

[0169] The memory 103 stores a computer program corresponding to the adaptive algorithm-based dredging vessel mud concentration control method of the above embodiments of the present invention. This computer program is executed by the processor 101. The processor 101 executes the computer program stored in the memory 103 to implement the content shown in the aforementioned method embodiments.

[0170] The electronic device 100 includes, but is not limited to, mobile terminals such as laptops and tablets, as well as fixed terminals such as desktop computers. The electronic device 100 shown in Figure 8 is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0171] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A dredging vessel mud concentration control system based on an adaptive algorithm, characterized in that, include: The data acquisition module is used to collect internal working condition data and external dynamic environmental data of dredging operations in real time. The data processing module is communicatively connected to the data acquisition module and is used to perform spatiotemporal alignment processing on the acquired data and calculate the external environment correction factor based on a nonlinear model. An adaptive control module, communicatively connected to the data processing module, is used to dynamically adjust the risk level of mud concentration control according to the external environment correction factor and generate collaborative control commands. An actuator module, communicatively connected to the adaptive control module, is used to execute mud concentration adjustment actions in response to the collaborative control commands. The process of the adaptive control module dynamically adjusting the risk level of mud concentration control includes: obtaining a preset risk identification threshold under static conditions, the risk identification threshold including a pressure gradient threshold and a vibration amplitude threshold; using the external environment correction factor to perform a division correction on the risk identification threshold to obtain a dynamic threshold; comparing real-time collected internal operating condition data with the dynamic threshold to determine the current risk level, the risk level corresponding to different pipe blockage probability intervals; and the adaptive control module generating execution parameters including mud pump speed, suction port opening, and cutter head speed based on the risk level and external environment type.

2. The system according to claim 1, characterized in that, The data acquisition module includes an internal operating condition acquisition unit, comprising an ultrasonic concentration sensor and an electromagnetic flow meter installed behind the suction port, for collecting real-time mud concentration data. Real-time flow rate of mud ; and pressure gradient sensors installed along the sludge discharge pipe to collect pipeline pressure gradients. The external environment acquisition unit includes a Doppler current meter installed on the bow of the dredging vessel for collecting water flow velocity data. and water flow direction ; and a tidal data parsing unit connected to the ship's GPS, used to obtain tidal phases. and tidal rate 。 3. The system according to claim 2, characterized in that, The spatiotemporal alignment processing performed by the data processing module includes: based on the mud flow velocity inside the pipeline. Physical distance from the sensor Calculate transmission time delay The formula is: ;in: ； The diameter of the pipe; the data collected by the external environment acquisition unit is processed according to the... Delaying the flow direction to align it with the data from the internal operating condition acquisition unit on the time axis; Converted to the relative angle with respect to the pipe axis The formula is: ;in: For the ship's heading; This refers to the direction of the pipeline axis.

4. The system according to claim 3, characterized in that, The external environment correction factors include water flow correction factors. The data processing module calculates using the following three-dimensional nonlinear model. ： ;in: Preset weighting coefficients; Let be the flow velocity influence function, when hour, Follow Linear increase; Let be the directional influence function, defined as ； The pulsation effect function is related to the water flow pulsation coefficient. Positive correlation; It is defined as the ratio of the difference between the maximum and minimum water flow velocities per unit time to the average water flow velocity.

5. The system according to claim 4, characterized in that, The external environment correction factor also includes a tidal correction factor. The data processing module calculates using the following multi-dimensional coupling model. ： ;in: Preset weighting coefficients; The phase influence function is given when the tidal phase... The value is greater than 1.0 during high tide and less than 1.0 during low tide; Let be the rate influence function, which varies with tidal rate. Increase and increase; The deviation between the real-time water depth and the reference water depth during slack tide. The water depth deviation influence function is calculated based on the ratio of the deviation between the real-time water depth and the reference water depth during slack tide.

6. The system according to claim 5, characterized in that, The dynamic threshold correction performed by the adaptive control module follows the following formula: ;in: For micro-feature thresholds in a static environment; The corrected dynamic threshold; when When the dynamic threshold decreases, the system's sensitivity to risk increases; when At that time, the dynamic threshold increases, and the system's tolerance for risk increases.

7. The system according to claim 6, characterized in that, The mud pump speed is among the execution parameters generated by the adaptive control module. Water intake opening and auger speed Calculated using the following linkage formula: ； ； ;in: These are the foundation mud pump speed, foundation suction port opening, and foundation cutter speed, respectively. The current risk level is represented by an integer value from 1 to 5, corresponding to safe, low risk, medium risk, high risk, and critical risk, respectively.

8. The system according to claim 1, characterized in that, It also includes a self-learning iteration module; the self-learning iteration module is used to build a case library, which stores historical cases including external environment correction factors, internal working condition data and corresponding control action parameters; the self-learning iteration module calculates the external environment similarity and internal working condition similarity between the current working condition and the historical cases in the case library, and when the weighted matching similarity reaches a preset threshold, it directly calls the control action parameters of the corresponding historical case, and periodically updates the calculation model parameters of the external environment correction factor based on the execution effect feedback.

9. The system according to claim 4, characterized in that, The data processing module also includes a motion decoupling and inertial filtering unit for hull lateral movement operations, used to correct the water flow correction factor. The calculation process specifically includes: obtaining the real-time transverse winch rotation speed fed back by the actuator, and calculating the transverse linear velocity vector at the bow in combination with the ship's geometric parameters; calculating the projection component of the transverse linear velocity vector in the direction detected by the Doppler current meter, and calculating the projection component in the direction of the water flow velocity. The proportion in is defined as the motion disturbance degree. When the motion disturbance degree exceeds the preset disturbance threshold, the system enters the inertial filtering mode. At this time, the final water flow correction factor is calculated by weighting the instantaneous calculation value at the current moment with the final output value at the previous moment. In the weighted calculation, the weight coefficient of the instantaneous calculation value at the current moment is negatively correlated with the motion disturbance degree to suppress the sudden change of the correction factor caused by the rapid lateral movement of the hull.

10. The system according to claim 6, characterized in that, The adaptive control module is also equipped with an oscillation suppression unit for the production capacity recovery phase. Its specific execution logic includes: real-time monitoring of the mud pump speed control command issued by the adaptive control module and calculating the rate of change of the speed command; when the rate of change of the speed command is positive and exceeds the preset acceleration judgment value, the system triggers a transient tolerance mode; in the transient tolerance mode, the system calculates the expected theoretical increment of physical parameters due to the increase in speed based on the performance curve of the mud pump and the current mud density; based on the theoretical increment, a transient compensation value that decays exponentially over time is generated, and the transient compensation value is superimposed on the dynamic threshold to form a final execution threshold for real-time comparison; the final execution threshold is temporarily raised during the mud pump acceleration to accommodate the natural increase of physical parameters, and falls back to the dynamic threshold after the speed stabilizes, thereby avoiding secondary false alarms caused by excessively low thresholds during the production capacity recovery process.

Images