Diving method and device for launching and controlling, storage medium and electronic equipment

Abstract

The application provides a method and device for controlling a submersible hoisting system, a storage medium and an electronic device. The method comprises: obtaining load interference and disturbance interference of the hoisting system within a first time period; determining a target working mode of the hoisting system based on the load interference and the disturbance interference; the hoisting system supports multiple different working modes, and the hoisting system has different working performances in different working modes; determining a first control parameter corresponding to the target working mode to control an actuator of the hoisting system based on the first control parameter; obtaining feedback parameters of the hoisting system within a second time period; the second time period is less than the first time period; and optimizing the first control parameter based on the feedback parameters to obtain a second control parameter to control the actuator of the hoisting system based on the second control parameter.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic control, and in particular to a method and device for hoisting and releasing control for diving, a storage medium and an electronic device. BACKGROUND

[0002] In diving operations, when the hoisting and releasing device performs the hoisting and releasing task of key devices such as diving bells, guide cable winches and umbilical winches in complex marine environments, the existing hydraulic hoisting and releasing control system adopts a fixed control strategy, which cannot be effectively adjusted according to actual working conditions, resulting in obvious oscillation, positioning drift and control lag of the hoisting and releasing device during operation, which seriously affects the accuracy of hoisting and releasing operation. SUMMARY

[0003] Therefore, the present application provides a method and device for hoisting and releasing control for diving, a storage medium and an electronic device to overcome the deficiencies in the prior art.

[0004] According to a first aspect of the present application, a method for hoisting and releasing control for diving is provided, which is applied to a first processor of a hoisting and releasing system in a ship, and the method comprises the following steps:

[0005] Obtaining load disturbance and disturbance disturbance of the hoisting and releasing system within a first time length;

[0006] Determining the working mode of the hoisting and releasing system as a target working mode based on the load disturbance and the disturbance disturbance; the hoisting and releasing system supports multiple different working modes, and has different working performances in different working modes;

[0007] Determining a first control parameter corresponding to the target working mode, so as to control the actuator of the hoisting and releasing system based on the first control parameter;

[0008] Obtaining feedback parameters of the hoisting and releasing system within a second time length; the second time length is less than the first time length;

[0009] Optimizing the first control parameter based on the feedback parameters to obtain a second control parameter, so as to control the actuator of the hoisting and releasing system based on the second control parameter.

[0010] Another aspect of the present application provides a device for hoisting and releasing control for diving, which comprises:

[0011] A first obtaining module is configured to obtain load disturbance and disturbance disturbance of the hoisting and releasing system within a first time length;

[0012] A first determining module is configured to determine the working mode of the hoisting and releasing system as a target working mode based on the load disturbance and the disturbance disturbance; the hoisting and releasing system supports multiple different working modes, and has different working performances in different working modes;

[0013] The second determining module is configured to determine a first control parameter corresponding to the target working mode, and control the actuator of the hoist and launch system based on the first control parameter.

[0014] The second obtaining module is configured to obtain a feedback parameter of the hoist and launch system within a second time length, and the second time length is less than the first time length.

[0015] The optimization module is configured to optimize the first control parameter based on the feedback parameter to obtain a second control parameter, and control the actuator of the hoist and launch system based on the second control parameter.

[0016] Another aspect of the present application provides an electronic device, comprising:

[0017] one or more processors;

[0018] a memory configured to store one or more programs,

[0019] When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described above.

[0020] Another aspect of the present application provides a computer-readable storage medium storing computer-executable instructions, which, when executed, implement the method as described above.

[0021] By adopting the technical solution of the present application, the hoist and launch system can accurately obtain the load disturbance and the disturbance disturbance within the first time length, realize real-time sensing of the complex changes of the marine environment and the dynamic characteristics of the equipment load, determine the target working mode based on the load disturbance and the disturbance disturbance, make the system intelligently select the most suitable control strategy according to the actual working condition, and avoid the problem of mismatch between the fixed control strategy and the actual working condition. The first control parameter corresponding to the target working mode is determined, and the actuator is controlled accordingly, thereby establishing a complete control link from working condition identification to parameter optimization and then to execution control, and ensuring high matching between the control strategy and the actual working condition.

[0022] The double-time-length control architecture of obtaining the feedback parameter within the second time length and optimizing the first control parameter to obtain the second control parameter realizes the combination of long-time-length coarse adjustment and short-time-length fine adjustment. Since the second time length is less than the first time length, the system can capture the overall change trend of the working condition and quickly respond and correct the slight deviation in the control process, thereby improving the dynamic response capability and control precision of the control system. The actuator is controlled based on the second control parameter to form a closed-loop adaptive control, so that the entire hoist and launch operation process can continuously optimize the control effect, effectively solve the problems of oscillation, positioning drift and control lag of the hoist and launch device in the prior art, and improve the precision of the hoist and launch operation.

[0023] It is to be understood that the details set forth herein do not limit the scope of the application. One of ordinary skill in the art would realize that other alternatives of the application also fall within the scope of the application. It is also to be understood that the following description is mainly for describing the principles of the present application and the application should not be limited thereto since the application is covered by the claims. BRIEF DESCRIPTION OF DRAWINGS

[0024] The above and other objects, features and advantages of the present application will become more apparent from the following description when taken in conjunction with the accompanying drawings, in which:

[0025] Figure 1 An architecture diagram of a launch and recovery system in a ship is illustratively shown;

[0026] Figure 2 A flow chart of a launch and recovery method for diving is illustratively shown;

[0027] Figure 3 A structure block diagram of a launch and recovery device for diving is illustratively shown;

[0028] Figure 4 A structure block diagram of an electronic device is illustratively shown. DETAILED DESCRIPTION

[0029] Hereinafter, embodiments of the present application will be described with reference to the accompanying drawings. It is to be understood, however, that the description is merely illustrative of the present application and is in no way intended to restrict the scope of the present application. Throughout the specification, numerous specific details are set forth in order to provide a thorough understanding of the present application. It will be apparent, however, to one skilled in the art that one or more embodiments of the present application can be practiced without such specific details. In other instances, well-known structures and functions are not described in detail in order to avoid obscuring the concept of the present application.

[0030] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. As used herein, the term "includes" and tautological expressions thereof, such as "including," "includes," "include," "contains," "containing," and so forth, shall be read expansively and without limitation. The terms "comprising," "comprise" and / or "comprises" and tautological expressions thereof (for example, "comprising a," "comprises an," etc.) shall be interpreted open as taking their broadest forms so as not to be limited to the elements thereof recited; that is, they will be interpreted to mean the elements recited therein, but also elements not specifically recited. Any process disclosed comprising steps can not be limited to those steps specifically discloses unless such limitation is explicitly recited. The terms "exemplary" and / or "example" are used herein to mean an instance of the general thing named, and are not used to imply a "preferred" or "superior" over other examples.

[0031] Embodiments of the present application provide a launch and recovery method and device for diving, a storage medium and an electronic device. The method can be applied to a first processor of a launch and recovery system in a ship. Hereinafter, the launch and recovery system in a ship will be introduced.

[0032] Figure 1 An architecture diagram of a launch and recovery system in a ship is illustratively shown.

[0033] As Figure 1As shown, the ship launching system 100 adopts a distributed control architecture, mainly including a first processor 110, a switch 120, a second processor 130 and a hydraulic system 140, and each component is interconnected and cooperatively controlled through an industrial Ethernet.

[0034] The first processor 110 serves as the global control center and decision core of the system, and undertakes advanced control functions such as working condition perception, prediction analysis, mode selection and parameter optimization.

[0035] The first processor 110 establishes a human-computer interaction interface with the upper computer 111 through the connection 110. The upper computer 111 usually adopts an industrial touch screen or an HMI device, and provides an intuitive operation interface for the operator to realize functions such as operation instruction input, system state monitoring and alarm information display. The operator can set the launching task parameters, select the operation mode, monitor the real-time running data through the upper computer 111, and perform manual intervention in abnormal conditions.

[0036] The first processor 110 is connected with the first panel 112 through the connection 112. The first panel 112 is configured with physical buttons, knobs and indicator lights, etc. operation elements, providing redundant operation channels and emergency control means. When the upper computer 111 fails or the network communication is interrupted, the operator can still realize basic start-stop control, speed adjustment and safety protection, etc. key operations through the first panel 112, to ensure the controllability and safety of the system in emergency situations.

[0037] The second processor 130 serves as the field execution control unit and hydraulic drive core, and is responsible for receiving the control instructions issued by the first processor 110, and converting them into specific hydraulic control signals to drive the execution mechanism to move. The second processor 130 is built-in with PID control algorithm and hydraulic proportional valve control logic, which can accurately adjust the pressure, flow and direction of the hydraulic system according to the control parameters.

[0038] The second processor 130 is connected with the second panel 131 through the connection 130. The second panel 131 is arranged near the hydraulic equipment field, and is equipped with field operation buttons, state indicator lights and emergency stop switches, etc. safety devices.

[0039] The second panel 131 provides an on-site emergency operation interface. When the upper control system fails or needs to be maintained on site, technicians can switch to the local operation mode through the second panel 131 to directly control the hydraulic system to run, ensuring the convenience of equipment maintenance and emergency disposal. The second processor 130 establishes a multi-way control connection with the hydraulic system 140 through the connection 140, to realize independent control and coordinated action of multiple hydraulic actuators.

[0040] Switch 120 serves as the communication hub and data exchange center of the system, bearing the core communication function of the distributed control network. Switch 120 establishes an uplink communication link with the first processor 110 through connection 110, responsible for receiving control instructions, work mode switching signals and parameter configuration data generated by the first processor 110. Switch 120 establishes a downlink communication link with the second processor 130 through connection 120, and forwards the control instructions to the second processor 130 in real time for execution, while receiving state feedback data, sensor collected information and fault alarm signals uploaded by the second processor 130, and forwarding these data to the first processor 110 for analysis and processing.

[0041] Switch 120 uses the industrial Ethernet Profinet protocol, with high bandwidth, low delay and strong real-time communication characteristics, ensuring synchronous data transmission and reliable communication between multiple processors in the distributed control system. Switch 120 also has network redundancy and fault self-healing functions, which can automatically switch to the standby path when a communication path fails, ensuring the communication stability of the control system.

[0042] Hydraulic system 140 serves as the power execution layer and mechanical operation terminal of the system, including bell winch 141, cable guide winch 142, umbilical winch 143, door-type davit 144 and compensation device 145, etc.

[0043] Bell winch 141, as the core equipment of subsea operation, is responsible for the accurate launching and safe recovery of the diving bell. Its hydraulic drive system has large torque output and precise position control capability, which can adapt to the operation requirements under different water depths and sea conditions.

[0044] Cable guide winch 142 is specially used for tension control and guidance management of cables and pipes. Through precise tension control, it prevents cable overstretching or relaxation, ensuring the safety and reliability of communication cables and power supply cables.

[0045] Umbilical winch 143 is responsible for managing the life support umbilical cable, including gas supply pipeline, communication cable and safety rope, etc. The control accuracy directly relates to the life safety of divers.

[0046] Door-type davit 144 provides stable mechanical support and precise guidance function. Through hydraulic drive, it realizes the lifting, rotation and positioning actions of the davit, providing reliable mechanical foundation for the whole launching system.

[0047] Compensation device 145 uses active wave compensation technology. By real-time sensing of ship motion and wave disturbance, it actively adjusts the hydraulic cylinder stroke to offset external interference, maintaining the stable position of the launching equipment relative to the seabed.

[0048] Each hydraulic actuator is equipped with a high-precision sensor system, including displacement sensors, pressure sensors, load sensors, inclination sensors, and temperature sensors, and other types of detection equipment. These sensors collect real-time operating state parameters of the actuator, including actual displacement, running speed, bearing load, hydraulic pressure, inclination angle, and oil temperature, and other key data, and transmit the data to the second processor 130 through the field bus.

[0049] After the second processor 130 pre-processes and filters the collected state data, it is uploaded to the first processor 110 for comprehensive analysis and decision calculation, forming a complete closed-loop control system for state perception, data transmission, analysis processing, and control execution, and realizing accurate control of the hoisting operation in complex marine environments.

[0050] On the basis of the above system architecture, the following describes the hoisting control method for underwater use provided by the embodiments of the present application.

[0051] Figure 2 A flowchart of a hoisting control method for underwater use provided by the embodiments of the present application is schematically shown.

[0052] As shown in Figure 1 The hoisting control method for underwater use includes steps S101-S103.

[0053] Step S201, acquiring load disturbance and disturbance disturbance of the hoisting system within a first time length;

[0054] Step S202, determining the working mode of the hoisting system as a target working mode based on the load disturbance and the disturbance disturbance; the hoisting system supports multiple different working modes, and has different working performances in different working modes;

[0055] Step S203, determining a first control parameter corresponding to the target working mode, to control the actuator of the hoisting system based on the first control parameter;

[0056] Step S204, acquiring feedback parameters of the hoisting system within a second time length; the second time length is less than the first time length;

[0057] Step S205, optimizing the first control parameter based on the feedback parameter to obtain a second control parameter, to control the actuator of the hoisting system based on the second control parameter.

[0058] In step S201, the load disturbance refers to the influence of the external load change on the system control performance during the operation of the hoisting system actuators, which can be understood in the embodiments of the present application as the system control deviation caused by the dynamic change of the load weight of the diving bell, the mooring winch, the umbilical winch and other equipment, and is used to represent the impact degree of the equipment on the control system stability when switching from the empty load to the full load state.

[0059] Exemplarily, the load disturbance includes but is not limited to the weight change in the diving bell hoisting process, the load increase and decrease when the winch winds the cable, the dynamic adjustment of the umbilical cable tension, and the load coupling effect when multiple devices operate simultaneously.

[0060] Similarly, the disturbance disturbance refers to the random influence of the external marine environment factors on the position stability and motion smoothness of the hoisting system, which can be understood in the embodiments of the present application as the degree of deviation of the hoisting equipment from the predetermined trajectory caused by the sea state changes such as wave fluctuation, current impact and ship swing, and is used to quantify the disturbance intensity of the complex marine environment on the system control accuracy.

[0061] Exemplarily, the disturbance disturbance includes but is not limited to the periodic swing caused by different levels of waves, the lateral deviation caused by the change of the current direction, the displacement disturbance transmitted by the six degrees of freedom motion of the ship, and the combined disturbance generated by the coupling action of wind and wave.

[0062] In the formula, the first duration refers to the time window length required for the system to perform working condition prediction and mode selection, which can be understood in the embodiments of the present application as the time span for the hoisting system to collect historical data and predict future load changes and disturbance trends, and is used to ensure that the system can capture the complete periodical characteristics of the working condition change and make accurate control strategy adjustment.

[0063] The setting of the first duration needs to consider the change period of the marine environment, the switching frequency of the equipment load and the response requirement of the control system, so as to ensure the accuracy of the prediction and meet the demand of real-time control.

[0064] In a feasible implementation, the first processor continuously collects state data from various sensors of the hydraulic system through industrial Ethernet, and establishes a data collection window covering the first duration. The first processor obtains real-time state information through multiple types of sensors installed on each actuator, wherein the load related data is collected through pressure sensors and load sensors, and the disturbance related data is collected through displacement sensors, inclination sensors and acceleration sensors. The first processor pre-processes and filters the collected data, eliminates abnormal values and noise disturbances, and forms a reliable historical data sequence.

[0065] Subsequently, the first processor predicts the load change trend and the disturbance change mode by using a time series analysis method based on historical data within the first time length. For the load disturbance, the first processor analyzes the change amplitude, change frequency and duration of the load data, identifies the regularity and suddenness of the load switching, and predicts the impact strength of the load change on the control system in the future period. For the disturbance disturbance, the first processor analyzes the statistical characteristics and frequency domain characteristics of the displacement and inclination data, identifies the periodicity and randomness of the sea state change, and predicts the influence degree of the environmental disturbance on the system stability in the future period.

[0066] In step S202, the working mode refers to a control strategy combination pre-configured by the launch and recovery system for different working conditions, which can be understood in the embodiments of the application as a set of differentiated control parameters set according to different combinations of load levels and disturbance levels, for enabling the launch and recovery system to adopt the most suitable control strategy to achieve the best control performance when facing a specific working condition.

[0067] Optionally, the launch and recovery system supports multiple different working modes, each corresponding to a specific load range and disturbance intensity, and has different response speed, control accuracy, stability performance and anti-interference ability in different working modes, so as to meet the needs of complex and changeable marine operating environment.

[0068] Exemplarily, the working modes include but are not limited to light load still water mode, light load moderate disturbance mode, light load severe disturbance mode, medium load still water mode, medium load moderate disturbance mode, medium load severe disturbance mode, heavy load still water mode, heavy load moderate disturbance mode, heavy load severe disturbance mode, etc.

[0069] In a feasible implementation, the first processor determines the target working mode by working condition analysis and mode matching based on the load disturbance and the disturbance disturbance. The first processor classifies the load disturbance into three levels of light load, medium load and heavy load according to the preset load threshold, wherein the light load level corresponds to the empty load or light load running state of the device, the medium load level corresponds to the normal load running state of the device, and the heavy load level corresponds to the full load or overload running state of the device.

[0070] Correspondingly, the first processor classifies the disturbance disturbance into three levels of still water, moderate disturbance and severe disturbance according to the preset disturbance threshold, wherein the still water level corresponds to the ideal sea state of windless and waveless, the moderate disturbance level corresponds to the wave influence under general sea state, and the severe disturbance level corresponds to the strong wave impact under severe sea state.

[0071] The first processor can determine a candidate working mode that best matches the current working condition from a plurality of preset working modes according to the combination of the load level and the disturbance level. The first processor calculates the matching degree between the current load level and the disturbance level and each candidate working mode, and determines the applicability weight of each candidate working mode through the matching degree evaluation. For the working mode with a higher matching degree, the first processor gives a larger weight coefficient; for the working mode with a lower matching degree, the first processor gives a smaller weight coefficient. Based on the weight coefficients of the candidate working modes, the control parameters of the corresponding working modes can be weighted and fused to generate a target working mode that adapts to the current working condition.

[0072] In step S203, the first control parameter refers to a combination of basic control variables used to drive the hydraulic actuator to move in the target working mode, which can be understood in the embodiments of the application as a set of proportional parameters, integral parameters and derivative parameters pre-adjusted and stored according to a specific working condition, used to ensure that the hoisting system can achieve the expected response speed, control accuracy and stability performance in the corresponding working mode.

[0073] Exemplarily, the first control parameter can be a high response speed parameter combination in the light load static water mode, a balance performance parameter combination in the medium load medium disturbance mode, a high stability parameter combination in the heavy load severe disturbance mode, and the like.

[0074] In a feasible implementation, the first processor retrieves the corresponding first control parameter from a pre-established control parameter database based on the target working mode. The control parameter database is a parameter knowledge base established through a large amount of sea trial data and simulation optimization, in which the optimal control parameter combination verified in different working modes is stored. The first processor locates the corresponding parameter record in the control parameter database through mode indexing according to the target working mode, and obtains the proportional parameter, integral parameter and derivative parameter corresponding to the working mode.

[0075] The first processor transmits the first control parameter to the second processor through the industrial Ethernet, and the second processor updates the internal control algorithm configuration immediately after receiving the first control parameter. The second processor recalculates the opening control signal of the hydraulic proportional valve and the pressure regulation instruction of the hydraulic pump station based on the first control parameter, to realize the control of the hoist winch, the mooring winch and the umbilical winch and the like.

[0076] During the actuator control process, the second processor can also continuously monitor the actual operating parameters of the hydraulic system, including the displacement of the hydraulic cylinder, the system pressure, the flow size and the load change and the like. These real-time monitoring data are compared and analyzed with the expected control effect of the first control parameter, the applicability of the control parameter is evaluated, and the feedback is fed back to the first processor.

[0077] In step S204, the feedback parameter refers to a state response signal and a performance evaluation index generated by the hoisting system actuator in actual operation, which can be understood in the embodiment of the application as real-time measurement data reflecting the current operation effect and deviation degree of the control system, used for evaluating the actual control effect of the first control parameter and providing a data basis for subsequent parameter optimization.

[0078] Exemplarily, the feedback parameter can be a position error, a speed error, a swing angle, a system response time, an overshoot, a steady-state error, an oscillation frequency, and the like comprehensive index reflecting control performance.

[0079] Similarly, the second duration refers to the time window length of the system for feedback data collection and parameter optimization adjustment, which can be understood in the embodiment of the application as a high-frequency control period shorter than the first duration, used for realizing rapid capture and timely correction of instantaneous deviation and dynamic response in the control process.

[0080] The second duration is shorter than the first duration, which can enable the system to perform working condition prediction and mode selection on a long time scale while performing fine parameter tuning on a short time scale.

[0081] In a feasible implementation, the first processor obtains the feedback parameter in the second duration from the second processor through the switch. The second processor continuously collects the running state data of the hydraulic actuator in the time window of the second duration, obtains real-time feedback information through the high-precision sensor array installed on each actuator, including the actual position signal measured by the displacement sensor, the movement speed data obtained by the speed sensor, the swing angle information detected by the inclination sensor, and the working pressure of the hydraulic system monitored by the pressure sensor, and the like multi-dimensional state parameters.

[0082] In step S205, the second control parameter refers to an optimized control variable combination obtained by adaptively adjusting the first control parameter based on real-time feedback information, which can be understood in the embodiment of the application as a fine adjustment set of the proportional parameter, the integral parameter and the differential parameter optimized through closed-loop feedback.

[0083] In a feasible implementation, the first processor constructs a comprehensive performance evaluation function based on the obtained feedback parameter, combines the feedback parameters such as position error, speed error and swing angle according to a preset weight, and forms an evaluation index that can comprehensively reflect the performance of the control system. The first processor gradually adjusts the first control parameter by using an iterative optimization algorithm, calculates the sensitivity of the performance evaluation function to each parameter component, and determines the optimization direction and adjustment amplitude of the proportional parameter, the integral parameter and the differential parameter.

[0084] Optionally, the first processor can also adjust the step size according to a preset learning rate control parameter, subtract the corresponding update amount from each parameter component to obtain an optimized parameter value. The first processor recombines the optimized parameter components to form a second control parameter, and performs boundary constraint checking to ensure parameter safety. The first processor sends the second control parameter to the second processor through the industrial Ethernet, and the second processor updates the control algorithm configuration based on the second control parameter, recalculates the hydraulic control instruction based on the second control parameter, and realizes the optimized control of the actuator.

[0085] By adopting the technical scheme of the present application, the load disturbance and the disturbance disturbance of the hoisting system can be accurately obtained within the first time length, and real-time sensing of the complex changes of the marine environment and the dynamic characteristics of the equipment load is realized. Based on the load disturbance and the disturbance disturbance, the target working mode is determined, so that the system can intelligently select the most suitable control strategy according to the actual working condition, avoiding the problem of mismatch between the fixed control strategy and the actual working condition. The first control parameter corresponding to the target working mode is determined, and the actuator is controlled accordingly, establishing a complete control link from working condition identification to parameter optimization to execution control, and ensuring high matching between the control strategy and the actual working condition.

[0086] The double-time-length control architecture of obtaining the feedback parameter within the second time length and optimizing the first control parameter to obtain the second control parameter realizes the combination of long-time-length coarse adjustment and short-time-length fine adjustment. Since the second time length is less than the first time length, the system can capture the overall trend of the working condition and quickly respond and correct the slight deviation in the control process, thereby improving the dynamic response capability and control precision of the control system. Based on the second control parameter, the actuator is controlled to form a closed-loop adaptive control, so that the entire hoisting operation process can continuously optimize the control effect, effectively solving the problems of oscillation, positioning drift and control lag of the hoisting device in the prior art, and improving the accuracy of the hoisting operation.

[0087] On the basis of the above-mentioned embodiments, as an optional embodiment, the process of obtaining the load disturbance and the disturbance disturbance of the hoisting system within the first time length can be realized by combining state signal acquisition and time series prediction. Specifically, step S201 can further include the following steps:

[0088] Step S301, obtaining the current state signal of the hoisting system; the state signal includes a load signal, a displacement signal and an inclination signal;

[0089] Step S302, taking the load signal as a load sequence, and taking the inclination signal and the displacement signal as a disturbance sequence;

[0090] Step S303, predicting the load disturbance within the first time length based on the load sequence;

[0091] Step S304, based on the disturbance sequence, predicting the disturbance interference in the first time length.

[0092] In step S301, the first processor obtains the current state signals of the launch system from the sensor network of the hydraulic system through industrial Ethernet. The state signals include load signals, displacement signals, and inclination signals.

[0093] The load signals are obtained by pressure sensors and load sensors installed on each actuator, which can reflect the current load bearing of devices such as winch, cable winch, and umbilical winch in real time.

[0094] The displacement signals are obtained by high-precision displacement sensors, which are used to monitor the actual deviation of the actuator from the set position.

[0095] The inclination signals are obtained by inclination sensors, which are used to detect the inclination of the actuator relative to the vertical direction.

[0096] Optionally, the first processor can perform digital filtering and denoising on the collected raw state signals to eliminate random noise and abnormal mutation values generated during the sensor collection process.

[0097] In step S302, the first processor directly processes the load signals as load sequences, and by establishing the load time sequence, it can effectively capture the load change pattern of the device from empty load to full load.

[0098] In contrast, the first processor constructs the inclination signals and displacement signals as disturbance sequences, because inclination changes and displacement deviations are usually caused by external oceanic disturbances, and the two have a strong coupling relationship in physical mechanism, and processing them can more comprehensively reflect the influence of external disturbances on system stability.

[0099] Optionally, the first processor adopts a weighted average method in the fusion process, sets corresponding weight coefficients according to the different influence degrees of inclination changes and displacement deviations on system control performance, and ensures that the disturbance sequence can accurately reflect the actual strength of external interference.

[0100] For example, the load sequence can be represented as:

[0101]

[0102] In the formula, represents the load signal at time t, represents the load signal at time t-k.

[0103] The disturbance sequence can be represented as:

[0104]

[0105] wherein, denotes the displacement signal at time t, denotes the tilt angle signal at time t, , denote weighting coefficients for adjusting the weight of displacement and tilt angle in the disturbance.

[0106] The set of disturbance sequences can be denoted as:

[0107]

[0108] wherein, denotes the disturbance sequence at time t, denotes the disturbance sequence at time t-k.

[0109] In step S303, since the load change often has certain periodicity and predictability, especially in regular operation process, the switching time and change range of the equipment load usually follow the established operation plan and operation process.

[0110] Optionally, the first processor can identify the time pattern, amplitude feature and duration cycle of the load change by analyzing the historical change data of the load sequence, and establish a mathematical prediction model of the load change. The prediction model can infer the change situation of the load that may occur in the future first time length according to the current load state and historical change trend, including the time point of load increase, the amplitude range of load change and the duration time of load stability. Through load disturbance prediction, the impact degree of the upcoming load change on the control system can be known in advance.

[0111] Exemplarily, the load disturbance prediction model can be denoted as:

[0112]

[0113] Exemplarily, the disturbance disturbance prediction model can be denoted as:

[0114]

[0115] wherein, denotes the disturbance disturbance prediction value at time t, denotes the load disturbance prediction value at time t, , denote the mean terms of load and disturbance respectively, , denote the autoregressive coefficients respectively, , denote the moving average coefficients respectively, , denote the moving average coefficients respectively, , respectively represent white noise terms.

[0116] Specifically, the first processor adopts an autoregressive integrated moving average model as the mathematical basis for load disturbance prediction, which establishes a complete mathematical framework for describing the dynamic law of load changes by comprehensively considering the autoregressive characteristics of historical load data, the difference stationarization processing and the moving average characteristics of random errors.

[0117] The load disturbance prediction model is expressed as a linear combination of the predicted load value equal to the weighted sum of the load mean term and the historical load value, the weighted sum of the historical error term and the current random error term. The autoregressive part captures the inertia characteristics of load changes by analyzing the time correlation of historical load data, and the moving average part modifies the accuracy of the prediction result by considering the influence of historical prediction errors.

[0118] When establishing the load disturbance prediction model, the first processor first performs stationarity test and difference processing on the collected load sequence to eliminate the trend and seasonal effects that may exist in the load data, and to ensure that the time series meets the stationarity requirement. Subsequently, the order parameters of the model are determined through autocorrelation function analysis and partial autocorrelation function analysis, wherein the autoregressive order reflects the dependence depth between the current load value and the historical load value, and the moving average order reflects the memory length of the prediction model for historical error information. The first processor optimizes and solves the model parameters using the maximum likelihood estimation method, and obtains the optimal parameter combination that minimizes the prediction error through iterative calculation, including autoregressive coefficients, moving average coefficients and error variance parameters.

[0119] In step S304, the same autoregressive integrated moving average modeling method can be used to predict the disturbance disturbance within the first time length based on the disturbance sequence. The disturbance disturbance prediction model has a similar mathematical structure to the load disturbance prediction model, i.e., the predicted disturbance value is a linear combination of the weighted sum of the disturbance mean term and the historical disturbance value, the weighted sum of the historical error term and the current random error term. However, since the disturbance sequence integrates the composite information of displacement signals and inclination signals, its change characteristics are more complex than single load signals, and the first processor needs to specially consider the mutual influence and coupling relationship between different types of disturbance signals when establishing the disturbance prediction model.

[0120] Optionally, during the disturbance disturbance prediction process, the main periodic components of the marine environment disturbance can be identified by analyzing the frequency domain characteristics of the disturbance sequence, and the disturbance components of different frequency bands such as low-frequency current changes, medium-frequency wave motions and high-frequency ship swings are separated. For periodic disturbance components with strong periodicity, the first processor increases the order of the autoregressive term to better capture its periodic characteristics; for random disturbance components with strong randomness, the first processor increases the order of the moving average term to improve the prediction robustness of random disturbances.

[0121] By adopting the technical scheme of the present application, the first processor can change passive state monitoring into active trend prediction, improving the foresight and adaptability of the control system to working condition changes. Compared with the traditional method of controlling based on the current state, the technical scheme of the present application can identify upcoming load changes and environmental disturbances in advance. Through differentiated sequence construction, the different influence characteristics of internal load changes and external environmental disturbances can be more accurately distinguished, avoiding the problem of reduced prediction accuracy caused by mixing the two types of disturbances for processing.

[0122] On the basis of the above-mentioned embodiments, as an optional embodiment, the process of determining the working mode of the hoisting system as the target working mode based on the load disturbance and the disturbance disturbance can be realized by combining the level division and weight fusion.

[0123] Specifically, step S202 can further include the following steps:

[0124] Step S401, respectively determining a load level of the load disturbance and a disturbance level of the disturbance disturbance;

[0125] Step S402, among a plurality of working modes supported by the hoisting system, determining at least two first working modes corresponding to the load level and the disturbance level;

[0126] Step S403, calculating the matching degree of the load level and the disturbance level with each working mode;

[0127] Step S404, determining the weight coefficient of each first working mode based on the matching degree;

[0128] Step S405, based on the weight coefficient of each first working mode, performing weighted summation on the control parameters corresponding to each first working mode to obtain the target working mode.

[0129] In step S401, since the load changes and environmental disturbances faced by the hoisting system in the actual operation process often have the characteristics of continuity and gradualness, and the traditional hard switching control is easy to produce frequent mode jumping and control oscillation problems at the critical state of the working condition, it is necessary to convert the continuous load disturbance and disturbance disturbance into discrete level division for subsequent mode matching and parameter selection.

[0130] Optionally, the predicted load disturbance can be divided into three load levels of light load, medium load and heavy load through a preset load threshold, wherein the light load level corresponds to the load disturbance range when the hoisting system is in an empty load or low load running state, the medium load level corresponds to the load disturbance range when the hoisting system is in a normal working load, and the heavy load level corresponds to the load disturbance range when the hoisting system is in a full load or overload running state.

[0131] Correspondingly, the first processor divides the predicted disturbance interference into three disturbance levels, i.e., static water, moderate disturbance and severe disturbance, by a preset disturbance threshold, wherein the static water level corresponds to weak environmental disturbance in good sea conditions, the moderate disturbance level corresponds to moderate wave impact in general sea conditions, and the severe disturbance level corresponds to strong wave impact and compound environmental interference in severe sea conditions.

[0132] In order to avoid the grade jump problem caused by the hard threshold judgment, the fuzzy membership function method can be used in the grade division process, so that the load interference and the disturbance interference can belong to multiple levels at the same time and have corresponding membership degrees. Through the calculation of the fuzzy membership function, the first processor can obtain the membership value of the current load interference to each load level and the membership value of the current disturbance interference to each disturbance level.

[0133] Exemplarily, the predicted load interference is , and the disturbance interference is .

[0134] The load level membership function can be set as:

[0135]

[0136] In the formula, , the load level membership function is represented, and the degree to which the load belongs to the kth level is represented.

[0137] The disturbance level membership function is:

[0138]

[0139] In the formula, , the disturbance level membership function is represented, and the degree to which the disturbance belongs to the mth level is represented, wherein k and m respectively represent different levels (for example, 1 corresponds to a low level, 2 corresponds to a medium level, and 3 corresponds to a high level).

[0140] Specifically, the predicted load interference can be set as the load input variable to be divided, the disturbance interference can be set as the disturbance input variable to be divided, and a corresponding fuzzy membership function system can be established. The load level membership function describes the membership degree of the predicted load interference to the light load, medium load and heavy load levels, wherein the function output value represents the membership degree of the load to the kth level, and k takes a value range corresponding to the light load, medium load and heavy load levels. The disturbance level membership function uses the same mathematical framework to describe the membership degree of the predicted disturbance interference to the static water, moderate disturbance and severe disturbance levels, wherein the function output value represents the membership degree of the disturbance to the mth level, and m takes a value range of one to three corresponding to the static water, moderate disturbance and severe disturbance levels.

[0141] In the execution of the membership function calculation, the center value, bandwidth and overlapping area of each level are determined by the preset membership function parameters, so that the load interference and the disturbance interference can obtain smooth membership change near the level boundary instead of abrupt hard switching results. For the load level membership function, when the predicted load interference is located in the center area of a certain level, the membership of this level is close to one and the memberships of other levels are close to zero; when the predicted load interference is located in the boundary area of two levels, the adjacent two levels both have non-zero membership values, and the size thereof reflects the relative distance relationship between the load interference and the center of each level. The disturbance level membership function follows the same calculation logic, and the belonging relationship between the disturbance interference and each disturbance level is described by continuous membership values, thereby avoiding the level jump and control discontinuity problems that may be caused by the traditional hard threshold judgment.

[0142] In step S402, according to the currently determined load level and disturbance level, the candidate working modes related to the current working condition are selected from the multiple working modes supported by the system as the first working modes.

[0143] Due to the fuzzy membership characteristics of the load level and the disturbance level, the current working condition usually corresponds to at least two first working modes. For example, when the load interference is in the transition state between light load and medium load and the disturbance interference is in the transition state between still water and medium disturbance, the system will simultaneously select the light load still water mode, the light load medium disturbance mode, the medium load still water mode and the medium load medium disturbance mode as the candidate first working modes.

[0144] In step S403, in order to quantify the adaptation degree between the current working condition and each first working mode, the first processor needs to calculate the matching degree of the load level and the disturbance level with each working mode. The calculation of the matching degree can be based on the fuzzy membership values of the load level and the disturbance level, and the matching degree value of the current working condition and the working mode can be obtained by multiplying the membership of the load level corresponding to the load component of the working mode and the membership of the disturbance level corresponding to the disturbance component of the working mode. The higher the matching degree value is, the better the adaptation degree of the working mode to the current working condition is, and the better the control effect of the corresponding control parameter on the current working condition is expected.

[0145] Exemplarily, assuming that the working mode set is Each mode corresponds to a different combination of load level and disturbance level.

[0146] The matching degree of mode is:

[0147]

[0148] In the formula, represents the matching degree of mode .

[0149] Specifically, the first processor pre-establishes a working mode set, which contains all working modes supported by the sling system, each of which corresponds to a specific load level and disturbance level combination. The working mode set can be constructed based on the Cartesian product combination of load levels and disturbance levels, for example, the combination of light load level and still water level forms a light load still water working mode, the combination of medium load level and medium disturbance level forms a medium load medium disturbance working mode, the combination of heavy load level and severe disturbance level forms a heavy load severe disturbance working mode, and so on to form a complete working mode set covering various working condition combinations. Each working mode not only defines the corresponding load level and disturbance level characteristics, but also pre-configures the optimized control parameters suitable for the working condition combination, including the targeted tuning of the proportional, integral and derivative parameter combination.

[0150] In calculating the matching degree of a specific working mode, first identify the load level identifier and the disturbance level identifier corresponding to the working mode, and then extract the membership value of the current predicted load disturbance to the load level and the membership value of the current predicted disturbance to the disturbance level from the fuzzy membership results calculated in advance. The matching degree calculation can use the multiplication operation to multiply the load level membership and the disturbance level membership to obtain the comprehensive matching degree of the working mode, that is, only when the current working condition is highly matched with a working mode in both load and disturbance dimensions, the working mode can obtain a higher matching degree value.

[0151] In step S404, the weight coefficient of each first working mode can be determined based on the matching degree value, which reflects the contribution degree of each working mode in the formation process of the final target working mode. In order to ensure the rationality and consistency of the weight coefficient, the first processor normalizes the matching degree values of all first working modes, thereby ensuring the rationality of the numerical range of the target working mode control parameters obtained by final fusion. The size of the weight coefficient directly determines the influence degree of the control parameters of the corresponding working mode in the final fusion process. The working mode with higher matching degree obtains larger weight coefficient, and its control parameter has greater contribution in the target working mode, and vice versa. The working mode with lower matching degree obtains smaller weight coefficient, and its influence degree is correspondingly reduced.

[0152] Exemplarily, the weight of each candidate mode obtained by normalization can be represented as:

[0153]

[0154] In the formula, represents the weight coefficient of mode .

[0155] Specifically, the matching degree values of all the first working modes can be summed to obtain a matching degree total as a denominator reference for normalization. The matching degree total reflects the comprehensive correlation degree between the current working condition and all candidate working modes, and provides a unified normalization reference for subsequent weight allocation. Subsequently, the first processor calculates the weight coefficient corresponding to each first working mode by dividing the matching degree value of the working mode by the matching degree total. This normalization calculation ensures that the sum of all weight coefficients is strictly equal to one, which meets the mathematical constraint condition of weight allocation, while maintaining the relative importance relationship between the working modes unchanged.

[0156] The calculation result of the weight coefficient directly reflects the order of the applicability of each first working mode to the current working condition. The working mode with a larger weight coefficient value indicates that its control strategy has a higher matching degree with the current working condition, and the corresponding control parameter will play a dominant role in the formation process of the target working mode. Although the working mode with a smaller weight coefficient value has a relatively low matching degree with the current working condition, it can still provide certain control strategy supplement and correction.

[0157] In step S405, the control parameters of the corresponding working modes can be weighted and summed based on the weight coefficients of the first working modes to generate a target working mode adapted to the current working condition. The weighted sum process is performed for the proportional parameter, the integral parameter and the differential parameter respectively. The proportional parameter of the target working mode is obtained by multiplying the proportional parameters of each first working mode by the weight coefficient and then summing them up. The integral parameter and the differential parameter of the target working mode are calculated in the same way.

[0158] For example, the control parameter vector of each mode may be:

[0159]

[0160] In the formula, respectively represent the PID control parameters of mode .

[0161] The control parameters of the target working mode are:

[0162]

[0163] That is:

[0164]

[0165]

[0166]

[0167] In the formula, respectively represent the control parameters corresponding to the target working mode.

[0168] Specifically, a control parameter vector can be established for each first working mode in advance, which contains three core control elements of proportional parameter, integral parameter and derivative parameter corresponding to the working mode. The control parameters can be empirical values obtained through a large number of sea trial data and simulation optimization, which are specially calibrated and verified for specific load level and disturbance level combinations, and have good control performance and stability guarantee.

[0169] The control parameter vector of each working mode represents the optimal control strategy configuration under the corresponding working condition. The proportional parameter determines the response strength of the system to the current error, the integral parameter determines the elimination ability of the system to the historical accumulated error, and the derivative parameter determines the prediction and suppression ability of the system to the error change trend.

[0170] When performing weighted summation operation, first, the control parameter vectors of all first working modes are traversed, and the component-by-component multiplication operation is performed on the control parameter vector of each working mode and the corresponding weight coefficient to obtain the weighted contribution of the working mode in the target working mode. Subsequently, the weighted contribution of all working modes is classified and accumulated according to the parameter type, that is, the weighted proportional parameters of all working modes are summed to obtain the final proportional parameter of the target working mode, the weighted integral parameters of all working modes are summed to obtain the final integral parameter of the target working mode, and the weighted derivative parameters of all working modes are summed to obtain the final derivative parameter of the target working mode.

[0171] By adopting the technical scheme of the present application, the traditional hard switching mode selection can be changed into a soft switching strategy based on fuzzy membership, effectively avoiding the control oscillation problem in the critical state of the working condition. The mechanism of multi-mode candidate and weight fusion enables the system to fully utilize the control experience of multiple preset working modes and generate a customized control strategy more suitable for the current working condition through weighted combination. The technical scheme of the present application can realize smooth transition of control parameters, thereby improving the stability and adaptability of the control system during the working condition change process.

[0172] On the basis of the above-mentioned embodiments, as an optional embodiment, the feedback parameters include position error, speed error and swing angle; the position error represents the difference between the set position and the actual position, the speed error represents the change rate of the position error; and the swing angle is the offset angle of the actuator of the hoisting system relative to the vertical direction.

[0173] Specifically, in step S205, the step of optimizing the first control parameter based on the feedback parameter to obtain the second control parameter can further include the following steps:

[0174] ​Step S501, constructing a performance index function based on the position error, the speed error, and the swing angle;

[0175] Step S502, calculating partial derivatives of the performance index function with respect to each parameter component in the first control parameter; the parameter component includes a proportional parameter, an integral parameter, and a differential parameter;

[0176] Step S503, iteratively optimizing the first control parameter based on the partial derivatives to obtain a second control parameter.

[0177] In step S501, the first processor constructs a comprehensive performance index function based on three key feedback parameters, i.e., the position error, the speed error, and the swing angle, to realize unified quantitative evaluation of the position accuracy, the dynamic response, and the swing suppression capability of the control system.

[0178] The position error reflects the deviation degree between the actual position of the hoisting system actuator and the set position, and the size of the position error determines the positioning accuracy of the hoisting operation. The speed error, as an index of the dynamic response characteristic of the system, can reflect the tracking ability and response speed of the control system to the working condition change through the rate of change of the position error. The swing angle represents the degree of deviation of the actuator from the vertical direction under the action of external disturbance.

[0179] Optionally, the integral form in a specific time window is adopted when constructing the performance index function, and the multi-dimensional performance parameters are comprehensively evaluated in the time interval. The position error, the speed error, and the swing angle are respectively squared, and the square processing can eliminate the influence of the parameter symbol on the evaluation result, and at the same time, a higher penalty weight is given to the larger deviation, which meets the actual demand of the control system being more sensitive to large deviation.

[0180] The weight coefficient in the performance index function reflects the relative importance of different performance indexes in the specific working environment. The first weight coefficient is used to adjust the importance of the position accuracy in the comprehensive performance evaluation, the second weight coefficient is used to control the proportion of the dynamic response performance in the overall evaluation, and the third weight coefficient is used to adjust the attention degree of the swing suppression capability in the performance evaluation. The first processor converts the discrete instantaneous performance data into continuous comprehensive performance index by integrating the weighted squared error in the evaluation time window.

[0181] Exemplarily, the position error can be expressed as:

[0182]

[0183] In the formula, represents the position error, represents the set position, represents the actual position.

[0184] The speed error can be expressed as:

[0185]

[0186] The performance index function is defined within a time window

[0187]

[0188] In the formula, The performance index function is denoted by , , respectively, and T represents the length of the time window.

[0189] In step S502, the first processor needs to calculate the partial derivative of the performance index function with respect to each parameter component in the first control parameter to determine the optimization direction and adjustment amplitude of the control parameter adjustment. Since the first control parameter includes three core components of the proportional parameter, the integral parameter and the differential parameter, the influence mechanism and the sensitivity of each parameter component on the performance of the control system are different, and therefore the partial derivative of the performance index function with respect to each parameter component needs to be calculated respectively.

[0190] Exemplarily, the chain rule can be used when calculating the partial derivative, and the partial derivative of the performance index function with respect to the control parameter is decomposed into the product of the partial derivative of the performance index function with respect to the error term and the partial derivative of the error term with respect to the control parameter. The partial derivatives of the performance index function with respect to the proportional parameter, the integral parameter and the differential parameter can be calculated respectively.

[0191] The proportional parameter is the response gain of the controller to the current error, and the change of its value directly affects the amplitude of the control signal, and further affects the action strength of the actuator and the correction speed of the position error. The integral parameter can effectively eliminate the steady-state deviation of the system through the accumulation of historical error information, and the response degree of the error integral term to the change of the integral parameter is calculated by analyzing the dependence of the position error on the integral parameter. The differential parameter can timely apply a reverse control action at the initial stage of error increase through the prediction and compensation of the error change trend, and the response strength of the error differential term to the change of the differential parameter is calculated by analyzing the sensitivity of the position error to the differential parameter.

[0192] In the calculation process, the numerical differentiation method is used to approximate the partial derivative of the error term with respect to each control parameter. By performing a small perturbation near the current parameter value and observing the corresponding error change, numerical differentiation is performed using the central difference format to obtain higher calculation accuracy and numerical stability. The perturbation step can be adaptively adjusted to ensure that the perturbation amplitude can produce a detectable error change without introducing a large numerical error. ​

[0193] Exemplarily, the PID control law can be expressed as:

[0194]

[0195] wherein, represents the control signal, represents the proportional parameter, represents the integral parameter, represents the derivative parameter.

[0196] The partial derivative of the performance index function with respect to the control parameters is approximated as:

[0197]

[0198] In step S503, the first control parameters can be iteratively optimized based on the calculated partial derivative information, and the numerical values of each parameter component are gradually adjusted by the gradient descent algorithm to minimize the performance index function. The core idea of the iterative optimization process is to adjust the control parameters along the negative gradient direction of the performance index function, so that each parameter update can promote the performance improvement, and after multiple iterations, it converges to the locally optimal control parameter combination.

[0199] Exemplarily, when performing iterative optimization, an update strategy is adopted for the proportional parameter, the integral parameter and the derivative parameter respectively, and the optimization process of each parameter component follows the basic principle of gradient descent, that is, the current parameter value is reduced by the product of the partial derivative of the performance index function with respect to the parameter and the corresponding learning rate. The proportional parameter is updated by subtracting the product of the proportional parameter learning rate and the partial derivative of the performance index function with respect to the proportional parameter from the current proportional parameter value, and the integral parameter can be processed in the same way.

[0200] The learning rate plays a role in controlling the step size of parameter adjustment in the iterative optimization process, and setting independent learning rates for different parameter components can fully consider the differentiated characteristics of the influence of each parameter on the performance of the control system. The setting of the proportional parameter learning rate needs to consider the influence of the proportional control action on the stability of the system, the selection of the integral parameter learning rate needs to balance the effect of the integral action on the improvement of the steady-state performance and the possible problem of integral saturation, and the setting of the derivative parameter learning rate needs to consider the trade-off between the noise sensitivity of the derivative action and the improvement effect of the dynamic response.

[0201] The learning rate is adaptively adjusted when the parameter update is performed, and the learning rate value is dynamically corrected according to the change trend of the performance index function in the optimization process. When the performance index function continuously decreases and the decrease amplitude is relatively stable for several times of continuous iterations, the learning rate of the corresponding parameter can be appropriately increased to accelerate the optimization convergence process, and when the performance index function appears to be in a fluctuation or upward trend, the learning rate of the related parameter is timely reduced to improve the stability of the optimization process. After the parameter update is completed, the optimization result is subjected to boundary constraint processing, the updated control parameters are ensured to be within the preset safety range, and the parameter components subjected to the constraint processing are recombined to form the second control parameter.

[0202] Exemplarily, the gradient descent update rule can be represented as:

[0203]

[0204]

[0205]

[0206] In the formula, and respectively represent the corresponding learning rate, which is used to control the update step of the parameter, respectively represent the corresponding second control parameter after optimization.

[0207] By using the technical solution of the present application, the control strategy can be automatically adjusted according to the feedback information of the actual control effect, and the change from passive execution of preset parameters to active optimization of control performance is realized. The construction of the multi-dimensional performance index function enables the system to simultaneously optimize multiple control targets such as position accuracy, dynamic response and swing suppression, and avoids the deterioration of other performances that may be caused by single performance index optimization. The gradient optimization method based on partial derivatives provides a mathematical theoretical basis for parameter adjustment, and ensures the correctness and convergence of the optimization direction.

[0208] On the basis of the above embodiment, as an optional embodiment, step S503 can further include the following steps:

[0209] Step S601, calculating the update amount of each parameter component based on the partial derivative and the preset learning rate; the learning rate represents the step size of parameter update;

[0210] Step S602, subtracting the corresponding update amount from each parameter component in the first control parameter to obtain the optimized parameter component;

[0211] Step S603, combining the optimized parameter components to obtain the second control parameter. ​​​​​

[0212] In step S601, the learning rate represents the step size of parameter update, and its value directly affects the convergence performance and stability of iterative optimization. A too large learning rate can cause the parameters to oscillate around the optimal value and fail to converge, and a too small learning rate can make the optimization process converge slowly or even fall into a local optimum.

[0213] Exemplarily, independent learning rates can be set for the proportional parameter, the integral parameter and the derivative parameter, fully considering the differentiated characteristics of different parameter components on the performance of the control system. The setting of the proportional parameter learning rate needs to comprehensively consider the promoting effect of the proportional control action on the system response speed and the potential impact on the system stability, and is usually set to a medium value to balance the response performance and stability requirements.

[0214] Optionally, the selection of the integral parameter learning rate focuses on the progressive characteristics of the integral action on the elimination of steady-state error. Since the cumulative effect of the integral action has a strong persistent influence, the corresponding learning rate is usually set to a small value to avoid integral saturation problems. The setting of the derivative parameter learning rate needs to consider the characteristics of the derivative action on noise sensitivity. Since the derivative term is prone to amplify high-frequency noise interference, the corresponding learning rate setting needs to be moderately conservative to ensure the robustness of the control system.

[0215] Specifically, the specific update amount of each parameter component in the current iteration step can be calculated by multiplying the corresponding partial derivative of each parameter component and the corresponding learning rate. The proportional parameter update amount is obtained by multiplying the proportional parameter learning rate and the partial derivative of the performance index function with respect to the proportional parameter. The numerical size of the update amount reflects the magnitude of the proportional parameter that needs to be adjusted in the current iteration, and the sign of the update amount indicates the adjustment direction of whether the proportional parameter should be increased or decreased. The integral parameter update amount and the derivative parameter update amount use the same calculation logic and are obtained by multiplying the corresponding learning rate and the partial derivative, respectively, to form a complete update amount set of each parameter component.

[0216] In step S602, each parameter component in the first control parameter can be subtracted by the corresponding update amount to perform a specific parameter value update operation to obtain the optimized parameter component. The parameter update process follows the basic principle of the gradient descent algorithm, that is, adjusting the control parameter along the negative gradient direction of the performance index function, and moving the parameter in the direction of performance improvement through subtraction operation. Since the partial derivative of the performance index function indicates the direction in which the function value increases, and the optimization goal is to minimize the performance index function, the parameter needs to be adjusted in the direction in which the function value decreases by subtracting the update amount.

[0217] Optionally, an update operation can be performed on the proportional parameter, and a new proportional parameter value is obtained by subtracting the proportional parameter update amount from the current proportional parameter value. The above update method can automatically adjust the gain strength of the proportional control according to the feedback information of the current control performance. When the position error is large, the proportional parameter will be adjusted in the increasing direction to improve the response speed. When the system overshoots or oscillates, the proportional parameter will be adjusted in the decreasing direction to improve the stability. The integral parameter and the derivative parameter adopt the same update logic, and the optimized parameter values are obtained by subtracting the corresponding update amount. The update of the integral parameter mainly affects the elimination ability of the system to the steady-state error, and the update of the derivative parameter mainly affects the prediction and suppression effect of the system to the dynamic change.

[0218] Specifically, a boundary constraint check is performed immediately after the parameter update to ensure that each optimized parameter component is within the preset safe range. The setting of the boundary constraint is based on the stability requirement of the control system and the actual engineering experience. When the update result of a certain parameter component exceeds the safe boundary, the first processor automatically limits the parameter value within the boundary value range, avoiding the control failure or system instability problem caused by excessively large or small parameters.

[0219] In step S603, the parameter combination process needs to ensure that the optimized proportional parameter, integral parameter and derivative parameter maintain a reasonable numerical matching relationship, so as to avoid the imbalance of the overall control strategy caused by the over-optimization of a certain parameter component. A consistency check mechanism of parameter combination can be established to verify whether the optimized parameter combination meets the basic design requirements and performance index constraints of the control system.

[0220] Specifically, the second control parameter can be issued to the second processor through the industrial Ethernet to realize real-time update and execution of the control strategy. The second processor updates the internal control algorithm configuration immediately after receiving the second control parameter, recalculates the hydraulic control instruction based on the new parameter combination, and drives each actuator to perform the lifting operation according to the optimized control strategy.

[0221] By adopting the technical scheme of the present application, efficient control performance optimization can be realized under the premise of ensuring the stability of parameter update. The introduction of the learning rate mechanism makes the parameter adjustment process more controllable and predictable. The independent learning rate setting of the parameter components fully considers the physical characteristic differences of different control parameters, avoiding the imbalance of parameter adjustment caused by the uniform learning rate. The boundary constraint and consistency check mechanism provide multiple safety guarantees for the parameter optimization process, ensuring the stability and reliability of the control system during the performance improvement process.

[0222] On the basis of the above embodiment, as an optional embodiment, the actuator of the lifting system is powered by a hydraulic module; the hydraulic module is controlled by the second processor; and the first processor and the second processor are communicatively connected through a switch.

[0223] On the basis of the above-mentioned embodiments, as an optional embodiment, after determining the first control parameter corresponding to the target working mode, the following steps can be further included:

[0224] Step S701, generating a control signal based on the first control parameter; the control signal is composed of the sum of the product of the proportional parameter, the integral parameter and the differential parameter and the position error, the position error integral term and the position error rate of change, respectively;

[0225] Step S702, converting the control signal into a hydraulic control instruction; the hydraulic control instruction includes a valve opening degree instruction and a pump station pressure instruction, the valve opening degree instruction being the product of the control signal and a first proportional coefficient, and the pump station pressure instruction being the product of the control signal and a second proportional coefficient;

[0226] Step S703, sending the hydraulic control instruction to the hydraulic module to drive the actuator to move;

[0227] Step S704, acquiring the motion state parameters of the actuator by the second processor; the motion state parameters include the actual displacement, the actual speed, the actual load and the actual pressure;

[0228] Step S705, calculating the position error based on the difference between the set position and the actual displacement, and feeding back the position error to the second processor to update the control signal.

[0229] In step S701, the control signal is composed of the sum of the product of the proportional parameter, the integral parameter and the differential parameter and the position error, the position error integral term and the position error rate of change, respectively.

[0230] Among them, the proportional term provides an immediate response to the current deviation through the product of the proportional parameter and the current position error, and when the actual position deviates from the set position, a correction signal of corresponding strength can be generated, and the strength of the proportional action directly affects the response speed and steady-state accuracy of the system.

[0231] The integral term realizes the elimination of historical cumulative error through the product of the integral parameter and the position error integral term, and when there is a persistent deviation in the system, the integral action can gradually eliminate the steady-state error, ensuring that the hoisting device can finally accurately reach the set position.

[0232] The differential term provides predictive compensation for the error change trend through the product of the differential parameter and the position error rate of change, and when the position error shows a rapid increasing trend, the differential action can timely apply a reverse control force to effectively suppress the system overshoot and oscillation phenomenon.

[0233] Specifically, real-time acquisition of the current position error, calculation of the integral accumulation value of the position error, and solving of the rate of change of the position error are required when generating the control signal. The position error is calculated by setting the difference between the actual displacement and the actual displacement, and the sign of the value represents the deviation direction, and the value size represents the deviation degree. The integral term of the position error is obtained by time integration of the historical position error, and the selection of the integral time window needs to consider the control response speed and the risk of integral saturation. The rate of change of the position error is calculated by time differentiation of the position error, and appropriate filtering measures need to be adopted in the differentiation calculation process to reduce the interference of high-frequency noise on the differentiated signal.

[0234] In step S702, the hydraulic control instruction includes two core components, valve opening degree instruction and pump station pressure instruction, which control the flow distribution and pressure regulation of the hydraulic system respectively, and realize accurate control of the motion state of the actuator through coordination.

[0235] The valve opening degree instruction can be calculated by the product of the control signal and the first proportional coefficient. The instruction directly determines the opening degree of the hydraulic proportional valve and the flow size of the hydraulic oil.

[0236] The setting of the first proportional coefficient needs to be calibrated according to the flow characteristics of the hydraulic proportional valve and the speed requirement of the actuator, to ensure that the control signal can be accurately converted into the corresponding valve opening degree value.

[0237] When the control signal is positive, the valve opening degree instruction drives the actuator to move in the positive direction, and when the control signal is negative, the valve opening degree instruction drives the actuator to move in the opposite direction. The opening degree is directly proportional to the amplitude of the control signal.

[0238] The pump station pressure instruction is calculated by the product of the control signal and the second proportional coefficient. The instruction is used to adjust the output pressure of the hydraulic pump station to adapt to the power demand under different load conditions.

[0239] The selection of the second proportional coefficient needs to consider the pressure regulation range of the hydraulic pump and the load characteristics of the actuator, to ensure that sufficient hydraulic power can be provided under various working conditions.

[0240] Optionally, the first processor also needs to perform safety check and boundary limitation when generating the hydraulic control instruction, to ensure that the valve opening degree instruction and the pump station pressure instruction are within the safe working range of the hydraulic system.

[0241] When the calculated instruction value exceeds the rated parameter of the device, the first processor automatically limits the instruction within the safety boundary to avoid overloading or damage of the hydraulic system. At the same time, considering the dynamic response characteristics of the hydraulic components, the rate of change of the instruction is appropriately limited to prevent hydraulic impact and mechanical vibration caused by sudden change of the instruction.

[0242] In step S703, the generated hydraulic control command can be sent to the hydraulic module through industrial Ethernet, realizing reliable transmission of the control command from the decision layer to the execution layer. After receiving the valve opening degree command, the hydraulic module immediately adjusts the opening degree of the hydraulic proportional valve, and controls the flow size and flow direction of the hydraulic oil by changing the valve port area.

[0243] Optionally, the hydraulic module simultaneously adjusts the output pressure of the hydraulic pump according to the pump station pressure command, and ensures sufficient driving force under different load conditions through pressure regulation. When the load is large, the pump station pressure is automatically increased to overcome the load resistance, and when the load is small, the pump station pressure is appropriately reduced to improve energy efficiency and reduce system heating.

[0244] The actuator starts to perform the corresponding motion action according to the control command under the action of the hydraulic driving force, including the lifting motion of the bell winch, the winding and unwinding action of the guide cable winch, the tension adjustment of the umbilical winch, and the active compensation of the compensation device. The coordinated motion of each actuator ensures that the entire lifting and launching system can maintain a stable working state in complex marine environment, and realizes accurate launching and safe recovery of the underwater equipment.

[0245] In step S704, the second processor continuously monitors the actual running state of the equipment through a high-precision sensor array installed on each actuator, and the sensor types include displacement sensor, speed sensor, pressure sensor and load sensor and other detection devices.

[0246] The motion state parameters include actual displacement, actual speed, actual load and actual pressure and other state quantities. The actual displacement can be measured by a high-precision displacement sensor, and this parameter represents the actual displacement of the actuator relative to the reference position, and the displacement measurement accuracy directly determines the accuracy of position control.

[0247] The actual speed can be obtained by differentiating the speed sensor or displacement sensor, and the speed information reflects the motion speed and dynamic response characteristics of the actuator, and provides an important basis for system dynamic performance evaluation.

[0248] The actual load can be indirectly measured by a load sensor or a pressure sensor, and the load information represents the current work intensity and external resistance, and provides real-time data support for load adaptability control.

[0249] The actual pressure can be directly measured by a pressure sensor of the hydraulic system, and the pressure parameter reflects the working state and energy output level of the hydraulic system.

[0250] The second processor pre-processes and filters the collected raw sensor data, eliminates the interference of measurement noise and abnormal values, and ensures the reliability and consistency of the state parameters. The data pre-processing process includes zero point calibration, range calibration, temperature compensation and linearization processing, etc. Through systematic data processing, the measurement accuracy and stability are improved.

[0251] In step S705, the second processor calculates the current position error based on the difference between the set position and the actual displacement, and feeds back the error information to the control system to support real-time updating of the control signal. The position error is the core feedback signal of the closed-loop control system, and its calculation accuracy and feedback timeliness directly affect the stability and control effect of the entire control loop.

[0252] Specifically, the position error is calculated by subtracting the actual displacement value measured by the sensor from the preset set position value. The second processor needs to consider the consistency of the coordinate system and the unity of the measurement reference when calculating the position error, to ensure that the set position and the actual displacement use the same reference frame and measurement unit.

[0253] The second processor can transmit the calculated position error information to the first processor in real time through industrial Ethernet, providing accurate feedback data for parameter optimization of the control algorithm and updating of the control signal. The feedback transmission process uses a high-priority communication mechanism to ensure the timeliness of the error information and avoid the adverse effects of communication delay on control performance. After receiving the position error feedback, the first processor immediately starts a new round of control signal calculation, generates a new control signal based on the updated error information, and forms a complete closed-loop control loop.

[0254] By using the technical solution of the present application, a complete control link from control parameter determination to actuator action to state feedback is realized, ensuring effective implementation and dynamic adjustment of the control strategy. The application of the standard PID control algorithm ensures the theoretical correctness and engineering reliability of the control signal generation, and the accurate conversion of the hydraulic control command realizes the effective connection between the control algorithm and the physical execution layer. Real-time state parameter acquisition and position error feedback mechanism enable the system to timely perceive the execution effect and make corresponding adjustments, forming a self-adaptive and optimized closed-loop control characteristic.

[0255] Figure 3 A structural block diagram of a submersible hoisting and releasing control device provided by the present application is schematically shown, which can include:

[0256] The first acquisition module is configured to acquire load interference and disturbance interference of the hoisting and releasing system within a first time length.

[0257] The first determining module is configured to determine, based on the load interference and the disturbance interference, the working mode of the slinging system as a target working mode; the slinging system supports multiple different working modes, and the slinging system has different working performances in different working modes;

[0258] The second determining module is configured to determine, based on the target working mode, a first control parameter corresponding to the target working mode, and control the actuator of the slinging system based on the first control parameter.

[0259] The second obtaining module is configured to obtain, within a second time length, a feedback parameter of the slinging system; the second time length is shorter than the first time length.

[0260] The optimization module is configured to optimize the first control parameter based on the feedback parameter to obtain a second control parameter, and control the actuator of the slinging system based on the second control parameter.

[0261] Figure 4 A structural block diagram of an electronic device suitable for implementing the method described above according to an embodiment of the present application is schematically shown. Figure 4 The electronic device shown is merely an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0262] As shown in Figure 4 The electronic device according to an embodiment of the present application includes a processor 401, which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 402 or loaded from a storage portion 408 into a random access memory (RAM) 403. The processor 401 can include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor, and / or a related chipset, and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 401 can also include an on-board memory for cache use. The processor 401 can include a single processing unit or multiple processing units for performing different actions of the method processes according to embodiments of the present application.

[0263] In the RAM 403, various programs and data required for the operation of the electronic device are stored. The processor 401, the ROM 402, and the RAM 403 are connected to each other through a bus 404. The processor 401 performs various operations of the method processes according to embodiments of the present application by executing the programs in the ROM 402 and / or the RAM 403. It should be noted that the programs can also be stored in one or more memories other than the ROM 402 and the RAM 403. The processor 401 can also perform various operations of the method processes according to embodiments of the present application by executing the programs stored in the one or more memories.

[0264] According to an embodiment of the present application, the electronic device can further include an input / output (I / O) interface 405 that is also connected to the bus 404. The electronic device can further include one or more of the following components connected to the input / output (I / O) interface 405: an input part 406 including, for example, a keyboard and a mouse; an output part 407 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage part 408 including, for example, a hard disk; and a communication part 409 including, for example, a LAN card, a modem, and the like. The communication part 409 performs communication processing via a network such as the Internet. A driver 410 is also connected to the input / output (I / O) interface 405 as necessary. A removable medium 411 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is mounted on the driver 410 as necessary, so that a computer program read therefrom is installed in the storage part 408 as necessary.

[0265] The present application also provides a computer readable storage medium, which can be included in the device / system described in the above embodiments; or can exist independently without being assembled into the device / system. The above computer readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of the present application.

[0266] According to an embodiment of the present application, the computer readable storage medium can be a non-volatile computer readable storage medium. For example, it can include, but is not limited to: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any appropriate combination thereof. In this application, a computer readable storage medium can be any tangible medium that contains or stores a program for use by or in connection with an instruction execution system, system or device.

[0267] For example, according to an embodiment of the present application, the computer readable storage medium can include one or more memories of the ROM 402 and / or the RAM 403 described above and / or one or more memories other than the ROM 402 and the RAM 403.

[0268] The embodiments of the present application also include a computer program product, which includes a computer program containing a program code for executing the method provided by the embodiments of the present application, and when the computer program product is run on the electronic device, the program code is used to make the electronic device implement the method provided by the embodiments of the present application.

[0269] When the computer program is executed by the processor 401, the above-described functions defined in the system / system of the embodiments of the present application are performed. According to the embodiments of the present application, the system, system, module, unit, etc. described above can be determined by computer program modules.

[0270] In one embodiment, the computer program can rely on a tangible storage medium, such as an optical storage device, a magnetic storage device, etc. In another embodiment, the computer program can also be transmitted in the form of a signal over a network medium, distributed, and downloaded and installed through the communication part 409, and / or installed from the detachable medium 411. The program code contained in the computer program can be transmitted by any suitable network medium, including but not limited to wireless, wired, etc., or any suitable combination of the above.

[0271] The embodiments of the present application are described above. However, these embodiments are only for illustrative purposes, and are not intended to limit the scope of the present application. Although each embodiment is described above separately, this does not mean that the measures in each embodiment cannot be used advantageously in combination. The scope of the present application is defined by the appended claims and their equivalents. Without departing from the scope of the present application, those skilled in the art can make various substitutions and modifications, which should all fall within the scope of the present application.

Claims

1. A method of handling a suspended submersion, characterized in that, A first processor applied to a ship's launch system, the method comprising: obtaining load disturbance and disturbance disturbance of the launch system within a first time length; determining the working mode of the launch system as the target working mode based on the load disturbance and the disturbance disturbance; the launch system supports multiple different working modes, and has different working performances under different working modes; determining the first control parameter corresponding to the target working mode to control the actuator of the launch system based on the first control parameter; obtaining the feedback parameter of the launch system within a second time length; the second time length is less than the first time length; optimizing the first control parameter based on the feedback parameter to obtain the second control parameter to control the actuator of the launch system based on the second control parameter.

2. The method of claim 1, wherein, Obtaining the load disturbance and disturbance disturbance of the launch system within a first time length comprises: obtaining the current state signal of the launch system; the state signal includes load signal, displacement signal and inclination signal; the load signal is taken as a load sequence, and the inclination signal and the displacement signal are taken as a disturbance sequence; based on the load sequence, the load disturbance within the first time length is predicted; based on the disturbance sequence, the disturbance disturbance within the first time length is predicted.

3. The method of claim 1, wherein, Based on the load disturbance and the disturbance disturbance, the working mode of the launch system is determined, comprising: determining the load level of the load disturbance and the disturbance level of the disturbance disturbance, respectively; determining at least two first working modes corresponding to the load level and the disturbance level in the multiple working modes supported by the launch system; calculating the matching degree of the load level and the disturbance level and each working mode; determining the weight coefficient of each first working mode based on the matching degree; based on the weight coefficient of each first working mode, the control parameters corresponding to each first working mode are weighted and summed to obtain the target working mode.

4. The method of claim 1, wherein, The feedback parameter includes position error, velocity error and swing angle; the position error represents the difference between the set position and the actual position, and the velocity error represents the change rate of the position error; swing angle is the offset angle of the actuator of the launch system relative to the vertical direction; Based on the feedback parameter, the first control parameter is optimized to obtain the second control parameter, comprising: constructing a performance index function based on the position error, the velocity error and the swing angle; calculating the partial derivative of the performance index function to each parameter component in the first control parameter; the parameter component includes proportional parameter, integral parameter and differential parameter; iteratively optimizing the first control parameter based on the partial derivative to obtain the second control parameter.

5. The method of claim 4, wherein, Based on the partial derivative, the first control parameter is iteratively optimized to obtain the second control parameter, comprising: calculating the update amount of each parameter component based on the partial derivative and the preset learning rate; the learning rate represents the step size of parameter update; subtracting the corresponding update amount from each parameter component in the first control parameter to obtain the optimized parameter component; Combine the optimized parameter components to obtain the second control parameter.

6. The method of claim 1, wherein, The actuator of the hoisting system is powered by a hydraulic module; the hydraulic module is controlled by a second processor; the first processor and the second processor are communicatively connected through a switch.

7. The method of claim 6, wherein, After determining the first control parameter corresponding to the target working mode, the method further includes: Generating a control signal based on the first control parameter; the control signal is composed of the sum of the product of the proportional parameter, the integral parameter and the derivative parameter and the position error, the position error integral term and the position error rate, respectively; Converting the control signal into a hydraulic control instruction; the hydraulic control instruction includes a valve opening degree instruction and a pump station pressure instruction, the valve opening degree instruction is the product of the control signal and a first proportional coefficient, and the pump station pressure instruction is the product of the control signal and a second proportional coefficient; Sending the hydraulic control instruction to the hydraulic module to drive the actuator to move; Obtaining the motion state parameter of the actuator through the second processor; the motion state parameter includes actual displacement, actual speed, actual load and actual pressure; Calculating the position error based on the difference between the set position and the actual displacement, and feeding back the position error to the second processor to update the control signal.

8. A deployment and handling device for submersion, characterized in that Comprise: The first acquisition module is used for acquiring the load disturbance and the disturbance disturbance of the hoisting system within a first time length; The first determination module is used for determining that the working mode of the hoisting system is a target working mode based on the load disturbance and the disturbance disturbance; the hoisting system supports multiple different working modes, and the hoisting system has different working performances in different working modes; The second determination module is used for determining the first control parameter corresponding to the target working mode based on the first control parameter, and controlling the actuator of the hoisting system based on the first control parameter; The second acquisition module is used for acquiring the feedback parameter of the hoisting system within a second time length; the second time length is less than the first time length; The optimization module is used for optimizing the first control parameter based on the feedback parameter to obtain a second control parameter, and controlling the actuator of the hoisting system based on the second control parameter. 9.An electronic device, comprising: one or more processors; memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1 to 7. 10.A computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the method of any one of claims 1 to 7.

Images