A control method for active suppression of ship roll
By calculating the effective matching score and energy consumption value of the ship's roll control system, the degree of servo action following the wave trend is identified, which solves the problem of poor control effect caused by mechanical response delay, realizes adaptive adjustment, and improves the roll reduction effect.
Patent Information
- Application Number
- CN202611626915.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-21
- Publication Date
- 2026-11-17
AI Technical Summary
In existing technologies for ship roll control, the mechanical response delay caused by high-frequency broken waves results in the reverse torque not being able to keep up with the command changes in a timely manner, thus reducing the roll reduction control effect.
By calculating the effective matching score and energy consumption value of each sampling beat, the degree to which the servo action follows the main wave trend is identified, and invalid energy consumption is corrected to achieve adaptive adjustment and avoid mechanical wear and energy waste.
It improves the roll control effect in complex sea conditions, protects mechanical components and maintains effective roll reduction, and avoids energy waste and mechanical wear caused by the delay of reverse torque.
Smart Images

Figure CN123104870A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of ship roll control technology, specifically to a control method for actively suppressing ship roll. Background Technology
[0002] When ships are sailing at sea, they are prone to rolling motion due to the vibration of ocean waves. To ensure navigation safety and operational efficiency, they are usually equipped with active roll damping systems driven by hydraulics or electric motors, which suppress hull rolling by outputting a counter-torque. However, the actual ocean wave environment has complex multi-scale characteristics, with low-frequency swells and high-frequency fractals superimposed, resulting in a large number of high-frequency components mixed in with the roll attitude signal.
[0003] Existing technologies typically rely on ship roll attitude signals for roll reduction control. Specifically, the control system directly reads the roll angle and its angular velocity as feedback quantities. After calculating the theoretical feedback torque using a proportional-derivative (PD) control law, it drives a servo mechanism to output a counter-torque to suppress hull sway. However, when encountering high-frequency breaking waves, the rate of change of the high-frequency components in the roll attitude signal exceeds the mechanical response capability of the servo mechanism. This causes the actual output counter-torque to fail to keep up with the command changes in a timely manner, resulting in a mechanical response delay and poor roll reduction control performance. Summary of the Invention
[0004] To address the technical problem that existing technologies typically rely on ship roll attitude signals for roll reduction control, resulting in poor roll reduction control performance because the actual output counter-torque cannot keep up with command changes in a timely manner, the present invention aims to provide a control method for active roll suppression of ships. The specific technical solution adopted is as follows: This invention proposes a control method for actively suppressing ship roll, the method comprising: At each sampling cycle, the roll angle, roll rate, feedback deflection angle, deflection rate, and drive current of the target ship and the servo mechanism are obtained. Based on the sign consistency between the roll angle and the roll angular velocity, and between the roll angle and the feedback deflection angle, a local matching score is calculated to determine the effective matching score for each sampling cycle; based on the deflection velocity and the drive current, the energy consumption value for each sampling cycle is determined; and the energy consumption values of all historical sampling cycles are summed up by invalid energy consumption to determine the invalid energy consumption for each sampling cycle. Based on the effective matching score, the invalid energy consumption is corrected to determine the invalid energy loss ratio for each sampling cycle; and active suppression is performed on each sampling cycle based on the invalid energy loss ratio.
[0005] Furthermore, the method for obtaining the effective matching score includes: Obtain the sign function values of the roll angle, roll angular velocity, and feedback deflection angle respectively; determine the roll direction identifier for each sampling beat based on the similarity of the sign function values between the roll angle and the roll angular velocity; determine the servo direction identifier for each sampling beat based on the similarity of the sign function values between the roll angle and the feedback deflection angle. Extract the roll direction identifier and servo direction identifier of each neighboring sampling beat within a preset time window for each sampling beat; for each neighboring sampling beat, if the roll direction identifier is opposite to the servo direction identifier, assign its single-beat matching score a preset positive value; if the roll direction identifier is the same as the servo direction identifier, assign its single-beat matching score a preset negative value. The single-beat matching scores of all neighboring sampling beats within the preset time window of each sampling beat are accumulated and normalized to obtain the effective matching score of each sampling beat.
[0006] Furthermore, the method for obtaining the ineffective energy consumption includes: For each sampling cycle, extract all historical sampling cycles with the same roll direction identifier and servo direction identifier as invalid energy consumption cycles; sum up the energy consumption values of all invalid energy consumption cycles to obtain the invalid energy consumption of each sampling cycle.
[0007] Furthermore, the method for obtaining the proportion of ineffective energy loss includes: For each sampling cycle, the invalid energy consumption is normalized and mapped to determine the basic energy loss ratio for each sampling cycle. Based on the basic energy loss ratio and the effective matching score, the invalid energy loss ratio for each sampling cycle is determined. The invalid energy loss ratio is positively correlated with the basic energy loss ratio and negatively correlated with the effective matching score.
[0008] Furthermore, the method for obtaining the basic energy loss ratio includes: For each sampling cycle, the energy consumption values of all historical sampling cycles are summed to obtain the total operating energy consumption of each sampling cycle; the proportion of the invalid energy consumption in the total operating energy consumption is used as the basic energy loss ratio for each sampling cycle.
[0009] Furthermore, the process of actively suppressing each sampling cycle based on the energy wastage ratio includes: When the energy wastage ratio is greater than a preset upper limit threshold, a gain coefficient is determined based on the energy wastage ratio; the gain coefficient is then used to attenuate the amplitude of the controller's output command. When the energy ineffective loss ratio is less than the preset lower threshold, the controller outputs the command at full power. When the energy ineffective loss ratio is between a preset lower threshold and a preset upper threshold, the operating mode of the previous sampling cycle remains unchanged.
[0010] Furthermore, the gain coefficient is the difference between a preset base value and the proportion of ineffective energy loss.
[0011] Furthermore, during the output command process of the full-power output controller, the gain coefficient is 1.
[0012] Furthermore, the energy consumption value is the absolute value of the product between the yaw rate and the drive current.
[0013] Furthermore, the method for obtaining the roll direction indicator is as follows: If the sign function value of the roll angle is the same as that of the roll angular velocity, then the roll direction identifier of each sampling beat is assigned a preset positive identifier; if the sign function value of the roll angle is different from that of the roll angular velocity, then the roll direction identifier of each sampling beat is assigned a preset negative identifier.
[0014] The present invention has the following beneficial effects: This invention effectively identifies the degree of following between servo actions and the main wave trend by calculating the effective matching score of each sampling beat, solving the problem of comparison calculation failure caused by mechanical response delay in existing technologies. By calculating the energy consumption value of each sampling beat and superimposing the invalid energy consumption of historical energy consumption, the invalid energy consumption of each sampling beat is determined, realizing the quantitative identification and accumulation of invalid energy consumption of the servo mechanism. By using the effective matching score to correct invalid energy consumption, the invalid energy loss ratio is output, and active suppression is performed on each sampling beat based on the invalid energy loss ratio. This enables the control system to adaptively adjust between protecting mechanical components and maintaining effective roll reduction according to the real-time energy consumption status of the servo mechanism, avoiding the energy waste and mechanical wear caused by the output reverse torque not being able to follow the command changes in time in existing solutions, thus improving the roll reduction control effect. Attached Figure Description
[0015] Figure 1 A flowchart of a control method for active suppression of ship roll provided in one embodiment of the present invention; Figure 2 This is a flowchart of an effective matching score acquisition method provided in one embodiment of the present invention. Detailed Implementation
[0016] The specific scheme of the active control method for suppressing ship roll provided by the present invention will be described in detail below with reference to the accompanying drawings.
[0017] Please see Figure 1The diagram illustrates a flowchart of a control method for active roll suppression of a ship according to an embodiment of the present invention. The method includes: S101: At each sampling cycle, acquire the roll angle, roll rate, feedback deflection angle, deflection rate, and drive current of the target ship and the servo mechanism.
[0018] During maritime navigation, ships frequently encounter a typical scenario where low-frequency swells and high-frequency fracturing waves overlap. Low-frequency swells cause the hull to roll slowly and significantly, while high-frequency fracturing waves exert rapidly changing excitation forces on the hull within a short period, resulting in a large number of high-frequency components mixed in with the roll attitude signal. In this scenario, the roll angle and roll velocity reflect the excitation state of the hull by the waves, while the feedback deflection angle, deflection velocity, and drive current reflect the current counter-current state of the servo mechanism. Since wave excitation and servo counter-current are manifestations of the same physical process in two dimensions, this embodiment of the invention acquires the roll angle and roll velocity of the target ship and the feedback deflection angle, deflection velocity, and drive current of the servo mechanism at each sampling cycle, thereby establishing a temporal correspondence between the wave excitation direction and the servo counter-current direction.
[0019] In one specific implementation of this invention, the system sets the sampling period to 0.02 seconds, meaning the controller generates a sampling cycle every 0.02 seconds. At the arrival of each sampling cycle, the controller concurrently performs the following read operations: reads the roll angle and roll velocity of the current sampling cycle from the inertial measurement unit via the bus interface; simultaneously reads the feedback deflection angle and deflection velocity of the same sampling cycle via the servo driver; and reads the drive current of the same sampling cycle via the current sensor. These five parameters are then uniformly bound to the timestamp of the current sampling cycle and stored in the controller's memory cache for subsequent steps.
[0020] S102: Based on the sign consistency between the roll angle and the roll angular velocity, and between the roll angle and the feedback deflection angle, perform local matching scores to determine the effective matching score for each sampling cycle; based on the deflection velocity and the drive current, determine the energy consumption value for each sampling cycle; perform invalid energy consumption summation on the energy consumption values of all historical sampling cycles to determine the invalid energy consumption for each sampling cycle.
[0021] In sea conditions where low-frequency swells and high-frequency breaking waves overlap, the essential relationship between wave excitation and servo countermeasures is directional: the consistency of the sign of the roll angle and roll velocity reflects whether the waves are currently intensifying the roll or naturally righting the ship; the consistency of the sign of the roll angle and feedback deflection angle reflects whether the servo torque is currently counteracting or conforming to the roll. Furthermore, considering the extreme amplitude fluctuations in single-beat scenarios with frequent breaking waves, if amplitude is used as the evaluation criterion, the instantaneous large amplitude caused by a single breaking wave will mask the effective countermeasure information of the previous dozens of beats, distorting the evaluation results. Therefore, this embodiment of the invention isolates the amplitude and analyzes from a directional perspective. Based on the consistency of the signs between the roll angle and roll velocity, and between the roll angle and feedback deflection angle, a local matching score is calculated to determine the effective matching score for each sampling beat, thereby quantifying the degree of following between the servo action and the main wave trend.
[0022] When waves frequently impact the hull, the servo mechanism needs to repeatedly change direction in a short period to try to follow the attitude changes. At this time, both the drive current and the yaw rate are at a high level, and energy consumption increases sharply. Therefore, this embodiment of the invention determines the energy consumption value of each sampling cycle based on the yaw rate and the drive current, quantifying the energy consumed by each sampling cycle to counteract roll. However, this increase in energy consumption does not represent an improvement in counteracting effectiveness. On the contrary, a considerable portion of the energy is consumed in erroneous movements aligned with the wave direction, meaning that instead of suppressing roll, it actually contributes to roll at a certain moment. Furthermore, considering that the energy consumption value of a single cycle only reflects the intensity of the movement at that instant and cannot reflect the cumulative trend of ineffective energy consumption over a period of time, this embodiment of the invention superimposes the ineffective energy consumption values of all historical sampling cycles to determine the ineffective energy consumption of each sampling cycle, thereby quantifying the energy loss of the servo mechanism in ineffective movements with incorrect direction over a past period.
[0023] S103: Based on the effective matching score, correct the invalid energy consumption and determine the invalid energy loss ratio for each sampling cycle; actively suppress each sampling cycle based on the invalid energy loss ratio.
[0024] When a ship encounters brief disturbances from broken waves, the servo mechanism can still effectively follow the low-frequency main wave for most of the time. However, some beats may experience directional misalignment due to the impact of broken waves, resulting in potentially high ineffective energy consumption, but also a high effective matching score. When the ship is exposed to high-frequency broken wave disturbances for an extended period, the servo mechanism continuously and frequently changes direction, leading to fragmented movements. In this case, ineffective energy consumption is high and the effective matching score is low. If only ineffective energy consumption is used as the evaluation criterion, both scenarios would be treated equally, and the system would trigger gain attenuation in both cases. However, in the brief broken wave scenario, the servo mechanism still possesses effective countermeasure capabilities, and attenuating the gain would actually weaken the suppression effect on the low-frequency main wave. Therefore, this embodiment of the invention integrates the evaluation at the energy consumption level with the evaluation at the direction level. Based on the effective matching score, the ineffective energy consumption is corrected to determine the proportion of ineffective energy loss for each sampling beat. Based on the proportion of ineffective energy loss, each sampling beat is actively suppressed, thereby ensuring that the system can actively reduce the amplitude of movement when the servo mechanism is caught in high-frequency ineffective directional changes. This achieves a dynamic balance between protecting mechanical components and maintaining effective roll reduction, improving the roll reduction control effect under complex sea conditions.
[0025] In summary, this invention effectively identifies the degree of following between servo actions and the main wave trend by calculating the effective matching score of each sampling beat, thus solving the problem of comparison calculation failure caused by mechanical response delay in existing technologies. By calculating the energy consumption value of each sampling beat and superimposing the historical energy consumption with invalid energy consumption, the invalid energy consumption of each sampling beat is determined, realizing the quantitative identification and accumulation of invalid energy consumption of the servo mechanism. By using the effective matching score to correct invalid energy consumption, the invalid energy loss ratio is output, and active suppression is performed on each sampling beat based on the invalid energy loss ratio. This enables the control system to adaptively adjust between protecting mechanical components and maintaining effective roll reduction according to the real-time energy consumption status of the servo mechanism, avoiding the energy waste and mechanical wear caused by the output reverse torque not being able to follow the command changes in time in existing solutions, thus improving the roll reduction control effect.
[0026] Preferably, in some possible implementations of the embodiments of the present invention, the specific process for obtaining the effective matching score includes: Please refer to Figure 2 The diagram illustrates a flowchart of an effective matching score acquisition method provided by an embodiment of the present invention, the method comprising: S201: Obtain the sign function values of the roll angle, roll angular velocity, and feedback deflection angle respectively; determine the roll direction identifier for each sampling beat based on the similarity of the sign function values between the roll angle and the roll angular velocity; determine the servo direction identifier for each sampling beat based on the similarity of the sign function values between the roll angle and the feedback deflection angle.
[0027] Low-frequency swells cause the hull to roll slowly and significantly, while high-frequency breaking waves exert rapidly changing excitation forces on the hull within a short period. At this time, the sign function values of the roll angle and roll velocity have clear physical meanings: when the sign function values are the same, it indicates that the roll angle is increasing and the roll velocity is in the same direction, meaning the waves are intensifying the roll; when the sign function values are different, it indicates that the roll angle is decreasing and the roll velocity is in the opposite direction, meaning the hull is naturally righting itself. Therefore, this embodiment of the invention obtains the sign function values of the roll angle, roll velocity, and feedback deflection angle respectively; based on the similarity of the sign function values between the roll angle and the roll velocity, the roll direction identifier for each sampling cycle is determined. Similarly, the similarity of the sign function values between the roll angle and the feedback deflection angle reflects whether the servo torque is currently resisting or yielding to the roll; therefore, this embodiment of the invention determines the servo direction identifier for each sampling cycle based on the similarity of the sign function values between the roll angle and the feedback deflection angle. Specifically, if the sign function value of the roll angle is the same as that of the roll angular velocity, then the roll direction identifier of each sampling beat is assigned a preset positive identifier; if the sign function value of the roll angle is different from that of the roll angular velocity, then the roll direction identifier of each sampling beat is assigned a preset negative identifier.
[0028] In one specific implementation of this invention, a sign function (Sgn function in this embodiment) is used to extract the sign function values of the roll angle, roll velocity, and feedback deflection angle. When the sign function values of the roll angle and roll velocity are the same, the roll direction identifier for each sampling cycle is marked as 1; when the sign function values of the roll angle and roll velocity are different, the roll direction identifier for each sampling cycle is marked as -1; specifically, when the sign function values of both the roll angle and roll velocity are 0, the roll direction identifier for each sampling cycle is marked as 0. Similarly, when the sign function values of the roll angle and feedback deflection angle are the same, the servo direction identifier for each sampling cycle is marked as 1; when the sign function values of the roll angle and feedback deflection angle are different, the servo direction identifier for each sampling cycle is marked as -1; specifically, when the sign function values of both the roll angle and feedback deflection angle are 0, the servo direction identifier for each sampling cycle is marked as 0.
[0029] It should be noted that when the roll direction indicator or servo direction indicator is marked as 0, it indicates that the hull is not rolling or the servo mechanism is not resisting, so this embodiment of the invention will not analyze it.
[0030] S202: Extract the roll direction identifier and servo direction identifier of each neighboring sampling beat within a preset time window for each sampling beat; for each neighboring sampling beat, if the roll direction identifier is opposite to the servo direction identifier, then assign its single-beat matching score a preset positive value score; if the roll direction identifier is the same as the servo direction identifier, then assign its single-beat matching score a preset negative value score.
[0031] Considering the superposition of high and low frequency sea conditions, the servo mechanism may experience brief timing misalignments due to mechanical response delays. For example, after effective countermeasures in multiple consecutive sampling beats, a single wave impact may cause the servo action to lag by one beat. In this case, if only the directional relationship of the current sampling beat is examined, this misalignment will lead the system to misjudge the countermeasures as a failure. Therefore, this embodiment of the invention extends the scope of examination to each neighboring sampling beat within a preset time window. By extracting the roll direction identifier and servo direction identifier of each neighboring sampling beat within the preset time window for each sampling beat, the overall relationship between servo action and wave trend can be observed on a longer time scale. Specifically: for each neighboring sampling beat, if the roll direction indicator is opposite to the servo direction indicator, it means that the servo torque direction in that beat is in opposition to the wave excitation direction, which is an effective counteraction action, and its single-beat matching score is assigned a preset positive score; if the roll direction indicator is the same as the servo direction indicator, it means that the servo torque direction in that beat is consistent with the wave excitation direction, which is an invalid counteraction action, and its single-beat matching score is assigned a preset negative score.
[0032] In one specific implementation of this invention, taking each sampling beat as the endpoint, all sampling beats within the previous 5 seconds are extracted (corresponding to 250 sampling beats in this embodiment), and a preset time window for each sampling beat is constructed. Simultaneously, to ensure that when the servo mechanism primarily engages in effective counter-attacks within the time window, the accumulation of positive scores should significantly offset the negative scores resulting from occasional ineffective counter-attacks, thereby maintaining the effective matching score at a high level; conversely, when the servo mechanism frequently falls into ineffective reversals, the accumulation of negative scores should quickly lower the total score, causing a significant decrease in the effective matching score. The reward for effective counter-attacks should be greater than the penalty for ineffective counter-attacks. Therefore, this embodiment sets the preset positive score to 2 and the preset negative score to -1. That is, for each neighboring sampling beat, if the roll direction indicator is opposite to the servo direction indicator, its single-beat matching score is assigned a value of 2; if the roll direction indicator is the same as the servo direction indicator, its single-beat matching score is assigned a value of -1.
[0033] It should be noted that for the sampling clock within the first 5 seconds of the initial startup of the servo mechanism, since the accumulated sampling data in its historical buffer is insufficient to cover a complete time window and cannot provide effective data reference, this embodiment of the invention does not analyze it.
[0034] S203: Accumulate the single-beat matching scores of all neighboring sampling beats within the preset time window for each sampling beat and perform normalization mapping to obtain the effective matching score for each sampling beat.
[0035] Based on the independent scoring described above, the single-beat matching scores of all neighboring sampling beats within the time window are accumulated and normalized to obtain the effective matching score for each sampling beat. The physical meaning is: if the servo mechanism primarily engages in effective countermeasures within the time window, the accumulated positive scores will result in a higher total score; if the servo mechanism frequently falls into ineffective commutation, the negative scores will cancel out the positive scores, significantly reducing the total score. This accumulation method allows for a direct quantification of the net contribution between effective and ineffective countermeasure actions of the servo mechanism over a given period. A higher effective matching score indicates a larger proportion of effective countermeasure actions within that time window.
[0036] In one specific implementation of this invention, the single-beat matching scores of all neighboring sampling beats within a preset time window for each sampling beat are accumulated to obtain the historical cumulative score value for each sampling beat. The min-max normalization algorithm is then used to map the historical cumulative score value to the range [0,1] to obtain the effective matching score for each sampling beat. The minimum and maximum values used in the min-max normalization algorithm can be obtained by statistically analyzing the historical minimum and maximum values of the historical cumulative scores of all historical sampling beats in the historical running database. If the historical cumulative score value is less than the historical minimum value, its effective matching score is set to 0; if the historical cumulative score value is greater than the historical maximum value, its effective matching score is set to 1.
[0037] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the ineffective energy consumption includes: Servo mechanisms typically output a continuous counter-torque to suppress ship roll. However, due to the rate of attitude change caused by breaking waves exceeding the mechanical response capability of the servo mechanism, the servo action may exhibit directional misalignment in some cycles. That is, the direction of the servo torque may align with the direction of wave excitation, not only failing to suppress roll but actually contributing to it at that moment. The energy consumed by the servo mechanism in this case is considered ineffective energy loss. Therefore, in this embodiment of the invention, for each sampling cycle, all historical sampling cycles where the roll direction identifier and the servo direction identifier are the same are extracted as ineffective energy loss cycles. The energy consumption values of all ineffective energy loss cycles are accumulated to obtain the ineffective energy consumption of each sampling cycle, thereby identifying and extracting this part of the directionally incorrect energy consumption, and truly reflecting the degree of ineffective loss of the servo mechanism.
[0038] Specifically, considering that during the actual operation of a ship roll reduction system, the servo mechanism outputs a deflection action under the drive of the controller command. Its deflection angular velocity reflects the speed of movement of the mechanical spindle or hydraulic fork, while the drive current reflects the electromagnetic torque or hydraulic torque level output by the servo motor or hydraulic servo valve to maintain this movement. Physically, the product of the drive current and the deflection angular velocity corresponds to the instantaneous mechanical power output by the servo mechanism per unit time, that is, the instantaneous energy consumed by the servo mechanism to complete the current deflection action. Therefore, in this embodiment of the invention, the energy consumption value is the absolute value of the product between the deflection angular velocity and the drive current.
[0039] It should be noted that the product of the driving current and the angular velocity is numerically proportional to the instantaneous mechanical power actually output by the servo mechanism. The proportionality coefficient is determined by the electromagnetic torque coefficient of the servo mechanism itself and is a constant for a given servo mechanism. At the same time, considering the sign difference of the product generated by forward and reverse motion, in order to eliminate the interference of the angular velocity direction, this embodiment of the invention uses the absolute value of the product as the energy consumption value, which can accurately reflect the energy loss level of the servo mechanism at the numerical level.
[0040] In one specific implementation of this invention, for each sampling cycle, the absolute value of the product between the yaw rate and the drive current is calculated as the energy consumption value of each sampling cycle; all historical sampling cycles with the same roll direction identifier and servo direction identifier are extracted, i.e., both the roll direction identifier and the servo direction identifier are 1 or both are -1; the energy consumption values of all invalid energy consumption cycles during the current navigation process are accumulated to obtain the invalid energy consumption of each sampling cycle.
[0041] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the proportion of ineffective energy loss includes: While the absolute value of invalid energy consumption can reflect the total amount of energy dissipated by erroneous actions over a period of time, this value itself is affected by various factors such as the output power level of the servo mechanism and the severity of sea conditions. There may be significant differences between different ships, different voyages, and even different time periods within the same voyage. Therefore, this embodiment of the invention normalizes the invalid energy consumption of each sampling cycle to determine the basic energy loss ratio of each sampling cycle, so that the evaluation index has a consistent physical meaning under different operating conditions.
[0042] Specifically, since the total operating energy consumption during the current voyage reflects the overall workload of the servo mechanism within the historical time window, normalizing the invalid energy consumption using this as a reference benchmark can eliminate the influence of external factors such as the output power level of the servo mechanism and the severity of sea conditions on the evaluation results. This ensures that the basic energy loss ratio has a consistent physical meaning under different operating conditions. Therefore, in this embodiment of the invention, for each sampling cycle, the energy consumption values of all historical sampling cycles are accumulated to obtain the total operating energy consumption for each sampling cycle. The proportion of the invalid energy consumption in the total operating energy consumption is used as the basic energy loss ratio for each sampling cycle.
[0043] In one specific implementation of this invention, for each sampling cycle in the current navigation process, the energy consumption values of all previous historical sampling cycles are accumulated to obtain the total operating energy consumption of each sampling cycle; the invalid energy consumption is used as the numerator, the total operating energy consumption is used as the denominator, and the ratio of the two is used as the basic energy loss ratio of each sampling cycle.
[0044] It should be noted that if the total operating energy consumption is 0, it means that the servo mechanism of the ship does not output any mechanical work during the current navigation process, that is, the system is in a completely silent state, so this embodiment of the invention will not analyze it.
[0045] Furthermore, considering that the basic energy loss ratio only reflects the proportion of ineffective energy consumption from an energy perspective, but cannot reflect the directional tracking quality of the servo action, for example, if the servo mechanism can still effectively follow the main wave for most of the time, and only experiences directional misalignment due to the impact of broken waves on individual beats, the basic energy loss ratio may be high, but the effective matching score is also high, indicating that the servo mechanism still has effective countermeasure capability. The former contributes positively to the degree of system degradation, while the latter contributes negatively to the degree of system degradation. If gain attenuation is triggered solely by the basic energy loss ratio, it will weaken the suppression effect on the low-frequency main wave. Therefore, this embodiment of the invention determines the energy ineffective loss ratio for each sampling beat based on the basic energy loss ratio and the effective matching score. The energy ineffective loss ratio is positively correlated with the basic energy loss ratio and negatively correlated with the effective matching score.
[0046] In one specific implementation of this invention, a mismatch correction factor is obtained by subtracting the effective matching score from a preset base value (set to 1 in this embodiment). This results in both the mismatch correction factor and the base energy loss ratio being positively correlated with the degree of system deterioration. That is, the more deteriorated the system, the larger the mismatch correction factor and the higher the base energy loss ratio. Finally, the correlation between the two and the energy ineffective loss ratio is analyzed using a product. The product between the mismatch correction factor and the base energy loss ratio is used as the energy ineffective loss ratio for each sampling cycle.
[0047] Preferably, in some possible implementations of the embodiments of the present invention, the process of actively suppressing each sampling cycle based on the energy wastage ratio includes: When a ship encounters high-frequency breaking waves while sailing at sea, the servo mechanism continuously outputs a reverse torque to suppress roll. If the proportion of ineffective energy loss is high, it indicates that the servo mechanism has consumed a large amount of energy in erroneous movements over a period of time. Continuing to output at full power at this point will not only fail to effectively reduce roll but will also exacerbate mechanical wear and energy consumption. Therefore, in this embodiment of the invention, when the proportion of ineffective energy loss exceeds a preset upper limit threshold, a gain coefficient is determined based on the proportion of ineffective energy loss. The gain coefficient is used to attenuate the amplitude of the controller's output command, causing the servo mechanism to actively reduce the amplitude of its movements and reduce the number of ineffective commutations, thereby protecting mechanical components and preventing further energy waste. In particular, considering that the proportion of ineffective energy loss reflects the current overall deterioration of the servo mechanism, the larger the value, the deeper the system is trapped in an ineffective commutation state, and the more urgent the need for gain attenuation; therefore, this embodiment of the invention uses a preset base value for negative correlation mapping, that is, the gain coefficient is the difference between the preset base value and the proportion of ineffective energy loss.
[0048] In one specific implementation of this invention, since the energy wastage ratio is a dimensionless value between [0,1], this embodiment sets a preset base value to 1. The energy wastage ratio is subtracted from the preset base value to obtain the gain coefficient for each sampling cycle. A multiplication method is used to achieve amplitude attenuation. Specifically, the gain coefficient is multiplied by the controller output command of the current sampling cycle, and the result is used as the actual control command for the servo actuator. This processing method allows the servo mechanism to still output torque in the original opposing direction, but the amplitude of the output torque is proportionally reduced. This retains the basic anti-roll capability while preventing the servo mechanism from undergoing violent commutation due to full-power pursuit of high-frequency waves.
[0049] Furthermore, the preferred range of the preset upper limit threshold is [0.6, 0.8], and in this embodiment of the invention, it is set to 0.7; the preferred range of the preset lower limit threshold is [0.2, 0.4], and in this embodiment of the invention, it is set to 0.3; both can be adjusted within the corresponding preferred range according to the specific implementation scenario. For example, when an increase in roll amplitude or deterioration of sea conditions is detected, the preset upper limit threshold can be appropriately lowered to allow the system to enter the protection mode earlier; when a decrease in roll amplitude or calmer sea conditions is detected, the preset lower limit threshold can be appropriately increased to allow the system to recover to the normal combat mode more quickly.
[0050] If the current energy utilization efficiency of the servo mechanism is acceptable and the proportion of invalid actions is within an acceptable range, attenuating the gain would actually weaken the servo mechanism's ability to effectively combat low-frequency main waves. Therefore, in this embodiment of the invention, when the proportion of invalid energy loss is less than a preset lower threshold, the controller outputs its commands at full power.
[0051] In one specific implementation of this invention, the gain coefficient serves as a multiplier factor for the controller's output command, and its value directly determines the retention ratio of the actual output torque of the servo mechanism relative to the theoretically calculated torque. When the gain coefficient is 1, the theoretical feedback torque calculated by the controller is transmitted to the servo actuator as is, allowing the servo mechanism to output at full power, thereby maximizing the countermeasure against the low-frequency main wave. Therefore, the gain coefficient is 1 during the process of outputting the controller's output command at full power.
[0052] When the servo mechanism is in the transition range between the above two states, if the operating mode is switched directly according to the current ratio, the system will frequently switch between full power output and gain attenuation when the energy wastage ratio fluctuates slightly around the threshold, causing unnecessary mechanical oscillations and control jitter in the servo mechanism. Therefore, in this embodiment of the invention, when the energy wastage ratio is between a preset lower threshold and a preset upper threshold, the operating mode of the previous sampling cycle is maintained unchanged.
Claims
1. A control method for actively suppressing ship roll, characterized in that, The method includes: At each sampling cycle, the roll angle, roll rate, feedback deflection angle, deflection rate, and drive current of the target ship and the servo mechanism are obtained. Based on the sign consistency between the roll angle and the roll angular velocity, and between the roll angle and the feedback deflection angle, a local matching score is calculated to determine the effective matching score for each sampling cycle; based on the deflection velocity and the drive current, the energy consumption value for each sampling cycle is determined; and the energy consumption values of all historical sampling cycles are summed up by invalid energy consumption to determine the invalid energy consumption for each sampling cycle. Based on the effective matching score, the invalid energy consumption is corrected to determine the invalid energy loss ratio for each sampling cycle; and active suppression is performed on each sampling cycle based on the invalid energy loss ratio.
2. The control method for active suppression of ship roll according to claim 1, characterized in that, The method for obtaining the effective matching score includes: Obtain the sign function values of the roll angle, roll angular velocity, and feedback deflection angle respectively; determine the roll direction identifier for each sampling cycle based on the similarity of the sign function values between the roll angle and the roll angular velocity; determine the servo direction identifier for each sampling cycle based on the similarity of the sign function values between the roll angle and the feedback deflection angle. Extract the roll direction identifier and servo direction identifier of each neighboring sampling beat within a preset time window for each sampling beat; for each neighboring sampling beat, if the roll direction identifier is opposite to the servo direction identifier, assign its single-beat matching score a preset positive value; if the roll direction identifier is the same as the servo direction identifier, assign its single-beat matching score a preset negative value. The single-beat matching scores of all neighboring sampling beats within the preset time window of each sampling beat are accumulated and normalized to obtain the effective matching score of each sampling beat.
3. The control method for active suppression of ship roll according to claim 2, characterized in that, The method for obtaining the ineffective energy consumption includes: For each sampling cycle, extract all historical sampling cycles with the same roll direction identifier and servo direction identifier as invalid energy consumption cycles; sum up the energy consumption values of all invalid energy consumption cycles to obtain the invalid energy consumption of each sampling cycle.
4. The control method for active suppression of ship roll according to claim 1, characterized in that, The method for obtaining the energy ineffective loss ratio includes: For each sampling cycle, the invalid energy consumption is normalized and mapped to determine the basic energy loss ratio for each sampling cycle. Based on the basic energy loss ratio and the effective matching score, the invalid energy loss ratio for each sampling cycle is determined. The invalid energy loss ratio is positively correlated with the basic energy loss ratio and negatively correlated with the effective matching score.
5. The control method for active suppression of ship roll according to claim 4, characterized in that, The method for obtaining the basic energy loss ratio includes: For each sampling cycle, the energy consumption values of all historical sampling cycles are summed to obtain the total operating energy consumption of each sampling cycle; the proportion of the invalid energy consumption in the total operating energy consumption is used as the basic energy loss ratio for each sampling cycle.
6. The control method for active suppression of ship roll according to claim 1, characterized in that, The process of actively suppressing each sampling cycle based on the energy ineffective loss ratio includes: When the energy wastage ratio is greater than a preset upper limit threshold, a gain coefficient is determined based on the energy wastage ratio; the gain coefficient is then used to attenuate the amplitude of the controller's output command. When the energy ineffective loss ratio is less than the preset lower threshold, the controller outputs the command at full power. When the energy ineffective loss ratio is between a preset lower threshold and a preset upper threshold, the operating mode of the previous sampling cycle remains unchanged.
7. The control method for active suppression of ship roll according to claim 6, characterized in that, The gain coefficient is the difference between a preset base value and the energy ineffective loss ratio.
8. The control method for active suppression of ship roll according to claim 6, characterized in that, During the process of outputting the full-power output controller's output command, the gain coefficient is 1.
9. The control method for active suppression of ship roll according to claim 1, characterized in that, The energy consumption value is the absolute value of the product between the yaw rate and the drive current.
10. The control method for active suppression of ship roll according to claim 2, characterized in that, The method for obtaining the roll direction indicator is as follows: If the sign function value of the roll angle is the same as that of the roll angular velocity, then the roll direction identifier of each sampling beat is assigned a preset positive identifier; if the sign function value of the roll angle is different from that of the roll angular velocity, then the roll direction identifier of each sampling beat is assigned a preset negative identifier.
