A method and system for hydrofoil angle blending control of a hydrofoil boat

Abstract

This invention relates to the field of hydrofoil motion control technology, and provides a hydrofoil angle fusion control method and system, comprising: real-time acquisition of the hydrofoil angle and calculation of the error and error rate of change with the target angle; parallel input of the error into a standard PID, an integral separation PID, and a derivative-first PID controller to obtain three control quantities, wherein the integral separation threshold is adaptively determined based on the error amplitude and rate of change, and the derivative term acts on the measured angle rate of change; using the absolute value of the error and the rate of change as dual inputs, fuzzy inference continuously outputs fusion weights to normalize and weight the three control quantities; then, the angular velocity is obtained to adaptively calculate the damping coefficient, and damping compensation is superimposed to obtain the final control quantity to drive the actuator, forming a closed loop. This invention coordinates three control strategies and enhances damping, taking into account response speed, resistance to integral saturation, smoothness, and dynamic stability, thereby improving the hydrofoil angle control performance.

Description

Technical Field

[0001] This invention relates to the field of hydrofoil motion control technology, specifically a hydrofoil angle fusion control method and system. Background Technology

[0002] Hydrofoils rely on underwater wing plates to generate lift, raising the hull above the water surface to achieve high-speed navigation. The angle of attack of the hydrofoil directly determines the hull attitude, navigation stability, and navigation efficiency. In order to maintain the ideal attitude of the hull, the hydrofoil angle control system needs to quickly and accurately track the target angle command and have good robustness to external wave disturbances and changes in its own parameters.

[0003] Currently, the most commonly used control strategy in hydrofoil angle control engineering is proportional-integral-derivative (PID) control and its improved forms, such as integral-separated PID control and derivative-leading PID control. However, when applying the above control strategies to hydrofoil angle control, the following problems exist: I. It is difficult to simultaneously address both integral saturation resistance and steady-state error elimination. Existing integral separation PID controllers often use fixed thresholds or linearly adjust only the error amplitude as a single variable, failing to consider the motion trend reflected by the rate of error change, and thus unable to distinguish between two operating conditions: one where a large deviation has stabilized and the other where a large deviation is still changing rapidly.

[0004] Second, differential impact occurs when the target angle changes abruptly. Standard PID calculates the derivative of the error; when the target changes abruptly, the derivative term is amplified dramatically, impacting actuators such as hydraulic servo valves or electric actuators, shortening their lifespan and causing vibration.

[0005] Third, a single control strategy is difficult to adapt to all operating conditions. Existing improved strategies mostly have fixed thresholds or are independent of each other and lack coordination; even if weighted scheduling is introduced, it is mostly a three-stage discrete switching based on error levels. Sudden changes in weight at the boundary cause jitter, and it only relies on single error information without taking into account the trend of error change.

[0006] Fourth, active damping lacks self-adaptation. Second-order hydrofoil systems have low natural damping, and the soft damping provided by the PID differential term is constrained by error signals and is easily affected by noise. Existing active damping systems mostly use fixed damping coefficients and rely on manual tuning, making it difficult to adapt to real-time conditions. Stability and response sensitivity are difficult to balance under different operating conditions. Summary of the Invention

[0007] Existing hydrofoil angle control technologies suffer from contradictions in response speed, anti-integral saturation capability, smoothness, adaptability, and dynamic stability. There is a lack of a hydrofoil angle control method that can synergistically integrate multiple control strategies, adaptively adjust control parameters based on the real-time state of the system, and enhance system damping.

[0008] To at least partially solve the above problems, this invention proposes a hydrofoil angle fusion control method and system for hydrofoils.

[0009] In a first aspect, the present invention proposes a hydrofoil angle blending control method for hydrofoil boats, comprising: The current angle of the hydrofoil is collected in real time as the measurement angle. The difference between the target angle and the measurement angle is used to obtain the error, and the rate of change of the error is obtained. The error is input in parallel to a standard PID controller, an integral separation PID controller, and a derivative-first PID controller to obtain the first control quantity, the second control quantity, and the third control quantity, respectively. With the absolute value and rate of change of the error as dual inputs, fuzzy inference continuously outputs the fusion weights corresponding to the three controllers respectively; The first control quantity, the second control quantity, and the third control quantity are normalized and weighted averaged using the fusion weights to obtain the fusion control quantity. The angular velocity of the hydrofoil is obtained, and the damping coefficient is adaptively calculated based on the angular velocity, error and its rate of change. The damping compensation term proportional to the angular velocity is superimposed on the fusion control quantity and limited to obtain the final control quantity. The final control quantity is output to the hydrofoil actuator to drive the hydrofoil to adjust its actual angle, and the adjusted actual angle is used as the acquisition object for the next cycle of angle measurement, forming a closed loop.

[0010] In a preferred embodiment, the integral separation threshold of the integral separation PID controller is a two-dimensional adaptive threshold. The two-dimensional adaptive threshold includes an amplitude term that increases with the magnitude of the error, and a trend term that changes with the rate of change of the error and decreases as the magnitude of the error increases.

[0011] In a preferred embodiment, the trend term is the product of the error rate of change influence coefficient, the absolute value of the error rate of change, and an exponential factor, wherein the exponential factor is a natural exponent with the negative of the product of the attenuation factor and the absolute value of the error as the exponent; the two-dimensional adaptive threshold, after being limited, is used as the integral separation threshold actually used at the current sampling time.

[0012] As a preferred embodiment, the continuous output of fusion weights via fuzzy inference includes: Multiple fuzzy sets are set for the absolute value of the error and the rate of change of the error, respectively. The membership degree of the current input to each fuzzy set is calculated based on the membership function of each fuzzy set. A fuzzy rule library is established to cover the combinations of fuzzy sets that cover the absolute value of the error and the rate of change of the error. Each fuzzy rule is based on a set of recommended weight coefficients corresponding to the three controllers. The product of the membership degree of the absolute value of the error and the membership degree of the rate of change of the error in each fuzzy rule is used as the activation degree of the fuzzy rule. The recommended weight coefficients corresponding to the same controller in each fuzzy rule are weighted and summed according to their activation degrees and normalized by the sum of all activation degrees to obtain the fusion weights corresponding to the three controllers.

[0013] In a preferred embodiment, among the multiple fuzzy sets corresponding to the absolute value of the error, the fuzzy sets representing the approach to the target and the transition stage adopt a Gaussian membership function, and the fuzzy sets representing the distance from the target adopt a monotonically increasing membership function; each fuzzy set corresponding to the rate of change of the error adopts a Gaussian membership function; the membership degree of the monotonically increasing membership function increases monotonically with the increase of the absolute value of the error and approaches its upper limit.

[0014] In a preferred embodiment, the adaptive calculation of the damping coefficient includes: The theoretical damping coefficient is determined by the difference between the target damping ratio determined by the optimal damping ratio of the second-order system and the inherent damping ratio of the hydrofoil system, combined with the undamped natural frequency of the hydrofoil system and the maximum allowable angular velocity of the hydrofoil. The basic damping coefficient is then corrected in real time based on the angular velocity, the error, and the rate of change of the error to determine the real-time adaptive damping coefficient. The larger of the theoretical damping coefficient and the real-time adaptive damping coefficient is taken as the adaptive damping coefficient.

[0015] In a preferred embodiment, the real-time correction of the basic damping coefficient includes an angular velocity term and an error term. The angular velocity term enhances damping as the angular velocity increases, while the error term enhances damping with the magnitude of the error and weakens the enhancement effect by an exponential factor with the rate of change of the error.

[0016] In a preferred embodiment, the damping compensation term is determined based on the angular velocity of the hydrofoil, independent of the standard PID controller, integral-separated PID controller, and derivative-first PID controller based on the error. The weighted fusion and the superposition of the damping compensation constitute two separate stages. The first stage weights and fuses the first, second, and third control quantities according to the fusion weights to obtain the fused control quantity. The second stage superimposes the damping compensation term onto the fused control quantity based on the angular velocity. The damping compensation term is superimposed on the damping provided by the derivative terms of the standard PID controller and the derivative-first PID controller.

[0017] In a preferred embodiment, the third control quantity of the derivative-first PID controller is obtained by summing and limiting the proportional term, integral term, and derivative term. The proportional term is determined based on the error and proportional gain, the integral term is determined based on the integral accumulation of the error and integral gain, and the derivative term is determined based on the rate of change of the measured angle and derivative gain, with its negative value participating in the summation. The rate of change of the measured angle is obtained by differentiating the measured angle and passing it through a first-order low-pass filter.

[0018] In a second aspect, the present invention provides a hydrofoil angle blending control system for a hydrofoil boat, comprising: The error calculation module is used to acquire the measurement angle of the hydrofoil in real time, calculate the error by subtracting the target angle from the measurement angle, obtain the rate of change of the error, output the error to the parallel control module and the fuzzy weight scheduling module, and output the measurement angle to the parallel control module. The parallel control module includes a standard PID controller, an integral-separated PID controller, and a derivative-first PID controller connected in parallel. It is used to perform parallel calculations on the error to obtain a first control quantity, a second control quantity, and a third control quantity, and output them to the weighted fusion module. The fuzzy weight scheduling module is used to take the absolute value and rate of change of the error as dual inputs, and continuously output the fusion weights corresponding to the three controllers through fuzzy inference and output them to the weighted fusion module. The weighted fusion module is used to perform a normalized weighted average of the first control quantity, the second control quantity, and the third control quantity using the fusion weights, to obtain a fused control quantity and output it to the adaptive active damping module. An adaptive active damping module is used to acquire the angular velocity of the hydrofoil, adaptively calculate the damping coefficient based on the angular velocity, the error and its rate of change, and superimpose a damping compensation term proportional to the angular velocity onto the fused control quantity and limit it to obtain the final control quantity. The final control quantity is used to drive the hydrofoil to adjust its actual angle, and the adjusted actual angle serves as the source of the measured angle in the next cycle, forming a closed loop.

[0019] Compared with the prior art, the present invention has the following advantages: 1. This invention employs a two-dimensional dynamic integral separation threshold, enabling the integral action to adaptively engage and disengage based on both error amplitude and error rate of change. This suppresses integral saturation when there is a large and rapidly changing deviation, and eliminates steady-state error when there is a small and stabilizing deviation, thus improving the system's resistance to integral saturation and its steady-state accuracy. Traditional integral separation PID controllers use fixed thresholds, making it difficult to accommodate different operating conditions; even simple linear adaptive methods do not account for the impact of the error rate of change. This invention introduces an error rate of change correction term, allowing the threshold to distinguish between two states: large deviations that have stabilized and large deviations that are still rapidly changing, achieving more refined integral management.

[0020] 2. This invention employs adaptive active damping control, calculating the damping coefficient using the hydrofoil angular velocity and real-time system status to suppress overshoot and residual oscillations, improve system dynamic stability, and eliminate the need for manual tuning of damping parameters. Conventional PID controllers provide soft damping through their derivative terms, and their fixed damping coefficients are ill-suited to varying operating conditions. This invention combines the theoretically optimal damping ratio with real-time status to automatically adjust the damping strength, ensuring basic stability while responding to sudden disturbances, thus achieving adaptive damping control.

[0021] 3. This invention employs dual-input continuous fuzzy weighted scheduling, smoothly adjusting the fused weights of the three controllers based on the error magnitude and error change rate. This allows for continuous transition of control strategies under different operating conditions, improving overall control quality. Existing three-stage discrete switching based on error grading is prone to weight abrupt changes at the grading boundaries and relies solely on single error information. This invention uses a Gaussian membership function and multiple fuzzy rules, which are normalized and defuzzified to ensure that weights change continuously with the system state, eliminating switching jitter. Furthermore, the introduction of the error change rate makes weight scheduling more consistent with the actual trend of error change.

[0022] 4. This invention employs derivative-first PID control, which applies the derivative term to the measured angle rather than the error, eliminating the derivative shock when the target angle changes abruptly, keeping the control command smooth, and avoiding instantaneous impact on the actuator. It is especially suitable for hydrofoils that frequently adjust the target angle.

[0023] 5. This invention adopts a two-stage architecture of weighted fusion followed by adaptive active damping, decoupling the outputs of the three PID controllers from the adaptive damping term in two stages. The fusion layer determines the emphasis of the control strategy based on the fusion weights, and the damping layer independently enhances the system damping based on the angular velocity. The two can be tuned independently, avoiding parameter coupling between the fusion strategy and the damping tuning. Without affecting the control strategy determined by the fusion, it further suppresses the inherent underdamped oscillations of the second-order hydrofoil system. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation on the scope of this application.

[0025] Figure 1 This is an exemplary flowchart of the hydrofoil angle fusion control method for hydrofoil boats provided in the embodiments of the present invention; Figure 2 This is a comparison chart of the absolute error values ​​of different control methods in the embodiments of the present invention; Figure 3 This is a comparison diagram of the control signals of various control methods in the embodiments of the present invention; Figure 4 This is a comparison chart of the integral separation effects of dynamic threshold and fixed threshold in an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the variation range of the adaptive damping coefficient in an embodiment of the present invention; Figure 6 This is a schematic diagram illustrating the assignment and change frequency of fuzzy weights in an embodiment of the present invention; Figure 7 This is a graph showing the change in the absolute value of the error of the fusion control in an embodiment of the present invention; Figure 8 This is a comparison chart of the proportion of integral saturation time in embodiments of the present invention; Figure 9 This is a comparison chart of disturbance recovery times in embodiments of the present invention; Figure 10 This is a comparison chart of peak times in an embodiment of the present invention; Figure 11 This is a comparison chart of adjustment time in an embodiment of the present invention; Figure 12 This is a comparison chart of the robustness of fusion control under different operating conditions in the embodiments of the present invention; Figure 13 This is a comparison diagram of the effects of having and not having active damping in the embodiments of the present invention; Figure 14 This is a control signal diagram of the fusion controller in an embodiment of the present invention; Figure 15 This is a tracking performance diagram of the fusion control in an embodiment of the present invention; Figure 16 This is a comparison diagram of three separate PID controls in an embodiment of the present invention; Figure 17 This is a comparison chart showing whether or not there is a control effect in the embodiments of the present invention. Detailed Implementation

[0026] To make the technical means, creative features, objectives, and effects of this invention easier to understand, the invention is further described below with reference to specific embodiments. However, the following embodiments are merely preferred embodiments of this invention and not all of them. Other embodiments obtained by those skilled in the art based on the embodiments described herein without creative effort are all within the protection scope of this invention.

[0027] A specific embodiment of the present invention discloses a method for hydrofoil angle fusion control of a hydrofoil boat, the overall process of which is as follows: Figure 1 As shown, the method includes: S1. The current angle of the hydrofoil is collected in real time as the measurement angle. The difference between the target angle and the measurement angle is used to obtain the error, and the rate of change of the error is obtained.

[0028] The current angle of the hydrofoil is collected in real time as the measurement angle. Specifically, an angle sensor located at the hydrofoil's pivot point collects the angle of attack of the hydrofoil relative to the water flow direction at a fixed sampling period to obtain the measurement angle at the current sampling moment, in degrees. In one embodiment, the sampling period is 0.01s. Then, the target angle issued by the driver or autopilot is read. The target angle represents the attitude the hydrofoil should reach under the current operating conditions. The difference between the target angle and the measurement angle is calculated using the following formula to obtain the error e at the current sampling moment: ; Where n represents the current sampling time, Let n be the target angle at sampling time n. Let e ​​be the measured angle at sampling time n. If e>0, it means that the measured angle is lower than the target angle and the hydrofoil has not yet reached its position. If e<0, it means that the measured angle is higher than the target angle and the hydrofoil has already reached its position.

[0029] Further, the rate of change of the error is obtained by subtracting the error at the current sampling time from the error at the previous sampling time using the following formula to obtain the original rate of change of error. : ; in This is the error from the previous sampling time. The sampling period.

[0030] The original error change rate is subjected to a first-order low-pass filter to remove the high-frequency components generated by the differential amplification of sensor noise, thus obtaining the filtered error change rate. Specifically, the original error change rate at the current sampling time and the filtered result at the previous sampling time are weighted and summed according to the filter coefficients to obtain the filtered error change rate at the current sampling time. The weight of the original error change rate at the current sampling time is the filter coefficient, and the weight of the filtered result at the previous sampling time is the difference between 1 and the filter coefficient. A larger filter coefficient value indicates a weaker filtering effect and a faster response to the original error change rate; a smaller value indicates a stronger filtering effect and stronger suppression of high-frequency components. The filter coefficient is determined by the cutoff frequency of the low-pass filter and the sampling period. Specifically, the filter coefficient is determined based on the cutoff frequency and sampling period of the low-pass filter, according to the discretization relationship of the first-order low-pass filter. That is, the filter coefficient is equal to 1 minus the negative exponent of the natural constant, where the exponent is the opposite of the product of 2π times the cutoff frequency and the sampling period. The higher the cutoff frequency or the larger the sampling period, the closer the filter coefficient is to 1 and the weaker the filtering effect; conversely, the closer it is to 0 and the stronger the filtering effect. In one embodiment, the cutoff frequency of the low-pass filter is 20Hz, and the corresponding filter coefficient is approximately 0.56.

[0031] Thus, the error and its rate of change are used in subsequent parallel PID control, fuzzy weight scheduling, and adaptive active damping correction. The measured angle is also used to calculate the rate of change of the measured angle in derivative-first PID control.

[0032] S2. Input the error into the standard PID controller, the integral separation PID controller and the derivative-first PID controller in parallel to obtain the first control quantity, the second control quantity and the third control quantity respectively.

[0033] The error is input in parallel to the standard PID controller, the integral separation PID controller, and the derivative-first PID controller, and the first control quantity is calculated respectively within the same sampling period. Second control quantity With the third control quantity All units are °; the three control laws operate in parallel without interfering with each other. Furthermore, the standard PID controller performs proportional, integral, and derivative operations on the error sequentially, sums the results, and limits the amplitude to obtain the first control quantity; firstly, it takes the proportional term of the current error according to the proportional gain. The error is then summed using a trapezoidal integral to obtain the accumulated integral. : ; And take the integral term Then, the rate of change of the filtered error is taken. Constitute differential terms At this point, the three terms are summed and the amplitude is limited to obtain the first control quantity. : ; in, For proportional gain, For integral gain, For differential gain, This indicates a range-limiting operation that restricts the values ​​within parentheses to a preset range of actuator instructions. In one implementation, the... The value is 15.0, the aforementioned The value is 10.0, the The value is 2.0, and the value of the limiting range is [-30°, 30°].

[0034] Specifically, the standard PID controller generates a comprehensive control quantity through three parallel channels: proportional, integral, and derivative. The proportional term represents the rapid response to the current deviation and is the most intuitive part, directly generating a proportional control force based on the magnitude of the current deviation. In hydrofoil angle control, if the current hydrofoil angle is 5° lower than the target angle, the proportional term will output a positive control command, causing the hydrofoil to rotate upwards; the larger the deviation, the stronger the rotation command. The proportional gain... Controlling this intensity, The larger the value, the faster the response; however, excessively large values ​​can easily cause overshoot or even oscillation. Therefore, the default value is... The value is 15, which enables rapid target tracking while maintaining a certain level of stability.

[0035] The integral term represents the elimination of steady-state error. Proportional control alone often cannot completely eliminate deviation because the proportional term becomes very small when the hydrofoil approaches the target, potentially being offset by friction or other resistance, leading to a permanent steady-state error, such as consistently being 0.5° below the target. The integral term addresses this problem by accumulating historical deviations. Traditional rectangular integrals are simple but generally inaccurate. This invention employs trapezoidal integrals, using the average deviation between the current and previous moments to calculate the integration step size. This is equivalent to approximating the rectangular area with the area of ​​a trapezoid, resulting in smaller errors, especially with higher integration accuracy when the deviation changes rapidly; Integral gain. The cumulative deviation is determined by the weight of its contribution to the control quantity. If the hydrofoil remains below the target position for an extended period, the integral term will gradually increase until the hydrofoil is moved to the target position. However, the integral term may also lead to saturation.

[0036] The differential term represents the trend of the prediction deviation and reflects the rate of change of the deviation, playing a role in advance adjustment. If the hydrofoil is rapidly approaching the target, the differential term will output a negative command to stop in advance and prevent overshoot; if the hydrofoil is moving away from the target, the differential term will output a positive command to strengthen the action. There is high-frequency noise in the actual sensor signal, which becomes very large after differential processing, causing the control quantity to jitter violently. To address this, a differential low-pass filter is provided to perform a first-order inertial filter on the original differential. The filter coefficient is calculated from the cutoff frequency and sampling period, which significantly reduces noise above 20Hz, making the differential action smooth and reliable.

[0037] Furthermore, the calculation method for the proportional and derivative terms of the integral-separated PID controller is the same as that of the standard PID controller. The difference lies in that the accumulation of the integral term is constrained by the integral separation threshold, and only when the absolute value of the error is less than the integral separation threshold... At that time, the accumulated amount of its independently maintained integrals If the integral is accumulated using the trapezoidal integral method, then the accumulated integral remains unchanged. ; Then the proportional term and the integral term Summing the differential term and limiting the amplitude yields the second control quantity. .

[0038] Furthermore, the integral separation threshold Instead of taking a fixed constant, and not merely varying linearly with the magnitude of the error, a two-dimensional adaptive threshold is simultaneously considered, taking into account both the magnitude of the error and the rate of change of the error, and is calculated in real time using the following formula: ; in, Based on the threshold, This is the coefficient affecting the magnitude of the error. The factor representing the influence of the rate of change of error. As an attenuation factor, in one embodiment, the The value is 5.0, the The value is 0.10, the The value is 0.40, the The value is 0.10, and the calculation result is limited to a preset range. , Within this range, the actual integral separation threshold used at the current sampling time is obtained; the first term... The integral separation threshold is increased as the magnitude of the error increases, raising the threshold for large errors and delaying the integration intervention; the second item The rate of change of the error is used for trend correction, and its exponential factor is... The effect of the trend term decreases as the magnitude of the error increases. For small and medium errors, the threshold is adjusted differently based on whether the hydrofoil is rapidly approaching or rapidly deviating from the target. For large errors, the influence of the rate of change is weakened to prevent the threshold from being excessively amplified. With the help of the two-dimensional adaptive threshold, the integral separation PID controller distinguishes between two operating conditions: large error that has stabilized and large error that is still changing rapidly. This achieves more precise integral switching than linear adjustment of a single error.

[0039] Furthermore, the calculation method for the proportional and integral terms of the derivative-first PID controller is the same as that of the standard PID controller, that is, the proportional term is taken as... Integral terms ,in The integral accumulation is calculated cycle by cycle according to the same trapezoidal integral rule as the standard PID controller; the difference is that the derivative term does not act on the rate of change of the error, but on the rate of change of the measured angle.

[0040] The rate of change of the measured angle is obtained by differentiating the measured angle. : ; Then, a first-order low-pass filter is applied to it to obtain the filtered rate of change of the measured angle. And its negative value forms the differential term. The three terms are then summed and limited to obtain the third control quantity. : ; The measured angle is constrained by the hydrofoil's own inertia, and remains continuous without abrupt change when the target angle undergoes a step change; the differential term accordingly maintains a finite value.

[0041] At this point, the first, second, and third control quantities have been calculated in parallel within the same sampling period and are output together to the subsequent weighted fusion step; the absolute value of the error and the rate of change of the error are also output to the continuous fuzzy weight scheduling step to determine the weights of the above three control quantities in the weighted fusion.

[0042] Specifically, all three share the same input and use the same gain, differing only in the switching conditions of the integral term and the object of action of the derivative term. Therefore, they are essentially three variants of the same set of proportional, integral, and derivative parameters. The first control quantity provides a comparison benchmark for the other two control quantities. The second and third control quantities are obtained by specifically modifying the integral and derivative terms based on the first control quantity. The standard PID controller accumulates the integral term while keeping the error constant and differentiates the rate of change of the error. Its advantages are that it responds quickly to changes in the target angle, can suppress overshoot in advance, and has comprehensive tracking performance. Its disadvantages are that when the error is large or the hydrofoil is subjected to strong disturbances, the integral term accumulates rapidly and saturates. Furthermore, when the target angle undergoes a step change, the rate of change of the error increases instantaneously, resulting in a differential shock.

[0043] The integral-separated PID controller uses the two-dimensional adaptive threshold constraint to switch the integral term. Its advantages are that, under large errors, it suspends integral accumulation, which is equivalent to degenerating into a control law containing only proportional and derivative actions, thereby suppressing integral saturation and accelerating the homing under large deviations; under small errors, it resumes integral accumulation and eliminates steady-state error; and by using the two-dimensional threshold that takes into account the magnitude and rate of change of the error, it distinguishes between two operating conditions: large error that has tended to stabilize and large error that is still changing rapidly, achieving more refined integral management than single error linear adjustment. Its disadvantage is that its derivative term still acts on the rate of change of the error, so there is also a derivative shock when the target angle steps.

[0044] The derivative-first PID controller changes the rate of change of the derivative term from the rate of change of the error to the rate of change of the measured angle. Its advantage is that the measured angle remains continuous due to the inertia of the hydrofoil itself and does not change abruptly with the step change of the target angle. Therefore, the derivative term is always a finite value, the control command is smooth, and instantaneous impact on the hydrofoil actuator is avoided. It is especially suitable for working conditions where the target angle is frequently adjusted. Its disadvantage is that, since the derivative term no longer directly reflects the change of the error, its predictability of the change of the target angle is slightly weaker than that of the standard PID controller.

[0045] The characteristics of the three control variables indicate that, under all operating conditions of the hydrofoil angle control, no single control variable is optimal at all times. When there is a large error and the target deviates rapidly, the anti-saturation characteristic of the second control variable is more advantageous. When the target angle jumps or tends to a steady state, the smoothing characteristic of the third control variable is more advantageous. When the target is approached rapidly with a moderate error, the fast response characteristic of the first control variable is more advantageous. Therefore, the subsequent continuous fuzzy weight scheduling step needs to allocate fusion weights to the three control variables in real time based on the absolute value of the error and the rate of change of the error, so that the fused control variable is close to the optimal value under each operating condition. This is the basis for the subsequent continuous fuzzy weight scheduling and weighted fusion steps.

[0046] S3. Using the absolute value and rate of change of the error as dual inputs, fuzzy inference continuously outputs the fusion weights corresponding to the three controllers respectively.

[0047] The absolute value and rate of change of the error are used as two inputs for fuzzy inference; the absolute value of the error represents the degree to which the hydrofoil deviates from the target angle, and the rate of change of the error represents the direction and speed at which the hydrofoil approaches or deviates from the target angle.

[0048] Three fuzzy sets—zero-small, medium, and large—are set for the absolute value of the error, and three fuzzy sets—negative-large, zero, and positive-large—are set for the rate of change of the error. Each fuzzy set provides a membership degree between 0 and 1 based on the current input. The three fuzzy sets for the rate of change of the error (negative-large, zero, and positive-large) and the two fuzzy sets for the absolute value of the error (zero-small and medium) all use Gaussian membership functions. The fuzzy set for the large absolute value of the error uses a monotonically increasing membership function. Specifically, the monotonically increasing membership function is a S-shaped function, where the membership degree is equal to 1 divided by the sum of the bases: the sum of the bases is 1 and the negative exponent of the natural constant; the exponent is the difference between the absolute value of the error and the center of the large fuzzy set divided by the width coefficient. The S-shaped function has a membership degree of 0.5 at the center, monotonically increasing and approaching 1 as the absolute value of the error increases, and monotonically decreasing and approaching 0 as the absolute value of the error decreases. The larger the width coefficient, the smoother the increase and the wider the coverage.

[0049] Specifically, the Gaussian membership function is symmetrical about the center of the fuzzy set, and its width coefficient represents the coverage range. When the input is equal to the center of the fuzzy set, the membership degree reaches its maximum value of 1. When the input deviates from the center, the membership degree decreases exponentially and smoothly, approaching 0, as the deviation increases. The larger the width coefficient, the slower the membership degree decreases with the deviation and the wider the input range it covers. The smaller the width coefficient, the faster the decrease and the narrower the coverage range. The monotonically increasing membership function is not symmetrical about the center. Its membership degree increases monotonically with the increase of the absolute value of the error and approaches 1. At the set center, the membership degree is 0.5, which is used to represent that the larger the absolute value of the error, the more completely it belongs to the large error state.

[0050] The centers of the two fuzzy sets representing the absolute value of the error (small and medium) are set sequentially from small to large, corresponding to the hydrofoil approaching the target and being in the transition phase, respectively. The center of the large fuzzy set corresponds to the large error state where the hydrofoil is moving away from the target. The centers of the three fuzzy sets representing the rate of change of the error (large negative, zero, and large positive) are set sequentially to negative, zero, and positive values, corresponding to the hydrofoil rapidly approaching the target, approximately uniform or steady state, and rapidly deviating from the target, respectively. In one embodiment, the centers of the two fuzzy sets representing the absolute value of the error (small and medium) are set to 0 and 6, respectively, and the width coefficients of the corresponding Gaussian membership functions are set to 1.5 and 2.5, respectively. The center of the monotonically increasing membership function of the large fuzzy set is set to 5, and the width coefficient is set to 4.0. The centers of the three fuzzy sets representing the rate of change of the error (large negative, zero, and large positive) are set to... The width coefficients of the Gaussian membership functions corresponding to 5, 0, and 5 are 3.0, 2.0, and 3.0, respectively. The center and width coefficients of the above fuzzy sets are design parameters tuned based on engineering experience to cover the dynamic range of typical hydrofoil angle control. In other embodiments, the center and width coefficients can be adjusted according to the actual dynamic characteristics of the hydrofoil to adapt to different response speed requirements.

[0051] Furthermore, a rule base consisting of nine fuzzy rules is established, which covers all combinations of the three fuzzy sets of the absolute value of the error and the three fuzzy sets of the rate of change of the error. Each fuzzy rule is based on the premise that the absolute value of the error belongs to a certain fuzzy set and the rate of change of the error belongs to a certain fuzzy set, and concludes with a set of recommended weight coefficients corresponding to the three controllers respectively. The recommended weight coefficients reflect the relative importance that the three controllers should be assigned under the working condition described by the premise.

[0052] The premise and conclusion configuration of the nine fuzzy rules are as follows: When the absolute value of the error is zero and the rate of change of the error is positive, the hydrofoil deviates rapidly from the target near the steady state, the derivative-first PID controller is enhanced, and the integral separation PID controller is moderately weakened, with recommended weight coefficients of (1.0, 0.8, 1.3); When the absolute value of the error is zero and the rate of change of the error is zero, the hydrofoil is near the steady state, and the derivative-first PID controller is used to suppress minor jitter, with recommended weight coefficients of (1.0, 1.0, 1.2); When the absolute value of the error is zero and the rate of change of the error is negative, the hydrofoil rapidly approaches the target, and the integral separation PID controller is introduced in advance to assist convergence, with recommended weight coefficients of (0.8, 1.4, 0.6); When the absolute value of the error is moderate and the rate of change of the error is positive, the standard PID controller dominates and quickly reverses the deviation trend, with recommended weight coefficients of (1.1, 1.0, 0.9); When the absolute value of the error is moderate... When the error rate of change is zero, all three components work in a balanced manner, with a slight increase in the integral-separated PID controller. The recommended weighting coefficients are (1.0, 1.1, 1.0). When the absolute value of the error is moderate and the error rate of change is negatively large, the integral-separated PID controller assists in rapid convergence. The recommended weighting coefficients are (0.7, 1.6, 0.5). When the absolute value of the error is large and the error rate of change is positively large, the standard PID controller is enhanced to quickly pull back, and the derivative-first PID controller is moderately weakened. The recommended weighting coefficients are (1.3, 1.2, 0.4). When the absolute value of the error is large and the error rate of change is zero, the integral-separated PID controller dominates, eliminating large steady-state errors and preventing integral saturation. The recommended weighting coefficients are (1.0, 1.5, 0.4). When the absolute value of the error is large and the error rate of change is negatively large, the integral-separated PID controller is fully enhanced to assist in homing. The recommended weighting coefficients are (0.5, 2.0, 0.3).

[0053] For each fuzzy rule, the product of the membership degree of the absolute value of the error in its premise and the membership degree of the rate of change of the error is taken as the activation degree of the fuzzy rule. The activation degree represents the degree to which the current input simultaneously satisfies the two premises of the fuzzy rule.

[0054] Furthermore, using the activation degree of each fuzzy rule as the weight, the recommended weight coefficients corresponding to the same controller in the conclusion of each fuzzy rule are weighted and summed, and then normalized by dividing by the sum of the activation degrees of all fuzzy rules to obtain the weight adjustment coefficients corresponding to the three controllers respectively. The weight adjustment coefficients change continuously with the continuous change of activation degree and there is no abrupt change point. When the sum of all activation degrees is zero, the weight adjustment coefficient takes the value corresponding to the preset base weight.

[0055] The weight adjustment coefficient can be calculated as follows: ; Where i is the controller number, i=1,2,3, corresponding to the standard PID controller, the integral-separated PID controller, and the derivative-first PID controller, respectively. The first weight, second weight, and third weight of the three controllers are respectively assigned to the current sampling time. For the corresponding controller's base weights, Let be the activation degree of the j-th fuzzy rule. For the recommended weight coefficient of the i-th controller in the conclusion of the j-th fuzzy rule, in one implementation, the base weights corresponding to the first weight, second weight, and third weight are... All values ​​are set to 1.0.

[0056] Thus, the first weight, second weight, and third weight corresponding to the three controllers at the current sampling time are obtained. The three weights are output to the subsequent weighted fusion step to perform normalized weighted averaging on the first control quantity, the second control quantity, and the third control quantity. Since the membership degree, activation degree, and weight adjustment coefficient all change continuously with the absolute value of the error and the rate of change of the error, the three weights also transition smoothly and continuously with the operating conditions, seamlessly switching control strategies between different error magnitudes and trends. This avoids the weight abrupt changes and control quantity jitter caused by discrete switching based on error levels at the level boundaries.

[0057] S4. The first, second, and third control quantities are normalized and weighted by the fusion weights to obtain the fusion control quantity.

[0058] Using the first weight, the second weight, and the third weight, the first control quantity, the second control quantity, and the third control quantity are normalized and weighted to obtain the fused control quantity at the current sampling time.

[0059] Specifically, the three weights are used as weighting coefficients for the first, second, and third control quantities, respectively. The three control quantities are summed according to their corresponding weights, and then normalized by dividing by the sum of the three weights to obtain the fused control quantity. ; in, , , These represent the outputs of the standard PID controller, the integral-separated PID controller, and the derivative-ahead PID controller, respectively, at the current sampling time. , , These are the weights assigned to the three control variables at the current sampling time.

[0060] Furthermore, the weighted average uses normalization processing instead of direct weighted summation. The division by the sum of the three weights ensures that the fused control quantity is always a convex combination of the three control quantities. Thus, regardless of how the three weights change, the amplitude of the fused control quantity is on the same order of magnitude as each control quantity and is not amplified as the overall weight increases. When a certain weight is zero, the corresponding control quantity does not participate in the fusion. When the three weights are equal, the fused control quantity degenerates into the arithmetic mean of the three control quantities.

[0061] Thus, the fused control quantity integrates the characteristics of the fast response of the standard PID controller, the anti-integral saturation of the integral separation PID controller, and the anti-derivative impact of the derivative-leading PID controller. Based on the three weights that change continuously with the operating conditions, the fused control quantity continuously emphasizes the appropriate control quantity under different error magnitudes and trends. The fused control quantity is output to the subsequent adaptive active damping correction step as the control quantity to be corrected before superimposed damping compensation.

[0062] S5. Obtain the angular velocity of the hydrofoil, adaptively calculate the damping coefficient based on the angular velocity, error and its rate of change, superimpose the damping compensation term proportional to the angular velocity onto the fusion control quantity and limit it to obtain the final control quantity.

[0063] First, the angular velocity of the hydrofoil is obtained. In one embodiment, the angular velocity is directly measured by an angular velocity sensor located at the hydrofoil's pivot. The adaptive damping coefficient is not a fixed constant, but is composed of the larger of the theoretical damping coefficient determined based on the optimal damping ratio of the second-order system and the real-time adaptive damping coefficient determined based on the real-time state. This allows the damping control to automatically adjust according to the operating conditions while meeting the minimum damping requirements, without the need for manual adjustment.

[0064] Subsequently, the theoretical damping coefficient is determined; based on the optimal damping ratio theory of second-order systems, the difference between the target damping ratio and the inherent damping ratio of the second-order dynamic characteristics of the hydrofoil is taken as the required supplementary damping ratio, and the theoretical damping coefficient is calculated accordingly. : ; in, The target damping ratio of the second-order system is... The inherent damping ratio of the hydrofoil system is given. The undamped natural frequency of the hydrofoil system is expressed in rad / s. The maximum permissible angular velocity of the hydrofoil, determined by mechanical limits; when ≤ When the theoretical damping coefficient is zero, in one embodiment, the target damping ratio is... The value is 0.707; the undamped natural frequency With the inherent damping ratio The system parameters characterizing the second-order dynamic characteristics of the hydrofoil are determined by parameter identification of the hydrofoil system or based on its design parameters; the maximum angular velocity... In one embodiment, the undamped natural frequency is determined by the mechanical limit of the hydrofoil actuator. The value is 10.0 rad / s, and the inherent damping ratio is... The value is 0.5, and the maximum angular velocity is... The value is 3.14 rad / s.

[0065] The theoretical damping coefficient is the equivalent damping coefficient required to supplement the closed-loop damping ratio of the second-order hydrofoil system relative to its inherent damping ratio, so that the closed-loop damping ratio reaches the target damping ratio. Its function is to provide a lower limit for the adaptive damping that does not change with the real-time state. In contrast, the real-time adaptive damping coefficient enhances the damping as needed based on the real-time state of the system. The larger of the two is taken so that the adaptive damping coefficient is not lower than the minimum value required to meet the optimal damping ratio under any operating condition, and is further enhanced under conditions of violent system motion or large deviation, thereby taking into account both basic stability and on-demand damping.

[0066] Subsequently, the real-time adaptive damping coefficient is determined; based on the absolute value of the angular velocity, the absolute value of the error, and the rate of change of the error, the basic damping coefficient is corrected in real time to obtain the real-time adaptive damping coefficient. : ; in, Based on the basic damping coefficient, Angular velocity, This is the angular velocity amplification factor. This is the error amplification factor. As the attenuation coefficient, in one embodiment, the... The value is 0.5, the The value is 0.10, the The value is 0.03, the The value is 0.20; where the first correction term... The real-time adaptive damping coefficient increases with the increase of the angular velocity, thereby enhancing damping to suppress overshoot and oscillation when the hydrofoil moves violently; second correction term. Damping is enhanced by the magnitude of the error, while the rate of change of the error is reduced by an exponential factor. This ensures that when the error is large and tends to stabilize, damping is moderately enhanced to prevent overshoot, and when the error changes rapidly, the contribution of the error term is reduced to avoid excessive damping and slowing down the response. The aforementioned basic damping coefficient, angular velocity amplification coefficient, error amplification coefficient, attenuation coefficient, and target damping ratio are all design parameters tuned based on engineering experience to adapt to the dynamic range of typical hydrofoil angle control. In other embodiments, the above parameters can be adjusted accordingly based on the actual dynamic characteristics of the hydrofoil to adapt to different stability and response speed requirements.

[0067] Therefore, the larger of the theoretical damping coefficient and the real-time adaptive damping coefficient is taken as the adaptive damping coefficient actually used at the current sampling time. : ; At this point, a damping compensation term is constructed by multiplying the adaptive damping coefficient by the angular velocity. The direction of this damping compensation term is opposite to the motion direction of the hydrofoil. This term is subtracted from the fused control quantity, and the subtraction result is limited to obtain the final control quantity. ; in, For the fusion control quantity, the value of the limiting range is [ [30°, 30°]. It should be noted that the angular velocity is measured in rad / s, while the control quantity is measured in degrees, and the adaptive damping coefficient... The conversion from radians to degrees has been taken into account to ensure that the damping compensation term is consistent with the dimensions of the fusion control quantity.

[0068] Furthermore, the damping compensation term is directly taken from the angular velocity of the hydrofoil rather than the error, so it is independent of the three-way PID control driven by the error and constitutes an independently tunable damping loop. The damping compensation term is superimposed with the soft damping provided by the derivative term in the standard PID controller and the derivative-leading PID controller, which is constrained by the error or the rate of change of the measured angle, to form double damping. Without affecting the control strategy determined by the aforementioned weighted fusion, it further suppresses the inherent underdamped oscillation of the second-order hydrofoil system.

[0069] The final control quantity simultaneously incorporates the control action obtained by fusing three PID controls through continuous fuzzy weight scheduling and the active damping action adaptively determined based on the real-time state. The final control quantity is output to the hydrofoil actuator to drive the hydrofoil to adjust its actual angle.

[0070] S6. Output the final control quantity to the hydrofoil actuator to drive the hydrofoil to adjust its actual angle, and use the adjusted actual angle as the acquisition object for the next cycle of angle measurement to form a closed loop.

[0071] The final control quantity is output to the hydrofoil actuator, which drives the hydrofoil to rotate and adjust its actual angle according to the final control quantity; in one embodiment, the hydrofoil actuator is a hydraulic servo valve or an electric push rod.

[0072] Furthermore, the actual angle of the hydrofoil after adjustment is acquired by the angle sensor at the next sampling moment and used as the measurement angle for the next control cycle. Accordingly, the error calculation, three-way PID parallel operation, continuous fuzzy weight scheduling, weighted fusion and adaptive active damping correction are executed sequentially in each control cycle. The measurement angle, error and the rate of change of error are updated cycle by cycle as the actual angle of the hydrofoil changes, forming a closed-loop control of the hydrofoil angle until the control ends.

[0073] In the closed-loop control, the integral separation threshold is adaptively updated periodically based on the magnitude and rate of change of the error; the first weight, the second weight, and the third weight are continuously adjusted periodically based on the absolute value and rate of change of the error; and the adaptive damping coefficient is adaptively calculated periodically based on the angular velocity and the real-time state of the system. Accordingly, the final control quantity continuously adjusts its control strategy and damping intensity according to the real-time state of the system under different hydrofoil speeds, sea states, and target angle commands.

[0074] Thus, the final control quantity integrates the control action obtained by continuous fuzzy weighted fusion of three PID controllers and the active damping action adaptively determined based on the real-time system state. These two are decoupled in two stages: a weighted fusion layer and an active damping layer. Specifically, the integral separation PID controller suppresses integral saturation under large deviations and eliminates steady-state error under small deviations through a two-dimensional adaptive threshold; the derivative-first PID controller avoids derivative shocks during target angle step changes; the continuous fuzzy weighted scheduling continuously allocates weights based on operating conditions, eliminating control quantity jitter caused by level switching; and the adaptive active damping, independent of the error loop, enhances system damping and suppresses underdamped oscillations. Therefore, this method achieves a balance between response speed, resistance to integral saturation, smoothness, adaptability, and dynamic stability. Compared to single PID control and control methods with fixed weights, fixed thresholds, and fixed damping, it achieves smaller overshoot, shorter settling time, and stronger disturbance rejection capability under all operating conditions of the hydrofoil, comprehensively improving the control performance of the hydrofoil angle.

[0075] This embodiment also discloses a hydrofoil angle fusion control system for a hydrofoil boat. The system includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, it implements an error calculation module, a parallel control module, a fuzzy weight scheduling module, a weighted fusion module, and an adaptive active damping module. Each module runs sequentially in each control cycle to jointly complete the closed-loop control of the hydrofoil angle.

[0076] The error calculation module is used to acquire the current angle of the hydrofoil in real time as the measurement angle, subtract the target angle from the measurement angle to obtain the error, and perform differential and first-order low-pass filtering on the error to obtain the rate of change of the error. The measurement angle is acquired by an angle sensor set at the hydrofoil shaft, and the target angle is issued by the driver or autopilot. The error and the rate of change of the error are output to the parallel control module, the fuzzy weight scheduling module, and the adaptive active damping module. The measurement angle is also output to the parallel control module for the calculation of the rate of change of the measurement angle under the derivative-first PID control.

[0077] The parallel control module includes a standard PID controller, an integral-separated PID controller, and a derivative-first PID controller connected in parallel. It is used to input the error into the three controllers in parallel, calculate the first control quantity, the second control quantity, and the third control quantity within the same sampling period, and output them to the weighted fusion module. The integral term of the integral-separated PID controller accumulates only when the absolute value of the error is less than the integral separation threshold, which is a two-dimensional adaptive threshold determined based on the magnitude of the error and the rate of change of the error. The derivative term of the derivative-first PID controller acts on the rate of change of the measured angle.

[0078] The fuzzy weight scheduling module is used to take the absolute value of the error and the rate of change of the error as dual inputs, and through membership degree calculation of each fuzzy set, fuzzy rule base reasoning and normalization defuzzification, continuously output the first weight, second weight and third weight corresponding to the three controllers respectively and output them to the weighted fusion module; the weights change continuously and smoothly with the absolute value and rate of change of the error.

[0079] The weighted fusion module is used to perform a normalized weighted average of the first control quantity, the second control quantity, and the third control quantity using the first weight, the second weight, and the third weight, to obtain a fused control quantity and output it to the adaptive active damping module.

[0080] The adaptive active damping module is used to acquire the angular velocity of the hydrofoil, calculate the adaptive damping coefficient in real time based on the angular velocity, the error, and the rate of change of the error, add a damping compensation term proportional to the angular velocity to the fused control quantity and limit it to obtain the final control quantity, and output the final control quantity to the hydrofoil actuator, which drives the hydrofoil to adjust its actual angle. The adaptive damping coefficient is the larger of the theoretical damping coefficient determined based on the optimal damping ratio of the second-order system and the real-time adaptive damping coefficient determined based on the real-time state of the system. The actual angle is acquired by the angle sensor in the next sampling cycle and serves as the source of the measured angle in the next control cycle, forming a closed-loop control. In one embodiment, the hydrofoil actuator is a hydraulic servo valve or an electric actuator.

[0081] To verify the control effect of this method, a second-order simulation model characterizing the dynamic characteristics of the hydrofoil angle was established. The method was simulated on this model and compared with a single standard PID control and a control method with fixed weight, fixed threshold, and fixed damping.

[0082] Angle tracking comparison when the three PID controllers operate individually Figure 16 As shown, the angle response with and without this control method is as follows: Figure 17 As shown, the angle tracking performance of this method with integrated control is as follows: Figure 15 As shown, the absolute values ​​of the errors of each control method are compared to, for example... Figure 2 As shown, the control signals of each control method are as follows: Figure 3 As shown, the control signals for fusion control in this method are as follows: Figure 14 As shown; by Figure 15 , Figure 16 and Figure 2 It is evident that no single controller can simultaneously achieve fast response, anti-saturation, and smoothness under all operating conditions, while the proposed method, which integrates control, exhibits faster absolute error convergence and smaller steady-state error. Figure 3 and Figure 14 It is evident that the control signal of this method is smooth and free from spikes caused by target step jumps.

[0083] A comparison of the integral separation effects of the two-dimensional dynamic threshold using this method and the fixed threshold is shown below. Figure 4 As shown, by Figure 4 It is evident that the two-dimensional dynamic threshold more thoroughly suppresses integral accumulation under large deviation conditions and recovers integrals in a timely manner under small deviation conditions, with both overshoot and steady-state error being superior to the fixed threshold.

[0084] The assignment and frequency of change of the three fusion weights during the control process are as follows: Figure 6As shown, the three weights transition smoothly and continuously with the absolute value and rate of change of the error, without any sudden changes at the working condition switching point, which verifies the effect of continuous fuzzy weight scheduling in eliminating the jitter of grade switching.

[0085] The effects of enabling and disabling the active damping are compared, for example... Figure 13 As shown, the range of variation of the adaptive damping coefficient is as follows: Figure 5 As shown. By Figure 13 It is evident that after superimposing the active damping on the weighted fusion, the residual oscillation amplitude is significantly reduced and the settling time is significantly shortened, indicating that the active damping independently generates an incremental effect in suppressing underdamped oscillations; Figure 5 It can be seen that the damping coefficient is automatically adjusted according to the real-time state of the system, increasing when the motion is intense and decreasing when the system is in a steady state.

[0086] This method integrates the change in the absolute value of the control error, as shown in... Figure 7 As shown, the integral saturation time accounts for, for example Figure 8 As shown, the disturbance recovery time is as follows Figure 9 As shown, peak time is for example Figure 10 As shown, the adjustment time is compared to... Figure 11 As shown, the robustness of fusion control under different operating conditions is compared as follows: Figure 12 As shown. By Figures 8 to 11 It is evident that the saturation time ratio, disturbance recovery time, peak time, and settling time of this method are significantly superior to the comparative methods; Figure 12 It is evident that this method maintains stable control performance under different operating conditions such as sailing speed and sea state, demonstrating strong robustness.

Claims

1. A method for hydrofoil angle blending control of a hydrofoil boat, characterized in that, include: The current angle of the hydrofoil is collected in real time as the measurement angle. The difference between the target angle and the measurement angle is used to obtain the error, and the rate of change of the error is obtained. The error is input in parallel to a standard PID controller, an integral separation PID controller, and a derivative-first PID controller to obtain the first control quantity, the second control quantity, and the third control quantity, respectively. With the absolute value and rate of change of the error as dual inputs, fuzzy inference continuously outputs the fusion weights corresponding to the three controllers respectively; The first control quantity, the second control quantity, and the third control quantity are normalized and weighted averaged using the fusion weights to obtain the fusion control quantity. The angular velocity of the hydrofoil is obtained, and the damping coefficient is adaptively calculated based on the angular velocity, error and its rate of change. The damping compensation term proportional to the angular velocity is superimposed on the fusion control quantity and limited to obtain the final control quantity. The final control quantity is output to the hydrofoil actuator to drive the hydrofoil to adjust its actual angle, and the adjusted actual angle is used as the acquisition object for the next cycle of angle measurement, forming a closed loop.

2. The hydrofoil angle blending control method for a hydrofoil boat according to claim 1, characterized in that, The integral separation threshold of the integral separation PID controller is a two-dimensional adaptive threshold. The two-dimensional adaptive threshold includes an amplitude term that increases with the magnitude of the error, and a trend term that changes with the rate of change of the error and decreases as the magnitude of the error increases.

3. The hydrofoil angle blending control method for a hydrofoil boat according to claim 2, characterized in that, The trend term is the product of the error rate of change influence coefficient, the absolute value of the error rate of change, and an exponential factor. The exponential factor is a natural exponent with the negative of the product of the attenuation factor and the absolute value of the error as its exponent. The two-dimensional adaptive threshold, after being limited, is used as the actual integral separation threshold used at the current sampling time.

4. The hydrofoil angle blending control method for a hydrofoil boat according to claim 1, characterized in that, The fusion weights corresponding to the three controllers, which are continuously output after fuzzy inference, include: Multiple fuzzy sets are set for the absolute value of the error and the rate of change of the error, respectively. The membership degree of the current input to each fuzzy set is calculated based on the membership function of each fuzzy set. A fuzzy rule library is established to cover the combinations of fuzzy sets that cover the absolute value of the error and the rate of change of the error. Each fuzzy rule is based on a set of recommended weight coefficients corresponding to three controllers. The product of the membership degree of the absolute value of the error and the membership degree of the rate of change of the error in each fuzzy rule is used as the activation degree of the fuzzy rule. The recommended weight coefficients corresponding to the same controller in each fuzzy rule are weighted and summed according to their activation degrees and normalized by the sum of all activation degrees to obtain the fusion weights corresponding to the three controllers.

5. The hydrofoil angle blending control method for a hydrofoil boat according to claim 4, characterized in that, Among the multiple fuzzy sets corresponding to the absolute value of the error, the fuzzy sets representing the approach to the target and the transition stage adopt a Gaussian membership function, and the fuzzy sets representing the distance from the target adopt a monotonically increasing membership function; each fuzzy set corresponding to the rate of change of the error adopts a Gaussian membership function; the membership degree of the monotonically increasing membership function increases monotonically with the increase of the absolute value of the error and approaches its upper limit.

6. The hydrofoil angle blending control method for a hydrofoil boat according to claim 1, characterized in that, The damping coefficient is adaptively calculated based on the angular velocity, error, and rate of change, including: The theoretical damping coefficient is determined based on the difference between the target damping ratio determined by the optimal damping ratio of the second-order system and the inherent damping ratio of the hydrofoil system, combined with the undamped natural frequency of the hydrofoil system and the maximum allowable angular velocity of the hydrofoil. The basic damping coefficient is then corrected in real time based on the angular velocity, error, and rate of change of error to determine the real-time adaptive damping coefficient. The larger of the theoretical damping coefficient and the real-time adaptive damping coefficient is taken as the adaptive damping coefficient.

7. The hydrofoil angle blending control method for a hydrofoil boat according to claim 6, characterized in that, The real-time correction of the basic damping coefficient includes an angular velocity term and an error term. The angular velocity term enhances damping as the angular velocity increases, while the error term enhances damping with the magnitude of the error and weakens this enhancement effect by an exponential factor with the rate of change of the error.

8. The hydrofoil angle blending control method for a hydrofoil boat according to claim 1, characterized in that, The damping compensation term is determined based on the angular velocity of the hydrofoil and is independent of the standard PID controller, integral-separated PID controller, and derivative-first PID controller based on the error. The weighted fusion and the superposition of the damping compensation constitute two separate stages. The first stage weights and fuses the first control quantity, the second control quantity, and the third control quantity according to the fusion weight to obtain the fused control quantity. The second stage superimposes the damping compensation term onto the fused control quantity based on the angular velocity. The damping compensation term is superimposed on the damping provided by the derivative term of the standard PID controller and the derivative-first PID controller.

9. The hydrofoil angle blending control method for a hydrofoil boat according to claim 1, characterized in that, The third control quantity of the derivative-first PID controller is obtained by summing the proportional term, integral term, and derivative term and limiting the amplitude. The proportional term is determined based on the error and proportional gain, the integral term is determined based on the integral accumulation of the error and integral gain, and the derivative term is determined based on the rate of change of the measured angle and derivative gain, with its negative value participating in the summation. The rate of change of the measured angle is obtained by differentiating the measured angle and passing it through a first-order low-pass filter.

10. A hydrofoil angle fusion control system for a hydrofoil boat, characterized in that, include: The error calculation module is used to acquire the measurement angle of the hydrofoil in real time, calculate the error by subtracting the target angle from the measurement angle, obtain the rate of change of the error, output the error to the parallel control module and the fuzzy weight scheduling module, and output the measurement angle to the parallel control module. The parallel control module includes a standard PID controller, an integral-separated PID controller, and a derivative-first PID controller connected in parallel. It is used to perform parallel calculations on the error to obtain a first control quantity, a second control quantity, and a third control quantity, and output them to the weighted fusion module. The fuzzy weight scheduling module is used to take the absolute value and rate of change of the error as dual inputs, and continuously output the fusion weights corresponding to the three controllers through fuzzy inference and output them to the weighted fusion module. The weighted fusion module is used to perform a normalized weighted average of the first control quantity, the second control quantity, and the third control quantity using the fusion weights, to obtain the fused control quantity and output it to the adaptive active damping module. An adaptive active damping module is used to acquire the angular velocity of the hydrofoil, adaptively calculate the damping coefficient based on the angular velocity, the error and its rate of change, and superimpose a damping compensation term proportional to the angular velocity onto the fused control quantity and limit it to obtain the final control quantity. The final control quantity is used to drive the hydrofoil to adjust its actual angle, and the adjusted actual angle serves as the source of the measured angle in the next cycle, forming a closed loop.

Images