一种用于水下仿生设备的多模块独立供电系统

Abstract

本发明属于水下仿生设备供电技术领域，具体为一种用于水下仿生设备的多模块独立供电系统，所述水下仿生设备包括尾鳍推进组件，供电系统集成于水下仿生设备中，供电系统包括第一电源组件和第二电源组件，通过解算实时尾鳍摆动相位，预测尾鳍推进组件预设时间窗口的离散功率需求序列，结合第二电源组件的安全输出功率阈值，生成涉及第一电源组件和第二电源组件的供电策略，经功率变化梯度约束处理，针对供电策略确定的响应电源组件，下发包含时间戳与目标输出功率值的响应指令集，并依据实际输出的跟踪误差进行稳定评估与闭环修正，实现水下仿生设备功率的主动、安全与平滑供给，提高高机动动作中功率需求快速变化的应对能力。

Description

Technical Field

[0001] This invention belongs to the field of underwater bionic equipment power supply technology, specifically a multi-module independent power supply system for underwater bionic equipment. Background Technology

[0002] Underwater biomimetic equipment is a type of special underwater operation equipment developed by imitating the movement mechanism and morphological characteristics of marine organisms. It usually adopts a modular design, and its power system, control system, and sensing system are often equipped with independent power modules to meet the requirements of layout, counterweight and safety isolation.

[0003] In existing technologies, solutions for powering multiple independent modules of underwater biomimetic devices often focus on power redundancy and static allocation. For example, a power supply system for an underwater power supply device and an underwater robot, as disclosed in Chinese Patent Publication No. CN218829225U, improves system reliability by setting up mutually isolated main power supply compartments and backup battery compartments to provide fixed or predefined power supplies to different functional units.

[0004] Another Chinese patent publication, CN112994124A, describes an underwater combined energy supply method and system that uses a combination of fuel cells and batteries. It performs macroscopic power switching and selection based on the working status of the electrical equipment or the remaining battery power to enhance the system's fault tolerance and flexibility.

[0005] However, when underwater biomimetic devices perform highly maneuverable actions requiring explosive propulsion, such as rapid ascent from the deep sea to the surface, power peaks occur concentratedly, primarily driven by the transient and fluctuating propulsion power demands generated by the high-frequency oscillation of the tail fin. Existing technologies exhibit the following shortcomings in addressing this scenario: 1. Existing technologies lack real-time analysis and utilization of the tail fin's oscillation motion patterns, and cannot predict future instantaneous power demand sequences based on the oscillation motion state. Therefore, they remain in a passive response state and cannot provide a basis for advance power scheduling and preparation.

[0006] 2. In the face of fluctuating propulsion power requirements, the power modules distributed in the body and tail of the existing solution mostly adopt simple parallel or fixed master and backup mode. They fail to dynamically plan the power output ratio and timing of each power supply according to the real-time calculated motion state and predicted power demand, resulting in limited peak power support capability or low energy utilization during valley periods. Summary of the Invention

[0007] To overcome the shortcomings of the prior art, embodiments of the present invention provide a multi-module independent power supply system for underwater biomimetic devices, which can effectively solve the problems involved in the prior art.

[0008] The objective of this invention can be achieved through the following technical solution: a multi-module independent power supply system for an underwater bionic device, the underwater bionic device including a tail fin propulsion component; the power supply system is integrated into the underwater bionic device, and the power supply system includes: a first power supply component, a second power supply component, a swing phase calculation module, a power supply strategy generation module, a response command issuance module, and a correction signal feedback module.

[0009] The swing phase calculation module and the power supply strategy generation module are connected; the power supply strategy generation module is connected to the response command issuing module; the response command issuing module is connected to the first power supply component and the second power supply component respectively, and is used to issue a response command set to the response power supply component determined by the power supply strategy; the first power supply component and the second power supply component are both connected to the correction signal feedback module, which is used to receive the actual output power value returned by each power supply component; the correction signal feedback module is connected to the power supply strategy generation module, and is used to transmit the correction signal to the power supply strategy generation module. The power supply strategy generation module, based on the execution deviation reflected by the correction signal and combined with the response power supply component determined by its power supply strategy, drives the response command issuing module to adjust the subsequent response command set through the connection relationship with the response command issuing module, thereby indirectly realizing the correction signal feedback to the subsequent response power supply component and the response command set.

[0010] The first power supply component and the second power supply component provide power to the tail fin propulsion component, respectively.

[0011] The swing phase calculation module calculates the real-time swing phase based on the tail fin swing angle data, and predicts the discrete power demand sequence of the tail fin propulsion component within a preset time window based on the continuous change trend of the swing phase.

[0012] The power supply strategy generation module extracts the power peak value of the discrete power demand sequence and compares it with the safe output power threshold of the second power supply component. Based on the comparison result, it generates a power supply strategy involving the first power supply component and the second power supply component.

[0013] The response command issuing module performs power change gradient constraint processing on the discrete power demand sequence and issues a response command set containing timestamps and target output power values ​​to the response power components determined by the power supply strategy.

[0014] The correction signal feedback module receives the actual output power value returned by each power supply component. Through tracking error calculation and stability performance evaluation, the correction signal is transmitted to the power supply strategy generation module. Through the linkage processing of the power supply strategy generation module and the response command issuing module, the correction signal is fed back to the subsequent response power supply components and response command set.

[0015] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) The present invention calculates the normalized phase of the tail fin swing in real time, and uses the swing phase and angular velocity to perform hydrodynamic calculations, predicts the discrete power demand sequence of the preset time window, realizes the transformation from passive response to active prediction, and predicts power fluctuations in advance based on the movement law of the tail fin itself, providing time margin and data basis for the pre-scheduling of power components, thereby effectively coping with the rapid changes in power demand during high-mobility maneuvers.

[0016] (2) This invention predefines the safe output power threshold of the second power supply component and compares it with the power peak of the discrete power demand sequence to dynamically generate a collaborative power supply strategy: when the peak exceeds the limit, the first power supply component supplements the difference; otherwise, the second power supply component works independently. This achieves a refined division of labor among power supply components based on real-time demand and their own capability boundaries, ensuring the supply of transient peak power while avoiding the risk of power overload.

[0017] (3) This invention performs power change gradient constraint processing on discrete power demand sequences, restricting the power change gradient within the physical allowable range of the power supply, and binds it with timestamps to form an instruction set. Furthermore, it generates a feedforward compensation signal through tracking error calculation and stability performance evaluation. This constitutes a closed-loop control of instruction smoothing, execution tracking, and error compensation, effectively suppressing output oscillations caused by instruction jumps or load disturbances, and ensuring the stability and tracking accuracy of power output under dynamic operating conditions. Attached Figure Description

[0018] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the module connection of the present invention.

[0020] Figure 2 This is a schematic diagram of the power variation gradient constraint processing of the present invention.

[0021] Figure 3 This is a schematic diagram of the corrected signal feedback of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Reference Figure 1 As shown, the present invention provides a multi-module independent power supply system for an underwater biomimetic device, the underwater biomimetic device including a tail fin propulsion assembly; the power supply system is integrated into the underwater biomimetic device, and the power supply system includes: a first power supply component, a second power supply component, a swing phase calculation module, a power supply strategy generation module, a response command issuance module, and a correction signal feedback module.

[0024] The first power supply component is a high-power-density power supply component, which is directly deployed in the tail fin area. Its design goal is to provide instantaneous high current output capability to meet the pulsed peak power requirements generated when the tail fin swings at high frequency.

[0025] The second power supply unit is a high-energy-density power supply unit, consisting of multiple independent power supply modules deployed within the main structure of the device. The main structure of the device refers to the main compartment of the non-oscillating propulsion section in the underwater biomimetic device body, and its specific spatial scope varies depending on the mechanical structure design of different biomimetic shapes (such as fish-shaped, whale-shaped, etc.). The design of the second power supply unit focuses on providing a continuous basic power supply.

[0026] This invention is applicable to the power supply control of underwater biomimetic devices when performing highly maneuverable actions such as high-speed surfacing and rapid turning. These actions rely on the high-frequency oscillation of the tail fin, and their power demand exhibits rapid pulsation characteristics that are strongly correlated with the oscillation motion.

[0027] The swing phase calculation module and the power supply strategy generation module are connected; the power supply strategy generation module is connected to the response command issuing module; the response command issuing module is connected to the first power supply component and the second power supply component respectively, and is used to issue a response command set to the response power supply component determined by the power supply strategy; the first power supply component and the second power supply component are both connected to the correction signal feedback module, which is used to receive the actual output power value returned by each power supply component; the correction signal feedback module is connected to the power supply strategy generation module, and is used to transmit the correction signal to the power supply strategy generation module. The power supply strategy generation module, based on the execution deviation reflected by the correction signal and combined with the response power supply component determined by its power supply strategy, drives the response command issuing module to adjust the subsequent response command set through the connection relationship with the response command issuing module, thereby indirectly realizing the correction signal feedback to the subsequent response power supply component and the response command set.

[0028] The first power supply component and the second power supply component provide power to the tail fin propulsion component, respectively.

[0029] The swing phase calculation module includes a continuous processing procedure of real-time swing phase calculation and discrete power demand sequence prediction, following the logical sequence from motion signal acquisition to power demand estimation.

[0030] Specifically, the real-time swing phase calculation process calculates the real-time swing phase of the underwater bionic device's tail fin based on the tail fin swing angle data collected in real time by the tail fin swing angle sensor. The specific execution steps include: extracting the tail fin swing angle data sequence according to a fixed sampling period.

[0031] The first-order difference calculation is performed on the tail fin swing angle data sequence to obtain the swing angle change rate, and the zero-crossing time point corresponding to the change rate of the swing angle changing from negative to positive is identified.

[0032] The time interval between two consecutively identified zero-crossing time points is defined as the oscillation period, and the zero-crossing time point is marked as the starting phase zero point of the current oscillation period.

[0033] Using the zero point of the starting phase of the current oscillation cycle as the time reference, calculate the time offset at any subsequent sampling time point.

[0034] Divide the time offset by the total duration of the current oscillation cycle, and then multiply by 360 degrees to calculate the real-time absolute phase value, expressed in angles, corresponding to the sampling time point.

[0035] Perform a modulo 360 operation on the real-time absolute phase value and map it to a preset angle range to obtain the normalized real-time oscillation phase, wherein the preset angle range specifically refers to .

[0036] Next, the discrete power demand sequence prediction process predicts the discrete power demand sequence of the tail fin propulsion component within a preset time window based on the continuous change trend of the swing phase. The specific execution steps include: S1. Calculating the first derivative of the normalized real-time swing phase to obtain the real-time angular velocity.

[0037] S2. Perform hydrodynamic calculations based on real-time oscillation phase and real-time angular velocity: S21. Establish a coordinate system for the tail fin itself, fixed to the tail fin, and a global coordinate system fixed to the underwater biomimetic device body. The origin of the tail fin coordinate system is located at the center of rotation of the tail fin, with its X-axis pointing towards the tip along the tail fin's span, and its Y-axis perpendicular to the tail fin plane pointing towards the dorsal side. The origin of the global coordinate system is located at the geometric center of the device body, with its X-axis pointing towards the standard forward direction of the device, and its Y-axis determined according to the right-hand rule, typically perpendicular to the forward direction of the device and pointing horizontally.

[0038] For the initial plane normal vector of the caudal fin in a neutral position, a rotational transformation is performed around the fixed rotation axis of the equipment, with a real-time swing phase angle, to calculate the real-time normal vector of the caudal fin plane in the global coordinate system. This rotational transformation operation is implemented using the Rodriguez rotation formula or rotation matrix, which is existing technology and will not be detailed here.

[0039] Obtain the position vector of a predetermined feature point (such as the midpoint of the leading edge) on the caudal fin relative to the rotation center in the caudal fin body coordinate system. Based on the real-time oscillation phase, transform the position vector from the caudal fin body coordinate system to the global coordinate system through a rigid body rotation transformation matrix.

[0040] The instantaneous velocity vector of the predetermined feature point due to the tail fin swing is calculated by performing a cross product operation between the real-time angular velocity vector and the transformed global coordinate system position vector.

[0041] The current forward velocity vector of the underwater biomimetic device body in the global coordinate system is obtained. The instantaneous motion velocity vector of the predetermined feature point is vector synthesized with the current forward velocity vector to obtain the relative inflow velocity vector perceived at the predetermined feature point of the tail fin.

[0042] The relative incoming flow velocity vector is projected onto the caudal fin plane determined by the current normal vector to obtain the incoming flow velocity components in the plane.

[0043] The angle between the in-plane velocity component of the flow and the predefined chord direction vector in the tail fin body coordinate system is calculated and used as the real-time fluid angle of attack.

[0044] S22. Retrieve the pre-constructed tail fin hydrodynamic coefficient table. The process of constructing the tail fin hydrodynamic coefficient table is as follows: Establish a three-dimensional geometric model of the tail fin based on the design drawings of the underwater biomimetic equipment.

[0045] The angle of attack of the fluid is determined based on the maximum design sway angle of the tail fin propulsion component. The symmetrical range of values, such as -30 degrees to 30 degrees, is used to determine the angular velocity based on the peak angular velocity designed for the tail fin propulsion component under the highest frequency oscillation condition. The symmetrical range of values, for example, from -10 rad / s to 10 rad / s.

[0046] The symmetric ranges of the fluid's angle of attack and angular velocity are discretized at fixed intervals (e.g., 1 degree interval for angle of attack and 0.5 rad / s interval for angular velocity), forming a two-dimensional parametric grid. Each grid point corresponds to a... Operating conditions.

[0047] For each grid point and its corresponding operating condition, taking the fluid dynamics simulation process as an example, the established three-dimensional geometric model of the tail fin is imported into the CFD preprocessing software to create the computational domain, generate the mesh, and set the boundary conditions. Using the CFD solver, numerical simulations of the flow field are performed on each operating condition within the two-dimensional parametric mesh to directly obtain the fluid forces acting on the tail fin under the corresponding motion state, including normal and tangential forces.

[0048] Inverse calculation based on fluid dynamics formulas Lift coefficient under operating conditions and drag coefficient .

[0049] The calculated The data pairs are stored as tables, which constitute the tail fin hydrodynamic coefficient table, storing the lift coefficient and drag coefficient of the tail fin profile under different combinations of fluid angle of attack and angular velocity.

[0050] During the query, the cell in the two-dimensional parametric grid is located based on the current fluid angle of attack and the current angular velocity value.

[0051] Read the pre-stored lift coefficient and drag coefficient at the four vertices of the cell.

[0052] Based on the proportional relationship between the current angular velocity and the angular velocities of the upper and lower boundaries of the cell, linear interpolation is performed on the two sets of coefficients located at the same angle of attack to obtain two sets of intermediate coefficient values.

[0053] Based on the proportional relationship between the current fluid angle of attack and the angle of attack of the left and right boundaries of the cell, the two sets of intermediate coefficient values ​​are linearly interpolated again to obtain the lift coefficient and drag coefficient that match the current fluid angle of attack and the current angular velocity.

[0054] S23. Calculate the normal and tangential forces acting on the tail fin at the current time point using fluid dynamics formulas. The fluid dynamics formulas are as follows: .

[0055] In the formula, The density of water, The calibration area of ​​the tail fin of the underwater bionic device. The relative incoming flow velocity is calculated by combining the forward velocity of the underwater bionic device and the linear velocity of its tail fin oscillation. or Contemporary The fluid normal force is obtained by solving the formula, and then introduced into the time equation. The fluid tangential force is obtained by solving the formula.

[0056] This is the hydrodynamic pressure term, representing the theoretical pressure per unit area generated by the relative motion between the fluid and the underwater biomimetic device.

[0057] The fluid dynamics formula is based on the principle of dynamic pressure in fluid mechanics. By multiplying the fluid dynamic pressure, fixed structural parameters, and the tail fin hydrodynamic coefficient obtained from real-time operating condition queries, it achieves quantitative calculation from motion state and geometric parameters to specific fluid forces. This is achieved through switching coefficients. Types ( or The same formula framework can be used to solve for the normal force that is beneficial to propulsion and the tangential force that hinders motion, providing a basic force input for subsequent calculations of propulsion torque and power requirements.

[0058] S24. Based on the position of the rotation center of the caudal fin under the current phase, decompose the fluid normal force and tangential force into torque components generated about the rotation center of the caudal fin.

[0059] S25. In the global coordinate system, the sum of the projection components of the torque components in the forward direction of the device is taken as the predicted propulsion torque required by the tail fin propulsion assembly to overcome fluid resistance in the current swing phase.

[0060] S3. Multiply the predicted propulsion torque by the current angular velocity to obtain the instantaneous mechanical power requirement of the tail fin propulsion assembly at the current time point. Divide the instantaneous mechanical power requirement by the mechanical transmission efficiency coefficient of the drive motor in the tail fin propulsion assembly to obtain the mechanical power required to be provided by the drive motor shaft end.

[0061] The mechanical power required to be provided by the drive motor shaft end is added to the static power loss of the drive motor to obtain the corresponding instantaneous power demand value.

[0062] S4. Starting from the current time point, and based on the continuous change trend of the swing phase, calculate the instantaneous electrical power demand value corresponding to each time point within the future set time period of the tail fin propulsion component.

[0063] S5. Integrate the series of instantaneous power demand values ​​obtained by calculation in chronological order to generate a discrete power demand sequence corresponding to a preset time window.

[0064] This invention provides a real-time solution for the normalized phase of the tail fin swing and uses the swing phase and angular velocity to perform hydrodynamic calculations, predicting the discrete power demand sequence within a preset time window. This enables a shift from passive response to active prediction. Based on the tail fin's own motion patterns, it anticipates power fluctuations in advance, providing a time margin and data foundation for the pre-scheduling of power components, thereby effectively addressing rapid changes in power demand during high-maneuverability maneuvers.

[0065] The power supply strategy generation module formulates a power supply strategy corresponding to a preset time window based on the predicted power demand of the tail fin propulsion component and the safety capability boundary of the power supply component.

[0066] The power supply strategy generation module is designed based on the following execution logic: First, a safe output power threshold is defined for the high energy density second power supply component to ensure that the power supply component is within the range of continuous power supply capability. The specific definition process includes: obtaining the output voltage time series data of the second power supply component under historical discharge conditions, arranging it in ascending order of output voltage to form an output voltage dataset, and extracting the preset low percentile value from the output voltage dataset as the feature voltage value.

[0067] Obtain the maximum permissible continuous discharge current calibrated at the rated operating temperature of the cell material in the second power supply component, multiply it by the characteristic voltage value, and define the product as the safe output power threshold of the second power supply component.

[0068] Secondly, the power peak of the discrete power demand sequence is extracted and compared with the safe output power threshold of the second power supply component. Based on the comparison result, a power supply strategy involving the first power supply component and the second power supply component is generated. The generation logic of the power supply strategy is as follows: when the power peak of the discrete power demand sequence is greater than the safe output power threshold of the second power supply component, the first power supply component is designated to provide the difference power exceeding the safe output power threshold, while the second power supply component provides the base power less than or equal to the safe output power threshold.

[0069] Conversely, when the peak power of the discrete power demand sequence is less than or equal to the safe output power threshold of the second power supply component, it is specified that the second power supply component alone provides all the power required at each time point in the discrete power demand sequence.

[0070] It should be noted that the aforementioned preset low percentile value is based on the following principle: when the second power supply component outputs a large current, its output voltage will drop due to the internal equivalent DC resistance; the larger the current, the more significant the voltage drop. Therefore, under the maximum continuous discharge current condition, the output voltage is at the theoretical lowest level throughout the entire task cycle. Using this low percentile voltage essentially selects a voltage valley value close to the actual voltage reached during historical high-load discharge periods. Using this as a benchmark to calculate the safe output power threshold ensures that even under extreme conditions of peak power demand and lowest output voltage, the system's required power output remains within the safe capability range of the second power supply component.

[0071] In this invention, the preset low percentile value can be exemplarily set as the 5th percentile. Those skilled in the art can make directional adjustments to this exemplary value according to the DC internal resistance characteristics of the second power component cell: if the DC internal resistance of the cell is low and the voltage drop under high current is not significant, it can be appropriately relaxed to a higher percentile, such as the 10th percentile. If the DC internal resistance is high and the voltage drop is significant, it needs to be tightened to a lower percentile, such as the 2nd percentile.

[0072] This invention, through pre-defining a safe output power threshold for a second power supply component and comparing it with the peak power of a discrete power demand sequence, dynamically generates a collaborative power supply strategy: when the peak power exceeds the limit, the first power supply component compensates for the difference; otherwise, the second power supply component operates independently. This achieves a refined division of labor among power supply components based on real-time demand and their own capability boundaries, ensuring the supply of transient peak power while avoiding the risk of power overload.

[0073] Given that the power supply components have a physical maximum allowable power change gradient limit, in order to ensure the executability of the instructions, the response instruction issuing module needs to first perform power change gradient constraint processing on the discrete power demand sequence. After the power change gradient is smoothly limited to within the maximum allowable power change gradient, a response instruction set containing timestamps and target output power values ​​is issued for the response power supply components determined by the power supply strategy.

[0074] Reference Figure 2 As shown, the power change gradient constraint processing of the discrete power demand sequence includes: sequentially traversing the discrete power demand sequence, and for any two adjacent instantaneous power demand values ​​in the sequence, dividing the absolute difference between the subsequent instantaneous power demand value and the preceding instantaneous power demand value by the sampling time interval to calculate the original power change gradient.

[0075] The time period to which consecutive data points whose original power change gradient is greater than the predetermined maximum allowable power change gradient of the power component belong is marked as the interval to be adjusted.

[0076] Starting with the instantaneous power demand value at the beginning of the interval, and using the maximum allowable power change gradient of the power supply components as a constant ramp rate, calculate the theoretical shortest time required to reach the instantaneous power demand value at the end of the interval.

[0077] If the theoretical minimum duration is greater than the span duration of the interval to be adjusted, the difference is used as the feedforward compensation duration, and the starting time point of the interval is shifted forward to obtain the expanded adjustment interval; otherwise, the original interval remains unchanged.

[0078] For the expanded adjustment range, a linear power ramp-up path is generated with the maximum allowable power change gradient of the power components as the slope, leading to the instantaneous power demand value at the end of the adjustment range.

[0079] The linear power ramp-up path is resampled at fixed time intervals according to the discrete power demand sequence to obtain the adjusted power point.

[0080] The adjusted power points are merged with the unlabeled original sequence points by timestamp. The merging follows the time coverage principle: for any given time point, if an adjusted power value exists, that adjusted value is used as the final value. If the same time point is calculated multiple times due to overlapping adjustment intervals, the last calculated adjusted value is used. Finally, a final power demand sequence that satisfies the rate of change constraint is generated.

[0081] It should be further noted that the predetermined maximum permissible power variation gradient of the power supply component is determined based on the transient response capability specified in the technical specifications of the corresponding power supply component, combined with the measured step response characteristics of the system. Specifically, the maximum safe power variation gradient under actual operating conditions can be calculated by applying a standard step load to the power supply component, measuring the shortest time required for its output voltage to recover to the specified tolerance range, and dividing the maximum power jump amplitude during the standard step process by the shortest time.

[0082] The response instruction set generation process includes: binding each power data point in the discrete power demand sequence after power change gradient constraint processing to its corresponding absolute time point within a preset time window to generate a series of timestamp-power value pairs.

[0083] All generated timestamp-power value pairs are encapsulated in chronological order to form a response instruction set, which is then sent to the response power components determined by the power supply strategy.

[0084] Reference Figure 3 As shown, in order to address the inherent response delay and steady-state deviation of the inner loop control under dynamic operating conditions, and to suppress the impact of external load disturbances on power supply quality, the correction signal feedback module receives the actual output power value returned by each power supply component, and through tracking error calculation and stability performance evaluation, feeds back correction signals to the subsequent response power supply components and response command set, thereby forming a closed-loop feedforward compensation mechanism.

[0085] The tracking error calculation process includes: collecting the actual output power values ​​from the local control units of the first power supply component and the second power supply component, respectively.

[0086] The actual output power value is compared point by point with the target output power value corresponding to the timestamp in the response instruction set. The difference between the actual output power value and the target output power value is used as the power tracking error, thereby obtaining the power tracking error at each time point within the preset time window of the response power supply component.

[0087] The stability performance evaluation includes: performing root mean square calculation on the continuous power tracking error sequence within a preset time window to obtain stability performance evaluation indicators.

[0088] If the stability performance evaluation index exceeds the preset tolerance threshold, a correction signal will be triggered.

[0089] The feedback correction signal to the subsequent response power supply component and response command set includes: based on the power tracking error at each time point within a preset time window of the response power supply component, using the sliding window extreme value method, a small window of fixed width (e.g., 50ms) slides on the power tracking error sequence, and each time the maximum absolute value of the error within the window is taken as the envelope amplitude estimate at the center time of the small window, thereby extracting the envelope of the power tracking error changing with time, and calculating the area enclosed between it and the zero error baseline through integration, which is denoted as the error energy.

[0090] The error energy is divided by the total duration of a preset time window to obtain the power compensation reference value.

[0091] The power compensation reference value for the response power supply component is encapsulated as a correction signal, prompting it to be algebraically superimposed with the target output power value at the corresponding time point in the subsequent time period to generate a corrected target output power value to replace the original instruction value.

[0092] This invention applies power change gradient constraint processing to discrete power demand sequences, limiting the power change gradient to within the physically permissible range of the power supply, and binding it with timestamps to form an instruction set. Furthermore, it generates a feedforward compensation signal through tracking error calculation and stability performance evaluation. This constitutes a closed-loop control system for instruction smoothing, execution tracking, and error compensation, effectively suppressing output oscillations caused by instruction jumps or load disturbances, and ensuring the stability and tracking accuracy of power output under dynamic operating conditions.

[0093] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A multi-module independent power supply system for underwater biomimetic devices, characterized in that, The underwater biomimetic device includes a tail fin propulsion assembly; the power supply system is integrated into the underwater biomimetic device, and the power supply system includes: The first power supply component and the second power supply component provide power to the tail fin propulsion component, respectively. The swing phase calculation module calculates the real-time swing phase based on the tail fin swing angle data, and predicts the discrete power demand sequence of the tail fin propulsion component within a preset time window based on the continuous change trend of the swing phase. The power supply strategy generation module extracts the power peak value of the discrete power demand sequence and compares it with the safe output power threshold of the second power supply component. Based on the comparison result, it generates a power supply strategy involving the first power supply component and the second power supply component. The response instruction issuing module performs power change gradient constraint processing on the discrete power demand sequence and issues a response instruction set containing timestamps and target output power values ​​for the response power components determined by the power supply strategy. The correction signal feedback module receives the actual output power value returned by each power supply component. Through tracking error calculation and stability performance evaluation, the correction signal is transmitted to the power supply strategy generation module. Through the linkage processing of the power supply strategy generation module and the response command issuing module, the correction signal is fed back to the subsequent response power supply components and response command set.

2. The multi-module independent power supply system for underwater bionic devices according to claim 1, characterized in that, The calculation of the real-time oscillation phase based on the tail fin oscillation angle data includes: Extract the tail fin swing angle data sequence according to a fixed sampling period; The first-order difference calculation is performed on the tail fin swing angle data sequence to obtain the swing angle change rate, and the zero-crossing time point corresponding to the change rate of the swing angle change from negative to positive is identified. The time interval between two consecutively identified zero-crossing time points is defined as the oscillation period, and the zero-crossing time point is marked as the starting phase zero point of the current oscillation period; Using the zero point of the initial phase of the current oscillation cycle as the time reference, calculate the time offset at any subsequent sampling time point; Divide the time offset by the total duration of the current oscillation cycle, and then multiply by 360 degrees to calculate the real-time absolute phase value in terms of angle corresponding to the sampling time point; The real-time absolute phase value is modulo 360 and mapped to a preset angle range to obtain the normalized real-time swing phase.

3. A multi-module independent power supply system for underwater bionic devices according to claim 2, characterized in that, The discrete power demand sequence for the predicted tail fin propulsion component within a preset time window includes: Calculate the first derivative of the normalized real-time oscillation phase to obtain the real-time angular velocity; Based on the real-time oscillation phase and real-time angular velocity, hydrodynamic calculations are performed to obtain the predicted propulsion torque required by the tail fin propulsion assembly to overcome fluid resistance under the real-time oscillation phase. Multiply the predicted propulsion torque by the current angular velocity to obtain the instantaneous mechanical power requirement of the tail fin propulsion assembly at the current time point. Divide the instantaneous mechanical power requirement by the mechanical transmission efficiency coefficient of the drive motor in the tail fin propulsion assembly to obtain the mechanical power required to be provided by the drive motor shaft end. Add the mechanical power required to be provided by the drive motor shaft end to the static power loss of the drive motor to obtain the corresponding instantaneous power demand value; Starting from the current time point, based on the continuous change trend of the swing phase, calculate the instantaneous electric power demand value corresponding to the tail fin propulsion component for each time series point within a set future time period; A series of instantaneous power demand values ​​obtained by integrating and calculating in chronological order are used to generate a discrete power demand sequence with a preset time window.

4. A multi-module independent power supply system for underwater bionic devices according to claim 1, characterized in that, The process for defining the safe output power threshold of the second power supply component includes: The output voltage timing data of the second power supply component under historical discharge conditions is obtained, and the output voltage is arranged in ascending order to form an output voltage dataset. The preset low percentile value is extracted as the feature voltage value used for the threshold calculation of the floating stage. Obtain the maximum permissible continuous discharge current calibrated at the rated operating temperature of the cell material in the second power supply component, multiply it by the characteristic voltage value, and define the product as the safe output power threshold of the second power supply component.

5. A multi-module independent power supply system for underwater bionic devices according to claim 1, characterized in that, The power supply strategy includes: When the peak power of the discrete power demand sequence is greater than the safe output power threshold of the second power supply component, it is specified that the first power supply component provides the difference power exceeding the safe output power threshold, while the second power supply component provides the base power less than or equal to the safe output power threshold. Conversely, when the peak power of the discrete power demand sequence is less than or equal to the safe output power threshold of the second power supply component, it is specified that the second power supply component alone provides all the power required at each time point in the discrete power demand sequence.

6. A multi-module independent power supply system for underwater bionic devices according to claim 3, characterized in that, The process of applying power change gradient constraints to the discrete power demand sequence includes: The discrete power demand sequence is sequentially traversed, and the original power change gradient is calculated for any two adjacent instantaneous power demand values ​​in the sequence. The time period to which consecutive data points whose original power change gradient is greater than the predetermined maximum allowable power change gradient of the power component belong is marked as the interval to be adjusted; Starting with the instantaneous power demand value at the beginning of the interval, and using the maximum allowable power change gradient of the power supply components as a constant ramp rate, calculate the theoretical shortest time required to reach the instantaneous power demand value at the end of the interval. If the theoretical shortest duration is greater than the duration of the interval to be adjusted, the difference is used as the feedforward compensation duration, and the starting time point of the interval is shifted forward to obtain the expanded adjustment interval; otherwise, the original interval remains unchanged. In the expanded adjustment range, a linear power ramp-up path is generated to the instantaneous power demand value at the end of the adjustment range, using the maximum allowable power change gradient of the power components as the slope. The linear power ramp-up path is resampled at fixed time intervals according to the discrete power demand sequence to generate the final power demand sequence that satisfies the power change gradient constraint.

7. A multi-module independent power supply system for underwater bionic devices according to claim 1, characterized in that, The response instruction set includes: Each power data point in the discrete power demand sequence after power change gradient constraint processing is bound to its corresponding absolute time point within a preset time window to generate a series of timestamp-power value pairs. All generated timestamp-power value pairs are encapsulated in chronological order to form a response instruction set, which is then sent to the response power components determined by the power supply strategy.

8. A multi-module independent power supply system for underwater bionic devices according to claim 1, characterized in that, The tracking error calculation process includes: The actual output power values ​​are collected from the local control units of the first power supply component and the second power supply component, respectively. The actual output power value is compared point by point with the target output power value corresponding to the timestamp in the response instruction set. The difference between the actual output power value and the target output power value is used as the power tracking error, thereby obtaining the power tracking error of the response power supply component at each time point.

9. A multi-module independent power supply system for underwater bionic devices according to claim 8, characterized in that, The stability performance assessment includes: The root mean square calculation is performed on the continuous power tracking error sequence within a preset time window to obtain a stability performance evaluation index. If the stability performance evaluation index exceeds the preset tolerance threshold, a correction signal will be triggered.

10. A multi-module independent power supply system for underwater bionic devices according to claim 9, characterized in that, The step of feeding back correction signals to subsequent response power components and response command sets includes: Based on the power tracking error at each time point within the preset time window of the response power component, its envelope that changes over time is extracted, and the area enclosed between it and the zero error baseline is calculated through integration, which is denoted as the error energy. The error energy is divided by the total duration of a preset time window to obtain the power compensation reference value; The power compensation reference value for the response power supply component is encapsulated as a correction signal, prompting it to be algebraically superimposed with the target output power value at the corresponding time point in the subsequent time period to generate a corrected target output power value to replace the original instruction value.

Images