Multi-target intelligent control system and method for hydrodynamic performance of underwater vehicle

Abstract

The invention belongs to the technical field of fluid mechanics and flow control, and discloses a multi-target intelligent control system and method for hydrodynamic performance of an underwater vehicle, and the system is characterized in that a control target is of an axisymmetric body structure, the tail part of the control target is provided with at least one active blowing jet outlet, and an execution unit is connected with the jet outlet of the control target; the jet flow blowing state is adjusted according to the control instruction output by the control logic module. The sensing unit is arranged on the surface of the tail of the axisymmetric body and used for collecting key parameters reflecting the flow control effect and the energy cost. And the control logic module is used for receiving the flow information output by the sensing unit and carrying out real-time or periodic optimization on the control parameters based on a preset optimization algorithm. And the optimized control parameters are sent to the execution unit by the control logic module, so that the system forms closed-loop operation. According to the technical scheme, collaborative optimization of resistance reduction and blowing energy consumption reduction can be achieved, and the adaptability and overall energy efficiency of the control system to different working conditions are improved.

Description

Technical Field

[0001] This invention belongs to the field of fluid mechanics and flow control technology, and in particular relates to a multi-objective intelligent control system and method for the hydrodynamic performance of underwater vehicles. Background Technology

[0002] Drag reduction design for axisymmetric or submarine-shaped vehicles such as aircraft and underwater vehicles is a key aspect of improving their hydrodynamic performance and propulsion efficiency. Currently, drag reduction technologies related to these shapes can be broadly classified into two categories: passive control and active control. Passive control typically improves flow characteristics through methods such as shape optimization, the installation of vortex generators, or the adjustment of surface roughness; active control mainly relies on methods such as jet blowing, suction, and plasma excitation to directly intervene in near-wall flow to suppress or delay flow separation.

[0003] In the field of active flow control, tailpipe blowing technology has been proven to effectively improve the tailpipe flow field structure, thereby significantly enhancing tailpipe pressure recovery performance. Existing technologies have conducted experimental studies on Ahmed-type vehicle body models, systematically comparing single-jet outlet and multi-jet outlet flow distribution schemes optimized using ant colony optimization. The results show that the optimized multi-jet distribution strategy can further promote tailpipe pressure recovery and reduce aerodynamic drag. However, this type of research object involves a vehicle body with significant three-dimensional asymmetry, whose wake structure differs fundamentally from that of typical axisymmetric bodies. Therefore, the applicability and generalization of the relevant conclusions in axisymmetric body flow control scenarios remain limited.

[0004] Some studies have attempted to introduce genetic algorithms into the active control of mechanical shapes, applying them to the active control of a D-shaped cylindrical flow field. By searching for the optimal control parameters under maximum drag reduction conditions, the feasibility of genetic algorithms in optimizing flow control parameters was verified. However, this type of optimization typically focuses solely on drag reduction rate, failing to consider blowing energy consumption or other control performance indicators comprehensively. This can easily lead to low control efficiency in engineering applications.

[0005] To further improve control efficiency, existing technologies have begun to incorporate energy factors into the objective function. For example, in the study of two-dimensional curved slope flow separation control, some researchers have used a genetic algorithm to optimize the airflow distribution of multiple jets and linearly combined the maximum drag reduction rate with the airflow energy consumption in the objective function through weighted coefficients, constructing an index in the form of J = J1 + φJ2. This achieved an energy consumption reduction of approximately 30% while maintaining a similar drag reduction effect. However, this method is essentially still a weighted single-objective optimization, and its results are highly sensitive to the selection of the weighting coefficient φ. Furthermore, it cannot provide a Pareto optimal solution set that reflects the trade-off between different drag reduction performances and energy consumption, and it still has shortcomings in terms of multi-objective collaborative optimization and flexibility in scheme selection.

[0006] For active control of the tail of axisymmetric bodies, previous studies have shown that arranging six steady tangential jet outlets circumferentially at the tail of the model can significantly suppress tail flow separation and achieve a drag reduction effect of approximately 24%. However, these methods generally employ steady open-loop control strategies, with control parameters mostly set manually based on experience. They lack systematic parameter optimization and adaptive adjustment mechanisms, making it difficult to balance drag reduction performance and energy utilization efficiency under different operating conditions.

[0007] In summary, existing technologies for active control of jet blowing at the tail of axisymmetric bodies lack a multi-objective optimization method that can simultaneously consider drag reduction and energy consumption indicators and provide a Pareto optimal solution set. They also lack specialized control structures and parameter design ideas for axisymmetric models, which need further improvement and refinement. Summary of the Invention

[0008] The purpose of this invention is to provide a multi-objective intelligent control system and method for the hydrodynamic performance of underwater vehicles, so as to solve the problems existing in the prior art.

[0009] To achieve the above objectives, this invention provides a multi-objective intelligent control system for the hydrodynamic performance of underwater vehicles, comprising: An axisymmetric body is provided with multiple independently arranged air jet outlets at its tail end. The air jet outlets are used to inject jets into the flow region around the tail end to change the tail separation characteristics and pressure distribution. A sensing module is installed at the tail of the axisymmetric body to collect the flow information of the axisymmetric body in real time. The flow information includes the tail surface pressure coefficient and the total blowing momentum coefficient. A control logic module, which is connected to the sensing module, is used to optimize the control parameters of each blowing jet outlet based on flow information and a preset optimization algorithm. The control parameters include the air supply pressure ratio, duty cycle, excitation frequency, and phase difference. The execution module is connected to the control logic module and each air jet outlet. It is used to adjust the air jet blowing state of each air jet outlet according to the optimized control parameters, so as to realize the real-time control of the flow state at the tail of the axisymmetric body.

[0010] Optionally, the blowing jet outlet includes an airflow inlet, a blowing chamber, and a jet outlet arranged sequentially.

[0011] Optionally, the sensing module includes a velocity measuring device and multiple pressure measuring holes, wherein each pressure measuring hole is disposed on the tail surface of the axisymmetric body to obtain the tail surface pressure coefficient, and the velocity measuring device is arranged at the jet outlet to obtain the total blowing momentum coefficient.

[0012] Optionally, the speed measuring device employs a hot-wire probe.

[0013] Optionally, the execution module includes a gas source and a control valve group, wherein the gas source is connected to the airflow inlet, and the control valve group is disposed on the connecting pipeline between the gas source and the airflow inlet.

[0014] Optionally, the air source is an air compressor.

[0015] Optionally, the control valve assembly includes an electro-proportional valve and a solenoid valve.

[0016] On the other hand, to achieve the above objectives, the present invention provides a multi-objective intelligent control method for the hydrodynamic performance of an underwater vehicle, applied to the aforementioned multi-objective intelligent control system for the hydrodynamic performance of an underwater vehicle, comprising: Step S1: Initialize the control logic module to generate an initial population based on the initial control parameters; send the initial control parameters corresponding to each individual in the initial population to the execution unit, which then applies them to the axisymmetric body. Step S2: The flow information of the axisymmetric body is acquired in real time through the sensing module, and the flow information is fed back to the control logic module as a performance evaluation index. The objective function is to improve the tail pressure recovery capability and reduce the blowing energy consumption, and a multi-objective optimization problem is constructed. Step S3: Iterate and update each individual in the initial population multiple times based on the non-dominated sorting genetic algorithm. When the preset termination condition is met, stop the iteration and output the optimized control parameters. Step S4: Adjust the jet blowing state of each blowing jet outlet according to the optimized control parameters; realize real-time control of the flow state at the tail of the axisymmetric body.

[0017] Optionally, in step S1, the control logic module uses the Latin hypercube sampling method to generate an initial population within a preset control parameter range.

[0018] Optionally, step S3 specifically includes: Step S31: Perform non-dominated sorting and crowding distance calculation on the previous generation population based on the performance evaluation index corresponding to each individual in the previous generation population, and generate the offspring population based on the calculation results through selection, crossover and mutation operations; Step S32: The control parameters corresponding to each individual in the offspring population are applied to the axisymmetric body through the execution unit, and the corresponding performance evaluation index is obtained by the sensing module. Step S33: Merge the previous generation population with the offspring population to form an intermediate population, and perform a fast non-dominated sort on the intermediate population. Combine the crowding distance criterion to select elites, and the selected individuals enter the next generation population. Repeat steps S31 to S33 for iterative updates. When the preset termination condition is met, stop the iteration and output the optimized control parameters to achieve adaptive iterative optimization of the control parameters.

[0019] The technical effects of this invention are as follows: 1. This invention addresses the problem of tail separation and increased flow resistance in axisymmetric models during flow. It proposes a jet-based active flow control structure based on tangential pulsed jets. This structure arranges four independent, controllable jets circumferentially along the tail of the model. The jet velocity, duty cycle, excitation frequency, and phase difference of each jet can be independently adjusted to achieve precise control over the tail pressure distribution and the separated flow pattern.

[0020] 2. This invention introduces the Non-dominated Sorting Genetic Algorithm II (NSGA-II) into the optimization design process of tangential blowing control parameters for axisymmetric body jets. With improving tail pressure recovery capability and reducing blowing energy consumption as dual objective functions, it uses non-dominated sorting and congestion distance criteria for global optimization to obtain the Pareto optimal solution set, thereby achieving synergistic optimization of active flow control performance and energy utilization efficiency. This provides a general method for the system design of efficient drag reduction control parameters.

[0021] 3. The jet tangential blowing control structure and its parameter optimization strategy proposed in this invention can be applied to the drag reduction and tail pressure recovery design of axisymmetric vehicles such as submarines, unmanned underwater vehicles (UUVs), torpedoes, and missiles; it is also applicable to active flow control optimization problems in the fields of aerodynamics and hydrodynamics, including but not limited to the drag resistance and pressure regulation design of high-speed carriers, shape-sensitive equipment, experimental models, etc. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a schematic diagram of the axisymmetric body artificial intelligence control system structure in an embodiment of the present invention; Figure 2This is a schematic diagram of the axisymmetric body jet blowing system and its air path control structure in an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the jet structure in an embodiment of the present invention. Detailed Implementation

[0025] Various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as a limitation of the present invention, but rather as a more detailed description of certain aspects, features, and embodiments of the present invention.

[0026] It should be understood that the terminology used in this invention is merely for describing particular embodiments and is not intended to limit the invention. Furthermore, with respect to numerical ranges in this invention, it should be understood that each intermediate value between the upper and lower limits of the range is also specifically disclosed. Every smaller range between any stated value or intermediate value within a stated range, and any other stated value or intermediate value within said range, is also included in this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0027] Various modifications and variations can be made to the specific embodiments described in this specification without departing from the scope or spirit of the invention, as will be apparent to those skilled in the art. Other embodiments derived from this specification will also be obvious to those skilled in the art. This application specification and embodiments are merely exemplary.

[0028] The terms “include,” “including,” “have,” “contain,” etc., used in this article are all open-ended terms, meaning that they include but are not limited to.

[0029] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0030] Large-scale separation and a low-pressure base region are common in the flow around the tail of axisymmetric bodies, resulting in a significant proportion of pressure drag. Existing passive control methods or open-loop blowing control strategies relying on manually set parameters struggle to simultaneously balance drag reduction performance and control energy consumption under complex operating conditions, and lack adaptive adjustment capabilities to changes in flow state. Furthermore, existing AI-based flow control methods mostly focus on single-objective optimization or non-axisymmetric body shapes, and have not yet developed an active control scheme for axisymmetric bodies that can combine multi-jet tangential blowing structures with multi-objective genetic algorithms (such as NSGA-II) to achieve adaptive blowing parameters and multi-objective collaborative optimization under experimental conditions. Therefore, it is difficult to systematically obtain the Pareto optimal trade-off between drag reduction and energy efficiency while ensuring system stability.

[0031] Based on the above shortcomings, this embodiment provides an axisymmetric body artificial intelligence active flow control system that combines a multi-jet blowing actuator with a multi-objective genetic algorithm optimization strategy. Through online feedback, it realizes automatic optimization of control parameters such as air supply pressure ratio PR, duty cycle dc, excitation frequency fe, and phase difference Δφ, thereby achieving synergistic optimization of resistance reduction and blowing energy consumption reduction, and improving the adaptability of the control system to different working conditions and overall energy efficiency.

[0032] like Figure 1 - Figure 3 As shown, the system provided in this embodiment includes a control target, an execution unit, a sensing unit, and a control logic module. These units are interconnected via signal connections and pneumatic connections to form a closed-loop control system.

[0033] 1. Control target: The control target is an axisymmetric body structure with at least one active blowing jet outlet (jet exciter) at its tail end, which is used to inject jet into the flow region around the tail end to change the tail separation characteristics and pressure distribution, thereby providing an intervention channel for drag reduction.

[0034] 2. Execution Unit: The execution unit is connected to the jet outlet of the controlled target and is used to adjust the jet blowing state according to the control commands output by the control logic module. The control commands include at least the supply pressure ratio PR, duty cycle dc, and excitation frequency f. e The phase difference Δφ is used to achieve active control of the jet. The electro-proportional valve directly controls the supply pressure ratio (the ratio of supply pressure to atmospheric pressure), indirectly affecting the amplitude of the blowing speed. The duty cycle and excitation frequency can also affect the blowing speed.

[0035] 3. Sensing Unit: The sensing unit is located on the tail surface of the axisymmetric body and is used to collect key parameters reflecting the flow control effect and energy cost. These parameters include at least the tail surface pressure coefficient and the total blowing momentum coefficient, and are used to construct the system's objective function.

[0036] 4. Control Logic Module: The control logic module is connected to both the execution unit and the sensing unit via signals. It receives flow information output from the sensing unit and optimizes the control parameters in real time or periodically based on a preset optimization algorithm. The optimized control parameters are then sent from the control logic module to the execution unit, enabling the system to operate in a closed loop.

[0037] Based on the above configuration, this embodiment achieves real-time control of the flow state at the tail of the axisymmetric body by actively adjusting the tail jet and combining sensor feedback with optimized calculations of the control logic module, thereby improving the tail pressure distribution and reducing the overall hydrodynamic resistance, thus achieving the basic objective of this embodiment.

[0038] like Figure 1As shown, the control target is an axisymmetric body with a hemispherical rear body structure. Multiple jets are arranged along the circumference of the rear body, which are denoted as S1, S2, S3 and S4 respectively. The outlets of each jet are evenly distributed in the circumferential direction.

[0039] like Figure 2 As shown, each jet outlet sequentially includes an airflow inlet, a blowing chamber, and a jet outlet. The airflow inlet is connected to the air path system of the actuator, and the blowing chamber is used to buffer and evenly distribute the incoming airflow, so that the jet velocity at the jet outlet is uniform along the circumferential direction, and the jet velocity direction is tangential to the surface.

[0040] Through the above structural design, the airflow at the jet outlet can adhere to the rear surface of the axisymmetric body and couple with the tail shear layer and the wake region, thereby effectively adjusting the tail separation morphology and base pressure distribution, and achieving active control of the tail flow structure.

[0041] like Figure 3 As shown, the execution unit consists of a gas supply system and a valve control system. The gas supply system is preferably an air compressor, used to continuously provide a stable gas supply to each jet. The valve control system includes an electro-proportional valve and a solenoid valve. The electro-proportional valve is used to adjust the gas supply pressure ratio PR entering each jet, thereby indirectly controlling the jet velocity amplitude. The solenoid valve is used to control the opening and closing process of the jet to achieve control over the jet duty cycle dc and excitation frequency f. e And the adjustment of parameters such as phase difference Δφ.

[0042] Each jet S1 to S4 corresponds to an independently configured air path branch, enabling independent control of the jets without interference. Through coordinated adjustment of the electro-proportional valve and the solenoid valve, the actuator can separately set the air supply pressure ratio PR, duty cycle dc, and excitation frequency f for each jet. e Control parameters such as phase difference Δφ are used to form an asymmetric and asynchronous multi-jet excitation mode, thereby improving the ability to control the flow structure at the tail of the axisymmetric body.

[0043] The sensing unit includes multiple sets of pressure measuring holes disposed on the rear surface of the axisymmetric body, and a velocity measuring device (hot wire probe) arranged at the jet outlet. The static pressure C on the tail surface is measured through the pressure measuring holes. p The expression for the pressure coefficient:

[0044] In the formula, P is the local static pressure (Pa) on the rear surface of the axisymmetric body; p ref The static pressure (Pa) is the free flow pressure upstream of the axisymmetric body.

[0045] Based on the definition of the total pressure drag coefficient of the model, the pressure drag coefficient of the rear body surface is calculated, and the expression is as follows:

[0046] In the formula, C is the coefficient of pressure drag at the rear of the body; p n is the pressure coefficient on the surface of the rear body; x Let n be the component of the unit normal vector n on the rear body surface in the x-direction; A is the area differential unit; A is the cross-sectional area (m²) of the axisymmetric solid model. 2 ).

[0047] The jet velocity is obtained by placing hot wire probes at each jet outlet, and the flow coefficients corresponding to the four jet outlets are calculated accordingly. Finally, the total flow coefficient is obtained by summing the four jet flow coefficients. The calculation formula is as follows:

[0048] In the formula, Let be the exit velocity of the i-th jet (m / s); The outlet area of ​​the i-th jet (m²) 2 ).

[0049] The control logic module is constructed based on the non-dominated sorting genetic algorithm NSGA-II and is used for multi-objective optimization of the blowing control parameters of multiple jets. The control logic module uses the blowing control parameters of each jet as optimization variables and incorporates the rear body pressure drag coefficient. With blowing flow coefficient As the objective function, a multi-objective optimization problem is constructed. Specifically, the multi-objective optimization problem can be constructed as follows:

[0050] in The control parameter matrix represents the supply pressure ratio, duty cycle, excitation frequency, and phase difference of the four jets, for example... . These are the domains of the control parameters. Control objective. and There is no specific function form; exploratory optimization is performed using NSGA-II.

[0051] During the initialization phase, the control logic module first uses the Latin hypercube sampling method to generate an initial population within the preset control parameter range. Each initial individual corresponds to a set of air blowing control parameters to ensure the uniformity and representativeness of the initial sample distribution in the parameter space.

[0052] Latin Hypercube Sampling (LHS) is a statistically significant experimental design method used to efficiently and uniformly extract sample points within a multidimensional parameter space. This method divides the range of each variable into equally probable intervals and randomly selects samples from each interval, ensuring that each variable uniformly covers its domain across all samples. This improves the efficiency of parameter space exploration and reduces the number of trials.

[0053] Subsequently, the control logic module sends the blowing control parameters corresponding to each individual in the initial population to the execution unit, which then applies them to the controlled axisymmetric model. Under the influence of these control parameters, the sensing unit collects real-time data on the tail pressure and blowing momentum, and calculates the corresponding... With C m The value is used as a performance evaluation indicator and fed back to the control logic module.

[0054] After obtaining the objective function value of the first generation population, the control logic module performs non-dominated sorting and crowding distance calculation on the first generation population based on the NSGA-II algorithm framework, performs tournament selection, and generates offspring population through crossover and mutation operations.

[0055] Specifically, the non-dominated ranking first stratifies individuals based on multi-objective dominance relationships, classifying individuals not dominated by any other individuals as the first non-dominated frontier, and then eliminating them layer by layer to obtain the second frontier, the third frontier, and so on, thereby assigning a corresponding non-dominated level to each individual.

[0056] Based on this, the crowding distance is used to characterize the local density distribution of individuals within the same non-dominated front in the target space. It is calculated by sorting each objective function separately and accumulating the normalized objective difference between adjacent individuals, so that individuals with sparser distribution (greater crowding) are preferentially retained in the selection, thereby maintaining the diversity of the Pareto solution set.

[0057] Subsequently, the control logic module constructs a mating pool using a binary tournament selection strategy based on "non-dominance level priority, followed by crowding distance". That is, it prioritizes individuals with lower non-dominance levels; when non-dominance levels are the same, it prioritizes individuals with greater crowding distance.

[0058] After establishing the mating pool, the crossover operator recombines the control parameters of paired individuals in the mating pool with a probability of 0.9, thereby generating new parameter combinations while retaining advantageous genes; the mutation operator randomly perturbs the individual parameters with a small probability (1 / 7) to enhance the ability to explore the parameter space and avoid the algorithm from getting trapped in local optima.

[0059] The control parameters corresponding to the offspring population generated by the above steps are also applied to the controlled object by the execution unit, and the corresponding performance index data are obtained by the sensing unit.

[0060] The control logic module merges the parent and offspring populations to form an intermediate population, and performs rapid non-dominated sorting on this intermediate population, combining it with a crowding distance criterion for elite selection. The specific process of elite selection involves filling the next generation of the population in descending order of non-dominated level. When the addition of a certain front would cause the population size to exceed a preset size, individuals within that front are sorted in descending order of crowding distance, prioritizing the retention of sparser individuals to ensure the coverage of the Pareto front.

[0061] Through the continuous execution and iterative updates of the above "sorting-selection-crossover-mutation-elite retention" steps, the control logic module continuously adjusts and optimizes the blowing control parameters until the preset termination conditions are met. The final set of non-dominated solutions constitutes the Pareto optimal control parameter set, realizing adaptive iterative optimization of the blowing control parameters between pressure recovery performance and blowing energy consumption.

[0062] The hypervolume (HV) index is used as the convergence criterion for the multi-objective optimization problem. When the HV increment is less than 2% for several consecutive generations, the Pareto front is considered to have stabilized and the optimization process is terminated.

[0063] For a given generation of population, its hypervolume index HV is calculated using the following formula:

[0064] in and These are the control targets under no-control conditions, and are used as reference values. and These are the control objectives of the i-th individual in the current population. n represents the total number of individuals in the current population.

[0065] Through the coordinated operation of the aforementioned control logic module, execution unit, and sensing unit, this embodiment can achieve comprehensive optimization between tail pressure recovery and blowing energy consumption of the axisymmetric body while ensuring stable system operation, thereby achieving the technical effect of reducing axisymmetric body resistance and lowering control energy consumption.

[0066] In summary, this embodiment addresses the problem of tail separation and increased flow resistance in axisymmetric models during flow, proposing a jet-type active flow control structure based on tangential pulse jets. The axisymmetric body artificial intelligence active control system of this embodiment uses multiple jet exciters to actively control the tail flow of the axisymmetric body. The air source is connected to each jet exciter through multiple independent flow control branches, allowing for different air supply pressure ratios (PR), duty cycles (dc), and excitation frequencies (f) at the outlets of different jet exciters. e The phase difference Δφ is adjusted independently.

[0067] By jointly controlling multiple jet actuators, the flow structure in different regions of the axisymmetric body's tail can be adjusted simultaneously, suppressing unfavorable separation phenomena and improving the pressure recovery level in the base region, thereby effectively reducing the aerodynamic drag of the axisymmetric body. At the same time, because the control parameters of each jet actuator are coordinated and configured through multi-objective optimization, significant drag reduction effects can be achieved while effectively controlling the energy consumption of the blowing gas, resulting in a high net energy gain from drag reduction.

[0068] The above description is merely a preferred embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A multi-objective intelligent control system for the hydrodynamic performance of an underwater vehicle, characterized in that, include: An axisymmetric body is provided with multiple independently arranged air jet outlets at its tail end. The air jet outlets are used to inject jets into the flow region around the tail end to change the tail separation characteristics and pressure distribution. A sensing module is installed at the tail of the axisymmetric body to collect the flow information of the axisymmetric body in real time. The flow information includes the tail surface pressure coefficient and the total blowing momentum coefficient. A control logic module, which is connected to the sensing module, is used to optimize the control parameters of each blowing jet outlet based on flow information and a preset optimization algorithm. The control parameters include the air supply pressure ratio, duty cycle, excitation frequency, and phase difference. The execution module is connected to the control logic module and each air jet outlet. It is used to adjust the air jet blowing state of each air jet outlet according to the optimized control parameters, so as to realize the real-time control of the flow state at the tail of the axisymmetric body.

2. The multi-objective intelligent control system for the hydrodynamic performance of an underwater vehicle according to claim 1, characterized in that, The air jet outlet includes an airflow inlet, an air blowing chamber, and a jet outlet arranged sequentially.

3. The multi-objective intelligent control system for the hydrodynamic performance of an underwater vehicle according to claim 2, characterized in that, The sensing module includes a velocity measuring device and multiple pressure measuring holes. Each pressure measuring hole is located on the tail surface of the axisymmetric body to obtain the tail surface pressure coefficient. The velocity measuring device is located at the jet outlet to obtain the total blowing momentum coefficient.

4. The multi-objective intelligent control system for the hydrodynamic performance of an underwater vehicle according to claim 3, characterized in that, The speed measuring device uses a hot-wire probe.

5. The multi-objective intelligent control system for the hydrodynamic performance of an underwater vehicle according to claim 2, characterized in that, The execution module includes a gas source and a control valve group. The gas source is connected to the airflow inlet, and the control valve group is disposed on the connecting pipeline between the gas source and the airflow inlet.

6. The multi-objective intelligent control system for the hydrodynamic performance of an underwater vehicle according to claim 5, characterized in that, The air source is an air compressor.

7. A multi-objective intelligent control system for the hydrodynamic performance of an underwater vehicle according to claim 5, characterized in that, The control valve assembly includes an electro-proportional valve and a solenoid valve.

8. A multi-objective intelligent control method for the hydrodynamic performance of an underwater vehicle, applied to a multi-objective intelligent control system for the hydrodynamic performance of an underwater vehicle as described in any one of claims 1-7, characterized in that, include: Step S1: Initialize the control logic module to generate an initial population based on the initial control parameters; The initial control parameters corresponding to each individual in the initial population are sent to the execution unit, which then applies them to the axisymmetric body. Step S2: The flow information of the axisymmetric body is acquired in real time through the sensing module, and the flow information is fed back to the control logic module as a performance evaluation index. The objective function is to improve the tail pressure recovery capability and reduce the blowing energy consumption, and a multi-objective optimization problem is constructed. Step S3: Iterate and update each individual in the initial population multiple times based on the non-dominated sorting genetic algorithm. When the preset termination condition is met, stop the iteration and output the optimized control parameters. Step S4: Adjust the jet blowing state of each blowing jet outlet according to the optimized control parameters; realize real-time control of the flow state at the tail of the axisymmetric body.

9. The multi-objective intelligent control method for the hydrodynamic performance of an underwater vehicle according to claim 8, characterized in that, In step S1, the control logic module uses the Latin hypercube sampling method to generate an initial population within a preset control parameter range.

10. The multi-objective intelligent control method for the hydrodynamic performance of an underwater vehicle according to claim 8, characterized in that, Step S3 specifically includes: Step S31: Perform non-dominated sorting and crowding distance calculation on the previous generation population based on the performance evaluation index corresponding to each individual in the previous generation population, and generate the offspring population based on the calculation results through selection, crossover and mutation operations; Step S32: The control parameters corresponding to each individual in the offspring population are applied to the axisymmetric body through the execution unit, and the corresponding performance evaluation index is obtained by the sensing module. Step S33: Merge the previous generation population with the offspring population to form an intermediate population, and perform a fast non-dominated sort on the intermediate population. Combine the crowding distance criterion to select elites, and the selected individuals enter the next generation population. Repeat steps S31 to S33 for iterative updates. When the preset termination condition is met, stop the iteration and output the optimized control parameters to achieve adaptive iterative optimization of the control parameters.

Images