Abstract:Hybrid energy storage has superior power controllability and is often used as a power buffer in island microgrids to compensate for the power output of distributed powers and loads. To address power response issues， a power control method for hybrid energy storage is proposed， and an adaptive power allocation strategy with multi-objective coordination for subunits is developed. Firstly， a master-slave parallel power control scheme is developed by analyzing the balance mechanism of bus power supply and demand. Secondly， the scheme achieved autonomous power distribution of subunits by utilizing the zeroing characteristic of the integrator. Based on this， a hierarchical management system for SOC and a power coordination method with multi-objective optimization of subunits are designed by analyzing the maximum power response of subunits under different SOCs. The simulation results indicate that the proposed method yields a power distribution effect that aligns with the power response characteristics of subunits. As the SOCscn offset increases， the proposed strategy provides a stronger optimization ability for SOCscn， as the performance can be improved by 10.1% when the initial SOCscn is 97%， which ensures that the power output capacity of the supercapacitor as a voltage source is guaranteed. The proposed strategy reduces the output strength of the battery under the normal condition of the supercapacitor as the maximum power depth of the battery is reduced by 46.7%. This improvement prolongs the service life of the battery.

Abstract:In view of the fact that the hybrid energy storage systems of optical storage and DC microgrids are susceptible to intermittent energy input， random load perturbations， and power flow switching， which may cause the problems of bus voltage fluctuations and system power instability， a complementary sliding mode control strategy based on a finite-time observer is proposed. First， according to the high-low frequency characteristics of the hybrid energy storage elements， the current equivalent distribution of the system differential power is carried out. Then， a finite-time dilated state observer is designed to observe the total disturbance to the system， and the disturbance observations are fed into the complementary sliding-mode controller as feedforward terms to compensate for the system disturbance， ensuring that the system state reaches convergence in finite time， improving the fast and anti-disturbance performance of the system， and proving the stability of the control system according to Lyapunov theory. Finally， based on the MATLAB simulation platform， the simulation results show that the proposed control has faster response speed and better immunity to disturbances than the traditional control strategy.

Abstract:To improve the impact of wind power output uncertainty and multi-energy transmission differences on the regulation process in integrated natural gas and electric power systems， a hybrid scale regulation method based on an improved wavelet fusion algorithm is proposed. Firstly， an interval mathematical approach is used to represent the wind power uncertainty in the system and give the wind power treatment strategy. Secondly， considering the differences of different energy transmission characteristics， an improved wavelet fusion algorithm is proposed to first decompose the sensor signal data in the power network with several different wavelet bases at multiple scales， and then implement a weighted data fusion algorithm on the hybrid scale for the signals decomposed using the same wavelet bases in the natural gas system signal data， and obtain the fused signals after the inverse conversion of different wavelet bases. Finally， based on the simulation model built， the modulation effects of different modulation methods are compared and analyzed. The results show that the proposed method outperforms the DMPC rolling optimization regulation and improves the system stability while improving the system operation economy.

Abstract:DC transmission lines usually take large inductance and DC filters as the boundary. When a fault traveling wave undergoes refraction and reflection at the boundary， its traveling wave energy will also be absorbed due to the inductance and DC filters， resulting in the difference of high-frequency traveling wave energy within and outside the region. When the positive direction fault occurs， the energy of the high-frequency forward traveling wave is less than that of the high-frequency backward traveling wave. When the opposite direction fault occurs， the energy of the high-frequency forward traveling wave is greater than that of the high-frequency backward traveling wave. Based on this feature， a pilot directional protection based on the difference of high-frequency energy of forward and backward traveling waves is proposed that uses the wavelet packet transform algorithm to extract the high-frequency energy of the traveling wave， and uses the difference of the high-frequency energy of the forward and backward traveling waves to distinguish the faults inside and outside the area. The simulation results show that the protection can quickly and accurately identify faults， and is not affected by fault type， transition resistance， fault location distance， and noises. It has certain practicability.

Abstract:Transmission conductors icing is a critical threat to the safety of power grid operation. Accurate prediction of the ice shape and growth trend of conductors icing can effectively prevent the occurrence of secondary disasters. Existing models for predicting the ice growth of transmission conductors seldom consider the effect of ice eccentricity torsion on the ice shape of transmission conductors. Therefore， the dynamic torsional icing prediction model of transmission conductor is established by considering the coupling torsional properties of transmission wire， such as ice eccentricity and wind speed. Based on the prediction model， the ice shape on the conductor's surface is simulated and compared with experimental data in the literature， and the validity of the model is verified. The effects of conductor inclination and temperature， wind speed， and MVD （median diameter of raindrops） on ice shape and torsional speed are further discussed with predictive models. It is found that the stratification of the ice thickness on the windward side of the conductor is gradually obvious with the increase of the tilt angle. Compared with the horizontal arrangement （zero inclination） conductor， when the inclination angle is 60°， the ice area of the wire increases by 18.26% （ice covering 125 minutes） and 26.30% （ice covering 245 minutes）， and the torsion angle increases by 10.4° （ice covering 125 minutes） and 16.3° （ice covering 245 minutes）， respectively， and the ice shape is "ravine".

Abstract:Accurate power load forecasting is crucial to the safe and economic operation of modern power systems. Power load forecasting can be expressed as a multivariable time series forecasting problem with certain potential spatial dependence. However， most existing power load forecasting work fails to explore this spatial dependency relationship. Based on this， this paper proposes a short-term power load forecasting method based on the spatiotemporal graph attention network. A spatiotemporal graph-based attention network module is proposed， which uses a graph attention layer to adaptively capture potential spatial dependencies between users. At the same time， a gated convolutional attention layer is used to adaptively fit the electricity consumption of each user in the time dimension to improve the prediction accuracy of the network. Actual data experiments show that the overall prediction accuracy of the model proposed is significantly improved， especially in alleviating the problem of deteriorating long-range prediction accuracy to a certain extent， verifying the effectiveness and feasibility of the proposed method.

Abstract:Photovoltaic power prediction plays an important role in the scheduling and operation of modern power systems. Aiming at the variability and complexity of photovoltaic power generation， a short-term PV power prediction method based on new similar day selection and northern Goshawk optimization （NGO） to optimize bidirectional gated recurrent unit （BiGRU） is proposed. The main meteorological factors are selected with the Spearman correlation coefficient， and the original PV power and maximum meteorological factor are decomposed into a series of sub-signals by variational mode decomposition （VMD）. Then， according to the construction of new evaluation indicators， the data set of similar days is screened out， a group of BiGRU is used to establish a deep learning model with similar day signals as network input， and NGO is used to effectively optimize the hyperparameters of each BiGRU network. Finally， the predicted value of PV power is obtained by synthesizing the predicted results of each sub-signal. Simulation results show that the proposed hybrid deep learning method is superior to other methods in terms of prediction accuracy and computational efficiency.

Abstract:An improved model prediction current control is proposed to address the disadvantages of the model predictive common-mode voltage suppression method for double-three-phase permanent magnet synchronous motors， including the size of the optimization calculation， high switching frequency， and poor steady-state performance. Firstly， the six-phase two-level inverter is improved for reducing the common-mode voltage of zero-vectors. Secondly， the virtual voltage vector is constructed by choosing small common-mode voltage vectors， which simplifies the value function and reduces the common-mode voltage and current harmonics. Then， the optimal voltage vector is directly selected by calculating the reference voltage vector to reduce the number of optimization-seeking times， and the duty cycle control is introduced to enhance the motor control accuracy and steady-state performance. Finally， the traditional model predictive current control， RCMV （reduced common mode voltage）-1， RCMV-2， and the proposed control method are simulated and compared. The results show that the proposed control method reduces the torque pulsations and harmonic currents while decreases the common-mode voltage， and the switching frequency is significantly lower than that of the RCMV-2 method. In addition， the execution time of the optimization-seeking code is reduced by about 91% and 65% compared with RCMV-1 and RCMV-2， respectively， which reduces the computational amount.

Abstract:Aiming at resistivity inversion and uncertainty quantification in electrical impedance tomography （EIT）， an uncertainty analysis method is proposed based on Bayesian theory. Firstly， the Back Propagation （BP） neural network model is used as a substitute model for the forward problem， the results with high calculation accuracy are obtained， and the calculation efficiency is greatly improved. Then， the Differential Evolution Adaptive Metropolis （DREAM_ZS） sampling algorithm based on Bayesian theory is used for the resistivity reconstruction， and different excitation modes and prior distributions are compared and analyzed. The inversion results of the four-layer concentric circle model simulating the head show that the DREAM_ZS sampling algorithm can accurately identify the four parameters， and the inversion effect of the relative excitation mode is the best. The uncertainty of the four parameters is different. The scalp resistivity has the minimum uncertainty and the strongest sensitivity， and then the skull， the brain， and the cerebrospinal fluid show the maximum uncertainty. Furthermore， the circular model with high-dimensional parameters is simulated， and the relative excitation mode is adopted. DREAM_ZS sampling algorithm can accurately invert the parameters of the two-dimensional circular model. When the prior distribution of the parameters is normal distribution， compared with the uniform distribution， the uncertainty of the inversion result is less， and the recognition effect of the algorithm is better.

Abstract:A compact fully integrated power amplifier for mobile devices based on the SiGe BiCMOS process is proposed. The design uses a cascode driver stage cascaded with a common emitter power stage to increase the power gain of the amplifier， integrates a CMOS power supply on-chip to provide bias current， and uses distributed ballast resistors and a bias circuit with thermal negative feedback effect to compensate the junction temperature so as to prevent amplifier failure at high-temperature operation. A compact circuit-level thermal coupling model is used to simulate and verify the proposed thermal stabilization measures. The post-simulation results show that PA has an output gain of 32.5 dB in the working range of 2.4~2.5 GHz under a 3.3 V power supply， S11&S22<-10 dB， and the output power at the 1 dB compression point is 25.4 dBm. The maximum junction temperature is less than 65 ℃ （saturation）. The chip area is only 1.25×0.76 mm2. The test results show that in the working environment of -45~ 85 ℃， the amplifier can work well in the application where the gain requirement is 26.5~ 32.9 dB. The output power at the 1 dB compression point is 24.3 dBm. Using a 20 MHz 64-QAM OFDM modulation signal test， the output power of DEVM reaching -30 dB is 18.1 dBm.

Abstract:A flotation process concentrate grade prediction method based on the Improved Sparrow Search Algorithm （ISSA） optimized Hybrid Kernel Least Squares Support Vector Machine （HKLSSVM）， a flotation process concentrate grade prediction method is proposed to address the issues of delayed variables， coupling characteristics， and limited modeling sample size in the flotation process， which make it difficult to accurately predict the concentrate grade. Firstly， collect data from the flotation site current carrying X-ray fluorescence grade analyzer as modeling variables and preprocess them to establish a prediction model based on the Least Squares Vector Machine. On this basis， a new mixed kernel function is constructed to map the input space to the high-dimensional feature space. Then， an Improved Sparrow Search Algorithm is introduced to optimize the model parameters， and an ISSA-HKLSSVM method is proposed to achieve concentrate grade prediction. Finally， a flotation concentrate grade prediction system based on LabVIEW is developed to verify the proposed method in practice. The experimental results show that the proposed method has a better fitting ability for small sample modeling in the flotation process. It can improve prediction accuracy compared to existing methods， and can achieve accurate online prediction of concentrate grade， providing real-time and reliable concentrate grade feedback information for intelligent control of the flotation process.

Abstract:The alpine and snowy environment causes massive snow accumulation in the bogie regions， which seriously endangers the running safety of high-speed trains. To analyze the icing characteristics of bogies under the alpine and snowy weather， an experimental investigation on the icing of high-speed train bogies is conducted in the Icing-snowing Wind Tunnel （ISWT） of Central South University （CSU）， using a scaled model containing a simplified train body and a motor bogie. A water-spray system is used to simulate the water-spraying phenomenon generated by the melt water around the brake calipers to reproduce the process of bogie icing due to water spraying from the wheels. The dynamic icing process and the overall icing distribution characteristics of the bogie regions， the ice mass weighting of each key region of the bogie and the icing rate model of the bogie are studied. The results show that the droplets thrown out by the wheels are spread to all regions of the bogie under the effect of the turbulent flow field， which generated icing at low temperatures and developed rapidly with time until the whole bogie region is covered with ice. For the bogie cabin， its rear end panel is seriously iced， and the icing mass accounts for 28% of the total icing mass of the bogie cabin； for the bogie， the frame and brake caliper regions have the highest ice distribution， accounting for 34% and 22% of the total icing mass of the bogie， respectively， while the ice distribution in the air spring， cross beam， and longitudinal beam is less. The bogie presents the characteristics of a large icing mass and a complex icing shape at the bottom. With the increase of icing time， the icing rate of each region is different， and the overall icing mass of the bogie cabin and bogie is a linear function of icing time. The bogie icing distribution characteristics and icing rate model obtained from the study can provide a reference basis for the rapid prediction of bogie icing mass within a certain operation time， which is of great significance to the safe operation of high-speed trains in alpine and snowy environments and bogie anti-icing/deicing.

Abstract:To investigate the effect of grilles in different directions on the ventilation performance of equipment cabins under high-speed trains， a wind tunnel test is conducted with a section of equipment cabin used in a high-speed train as the test object， and a numerical model for the wind tunnel test is constructed based on the Realizable k-ε turbulence model. The accuracy of the numerical model is verified by wind tunnel test results， and a set of meshing parameters for the equipment cabin apron grill region is obtained. Based on the numerical model and the meshing parameters verified by the wind tunnel test， a three-car grouping numerical model with the head car including the equipment cabin is established. The ventilation volume of the equipment cabin grille vents and the velocity distribution inside the equipment cabin at 300 km/h speed level are investigated with different directions of grilles （vertical and horizontal）. The results show that the total inlet air mass flow rate of the equipment cabin is basically the same for the two types of grilles， among which the air inlet mass flow rate of both sides are basically the same when the vertical grille is installed， but the air inlet mass flow rate of both sides differs when the horizontal grille is installed， and the mass flow rate of the y-side decreases by 6.59% and that of the y+ side increases by 6.79% compared with the former. When the vertical grille is assembled， the airflow enters uniformly from the whole grille region， and the entrance direction is the same as the train operation direction， and the speed is lower. When the horizontal grille is assembled， the airflow concentrates on the right side of the grille region， and the entrance direction is opposite to the train operation direction， and the speed is higher. By comparing the mass flow rate and the airflow velocity in the equipment cabin， it is found that the ventilation performance is better when the horizontal grille is assembled.

Abstract:Four-wheel steering （4WS） vehicles are more flexible than front-wheel steering （FWS） vehicles. However， the steering resistance torque varies as the improvement of vehicle handling and stability with rear-wheel steering， which makes the power strategy of the original electric power steering system mismatch with 4WS vehicles and affects driving safety. Based on the 2-DOF linear bicycle model， the steering characteristics of front-wheel steering vehicles and four-wheel steering vehicles are compared and analyzed， and the modified control strategy of electric power steering is proposed. The simulation results show that the steering wheel torque of the 4WS vehicle with power correction is basically the same as that of the FWS vehicle under the condition of angle step， which not only ensures the handling portability of the four-wheel steering vehicle at low speed but also takes into account the handling stability at high speed.

Abstract:Intelligent vehicle path tracking control is faced with interferences such as vehicle parameter perturbation and scene variability， which affects the accuracy of path tracking， and even causes the control system instability. In this paper， a second-order super-twisting sliding mode robust control algorithm considering uncertainty is designed， the convergence of control system is proved， and a feedforward compensation controller is designed for interference problems to further improve the accuracy of the control system. Then， to validate the accuracy and robustness of the proposed robust control algorithm， several cases including double-lane change， sine path， as well as parameter perturbation scenes， are simulated under the Carsim-Matlab/Simulink co-simulation environment， and the traditional first-order sliding mode control is selected as a comparison . The results show that， the proposed super-twisting sliding mode algorithm has better robustness and tracking accuracy than the traditional sliding mode algorithm when considering the above interferences. Meanwhile， the proposed super-twisting algorithm can effectively reduce the steering wheel buffeting problem which exists in the traditional sliding mode algorithm. Finally， the large curvature scene travelling at low speed is tested on the real vehicle platform to verify the more accurate path-tracking effect of the designed super-twisting sliding mode algorithm.

Abstract:In order to address the issues of long design cycles， high design complexity， and excessive reliance on engineer experience in traditional lightweight design of tractor chassis， an intelligent lightweight design method is proposed. Firstly chassis performance data is obtained through Design of Experiments （DOE） combined simulation. Then， the BS-TabNet model is constructed based on the TabNet algorithm， Bayesian optimization algorithm， and SHapley Additive exPlanation （SHAP） theory. This model is used to learn from the chassis performance data and generate a surrogate model for the chassis. Finally， the Levy flight strategy is applied to improve the Sparrow Search Algorithm （SSA）， resulting in the Levy Sparrow Search Algorithm （LSSA）， which is used to solve the lightweight design task and find the optimal structural parameters for the chassis. Compared with traditional machine learning algorithms， the BS-TabNet model shows higher accuracy， stability， and interpretability. Its accuracy reaches around 0.98， stability is improved by over 50%， and it has stronger interpretability， addressing the poor performance of deep learning on tabular data. Compared with traditional swarm intelligence optimization algorithms， the LSSA algorithm can obtain better optimization results. While meeting other performance requirements， it achieves a 5.64% reduction in chassis weight. The intelligent lightweight design method combines artificial intelligence with chassis lightweight design， and it can save a significant amount of design time and improve design efficiency.

Abstract:Most of the existing parking spot detection solutions simply combine the target detection scheme and with manually designed post-processing modules， and there is a large amount of redundant information in the features extracted at each stage. Moreover， the manually designed post-processing modules are usually narrowly adapted and computationally intensive， which ultimately makes the parking space detection effect difficult to be practical. To address these problems， this paper introduces panoramic vision and combines the advantages of existing algorithms with the characteristics of surround-view images to design an end-to-end anchorless frame parking spot detection algorithm. The algorithm models the entry line orientation of parking spaces instead of considering two entry points separately， eliminating the process of parking space entry point matching and orientation judgment， and finally realizing realizes fully integrated parking space location， orientation， and occupancy detection. Considering the practicality， the network structure design is optimized in many aspects， such as the balance of speed and accuracy， positive and negative sample balance， and no post-processing. Finally， on the ps2.0 dataset， the AFPSD model proposed in this paper achieves 68.7% AP with a FPS（Frames Per Secend） of 88.7， which is 1.2% and 2.1% higher accuracy compared to the VPS-Net and DMPR-PS schemes， respectively. It can be seen that the one-stage end-to-end scheme designed in this paper can replace the three-stage scheme to achieve stable detection of parking slots on the surround-view image.

Abstract:The existing vehicle adaptive cruise control mostly focuses on a single trajectory tracking target， and the key to determin the practical application benefit of the control technology is the energy-saving performance of the system. In order to ensure the tracking error's convergence and fuel economy at the same time， this paper studies a design method forof vehicle energy-saving adaptive cruise controller. Firstly， a state feedback control law with time-varying feedback gains is designed with trajectory tracking as the goal. Secondly， taking the feedback gain as the optimization variable and the accumulation of the fuel consumption per unit displacement of the own vehicle at each sampling time in the entire prediction time domain as the cost function， an energy-saving control algorithm is designed based on the economic model predictive control method， and the selection range of the feedback gain to ensure that the tracking error converges at an exponential speed is proposed. Finally， the effectiveness of the controller design is verified by numerical simulations， and the effectiveness of the method proposed in this paper in reducing fuel consumption is quantitatively demonstrated.

Abstract:Aiming at the problems of poor temperature uniformity of lithium-ion batteries and high energy consumption of liquid cooling systems， a square lithium-ion battery was taken as the research object. Based on the verification of the battery cell model， a lithium-ion battery heat dissipation structure with a series-parallel symmetrical liquid cooling channel is designed. Five flow channel schemes are compared. Based on the fourth optimal scheme， the effects of the liquid cooling flow rate in the liquid cooling plate， the thickness combination of the aluminum plates， and the start-up time of the liquid cooling system on the battery heat dissipation and the energy consumption of the liquid cooling system are analyzed. The results show that compared with the channel shape S0 of scheme 1， the channel shape S3 of scheme 4 can reduce the maximum temperature difference of the cells in the battery module by 15%. In addition， the maximum temperature of the battery tends to decrease and then be gentle with the increase of the liquid cooling flow rate. Under the premise of ensuring the total mass of the liquid cooling system is unchanged， compared with the initial aluminum plate thickness combination h0， the adjusted aluminum plate thickness combination h4 can reduce the maximum temperature difference of the battery module by 12%. When the battery is discharged at 2.5C， the start-up time of the liquid cooling system is delayed to 563 s， which can ensure that the battery is within the best working temperature range and can save about 39% of the energy consumption cost of the liquid cooling system.

Abstract:Focusing on the applicability of the frequency domain method and time domain method in random vibration fatigue calculation， the accuracy and efficiency of the two methods are comprehensively studied with a certain type of battery pack as the research object. First， a finite element model of the battery pack is established and verified through the modal test. Secondly， the acceleration Power Spectral Density （PSD）of GB 38031―2020 is taken as the load spectrum in the frequency domain and converted into the time domain load using Fourier inverse transformation technology. Finally， the vibration acceleration， stress， and fatigue life of the battery pack are calculated based on the frequency domain method and the time domain method， respectively. The calculation accuracy and efficiency of both methods are compared and verified by the test. The results show that in regard of the total vibration level and stress RMS value of the battery pack， the calculated values of the frequency domain method and the time domain method are similar， and the relative error is less than 16%. For the peak value of acceleration and stress， the "3σ" calculation results of frequency domain method are much different from those of time domain method. With the adoption of the "4σ" or "5σ" principle， the calculation results of both methods are similar， and the relative error is less than 15%. In concern of vibration fatigue， the calculated life by the frequency domain method is about as 3 to 6 times long as that by the time domain method. The main reasons include Dirlik model and stress response PSD， in which the difference caused by the Dirlik model is less than 1.5 times. In terms of computational efficiency， the frequency domain method is about 134 times faster than the time domain method. The test data show that the time domain method is with higher accuracy and suitable for the accurate calculation of vibration fatigue at dangerous positions， while the frequency domain method is more efficient and suitable for the rapid prediction of structural high-risk positions.

Abstract:Aiming at the problem that of low machining accuracy of a certain type of small five-axis high-precision CNC machine tool is not high， the dynamics of its beam is studied. The natural frequency of the beam is studied by combining the hammering mode test experiment and finite element simulation calculation. Measure the vibration speed and vibration noise at different speeds of the spindle， and the vibration evaluation index equation is established through weighting processing to obtain the optimization target. The shape of the studied beam is optimized. The maximum error of the natural frequency obtained by the test and simulation is 5.76%， and the frequency range of 83~117 Hz is determined based on the spindle vibration， which is the low-order natural frequency range that needs to be avoided in the design of the beam. After shape optimization， the first-order natural frequency of the beam is 139 Hz， which is an increase of about 20.870% compared with the original beam， and the dynamics of the beam is improved.

Abstract:In this paper， a dynamic vibration absorber with negative stiffness （DVA-NS） is proposed to control the rest tremor of human arm in order to effectively reduce the dynamic response transmitted to the forearm. Firstly， the coupling dynamic model of human arm with DVA-NS is established. Then， the amplitude-frequency response expressions of each joint positions are derived， and the optimal design parameters of the DVA-NS are determined by using the sequential quadratic programming algorithm. On this basis， the above optimal design parameters are used to calculate the dynamic response curves under sinusoidal and random excitations， respectively. For ease of comparison， the dynamic responses of the coupling dynamic systems without the DVA-NS and with traditional linear dynamic vibration absorber （LDVA） are also presented， respectively. The calculation results show that the suppression effect on dynamic responses of different joints induced by the rest tremor of human arm by using the DVA-NS is significantly better than that of traditional LDVA.

Abstract:In order to study the effect of jet hole structure parameters on underwater jet drag reduction， this paper establishes a bionic fish model with tuna as the bionic object， and adds jet holes to the side of the bionic fish model by simulating shark gills. The effects of the velocity of the main stream field， the shape， height， position and aspect ratio of the jet hole on the surface drag reduction of the bionic fish are analyzed by numerical simulation. The response surface multi-objective parameter optimization is carried out on the structural parameters of the jet hole through Design-expert software， and the influence of the structural parameters of different jet holes on the surface drag reduction of the bionic fish is further analyzed. Finally， it is determined that the shape of the back triangle is added at the distance of 5 mm from the head of the fish， and the height is 6 mm. The ideal drag reduction effect can be achieved when the jet hole with the aspect ratio of 4 is used. In this case， the total resistance of the model is 2.510 21 N， and the corresponding drag reduction rate is 6.49%. Through in-depth analysis of the influence of jet hole structure parameters， this paper provides an important theoretical basis and experimental guidance for underwater jet drag reduction technology， and expands a new possibility for the application of bionic technology in the field of underwater fluid mechanics.